sport betting research

The #1 Sports Betting Research Platform In The World

Find Your Edge

We make sports betting easier for you.

EdgHouse is a premium sports betting research platform. We currently cover NFL, NBA, MLB, NCAAM, & NCAAF betting and provide users with betting tools designed to help them make smarter data-driven betting decisions

Using our betting tools will cut your research time down significantly and help you gain a competitive advantage while finding an edge that is needed to beat the sportsbooks in the long run

"@EdgHouse really is the best. All the info you could want on.....I can do all the research I need for game bets and player props all in one place AND they have the best, most responsive customer service. What a treat to start my week."

"you guys seriously make it so easy to find high probability plays".

@DelphiCommish

"2nd to NOBODY, you folks are amazing!!!"

"love using every resource you guys supply. been loving it since day 1".

@donnie_roos

"With the stats I got from y’all I just need to sit back and watch the money roll in.... Appreciate it"

@WeinbergerBlake

"You beauty. Thanks for all your hard work!"

@Arana_Taha

"@EdgHouse strikes again...trend is my friend."

"love the platform".

@ZonaJsPlays

Transforming The Way You Research

Our users are given exclusive access to our sports betting research platform (mobile & web) where all betting cheatsheets, tools, & insights are provided. EdgHouse makes finding and researching sports bets much easier and faster

Our Purpose

To provide you with the best sports betting research tools and make you a better bettor

The Best Betting Research Tools

Our betting tools are designed so you can research any bet, from any angle, at any time

Designed For Sports Bettors

EdgHouse is designed for sports bettors by sports bettors

Scan & Analyze

Scan by swping through bets available, find trends, and analyze further

Visualize & Explore

See the data and research through a new lens like never before

Multip le Sports & Leagues

We currently cover the NFL, NBA, MLB, NCAAM, & NCAAF betting

Easy To Use

Our platform is easy to understand, interactive, and bettor friendly

Web & Mobile

When you sign up for EdgHouse you get access to both our web platform and mobile app (iOS & Android)

Make better data-driven betting decisions and find your true edge

Plans For Everyone

Access to everything.

Pro Plan ($9.99 / week)

All-Star Plan ($19.99 / month)

Hall of Fame Plan ($149.99 / year)

• A ll sports betting research cheatsheets, tools, & insights
• Lines & odds updated in near real-time
• Web and mobile app (iOS & Android)
• NFL, NBA, MLB, NCAAM, & NCAAF coverage
• All future tools added, improvements, & updates
• All future leagues & sports we add
• Dedicated customer service

Have any questions? Please reach out to our support team and we will respond as soon as possible!

Our users are given exclusive access to our sports betting research platform (mobile & web) where all betting cheatsheets, tools, & insights are provided. EdgHouse makes finding and researching sports bets much easier and faster

Join The EdgHouse Community!

 Subscribe to our mailing list to receive the latest news and updates from the EdgHouse team!

Homepage Subscribers Homepage Subscribers

You have successfully subscribed to receive EdgHouse news and updates!

The first sports betting super app

Browse, analyze, and execute picks on major sportsbooks. All in one place.

Integrated with your favorite sportsbooks to place bets faster than ever

The community has spoken

Get inspired. Comb through a curated feed of trending picks and insights to uncover smart betting opportunities.

Explore the markets. Browse thousands of gamelines and player props across leagues to find the perfect pick.

Evaluate the opportunity. Analyze performance trends, injuries, matchup data, public sentiment, and line movement to understand your picks.

Find Your Edge with Outlier Pro. Take action on a live feed of Positive EV bets, filtered to your strategy.

Uncover the insights. Act on player performance and injury insights presented in the context of available betting markets.

Maximize your upside. Secure picks at the best price by comparing odds across your favorite sportsbooks & DFS apps.

Execute seamlessly. Place bets on FanDuel, DraftKings, BetMGM, or Caesars, in just two clicks.

Outlier Pro: Positive EV Betting Tools. Simplify betting and boost profits with Outlier's EV+ tools.

Which plan is right for you?

Our sports betting research platform will make you a smarter, more efficient, and more confident bettor. Get started with a 7 day free trial to all our tools.

Become a Smarter Sports Bettor

Merch. Loving Outlier? Stock up on gear and show off your favorite betting app.

Odds and Evens: The Science and Strategy of Successful Sports Betting

staffwriter

• Published January 22, 2024
• Updated January 2024

In the thrilling world of sports betting, enthusiasts often find themselves walking a fine line between chance and strategy. While luck undoubtedly plays a role, successful sports betting is far from a shot in the dark. Behind the scenes, a complex interplay of mathematics, statistics, and strategic decision-making shapes the landscape of profitable wagering. In exploring the science and strategy behind successful sports betting, we delve into the intricacies that can elevate your betting game and increase your chances of walking away with a win.

Understanding odds: The fundamental building blocks

At the heart of sports betting at 4 pound deposit casino sites in the United Kingdom lies the concept of odds, the numerical representation of the probability of a particular outcome. Whether expressed as fractional, decimal, or moneyline odds, grasping the fundamentals of these numerical representations is essential for any aspiring sports bettor.

Probability and expected value: The science behind the odds

While odds provide insight into potential winnings, understanding probability is the key to making informed betting decisions. The relationship between odds and probability is reciprocal—the higher the odds, the lower the implied probability, and vice versa. Calculating a bet’s expected value (EV) involves assessing the potential return against the likelihood of success.

Expected Value Formula: EV = (Probability of Winning * Potential Profit) – (Probability of Losing * Stake)

In essence, the Expected Value illuminates the path towards making well-informed betting decisions by balancing risk and reward in the intricate dance between odds and probability.

Bankroll management: Protecting your capital

Successful sports betting isn’t just about picking winners; it’s about managing your funds wisely. Bankroll management is a crucial aspect that prevents impulsive decisions and helps bettors navigate the inevitable swings in fortune.

• Set realistic goals : Establish achievable and measurable goals for your betting endeavours, whether it’s a daily, weekly, or monthly target.
• Risk percentage : Avoid risking more than 1-5% of your total bankroll on a single bet to protect against significant losses.
• Bet sizing : Adjust your bet sizes based on your confidence in a particular wager. Higher confidence can justify a larger bet, while uncertainty should lead to smaller stakes.
• Diversify wager types : Spread your bets across different types—straight bets, parlays, and prop bets—to mitigate risk and maximise potential returns.
• Track and analyse performance : Record your bets, wins, and losses meticulously. Regularly analyse your performance to identify trends and areas for improvement.
• Periodic assessments : Conduct periodic evaluations of your bankroll strategy, adjusting it as needed based on your evolving betting style and financial circumstances.
• Embrace patience : Recognise that success in sports betting is a marathon, not a sprint. Practice patience, and avoid chasing losses or overcommitting during hot streaks.
• Research and informed decisions : Invest time in thorough research before placing bets. Informed decisions based on solid analysis enhance your chances of success and contribute to effective bankroll management.
• Emergency fund : Set aside a portion of your bankroll as an emergency fund to cover unexpected expenses. This ensures that your betting activities don’t encroach on essential financial responsibilities.

Research and analysis: The foundation of informed betting

Research and analysis are the bedrock of informed decision-making in the intricate realm of sports betting. Thorough research unveils the nuances that escape casual observation, offering a deeper understanding of the variables influencing a game. Analytical tools and methodologies further refine this information, transforming raw data into actionable insights. Bettors gain a strategic edge by delving into injury reports, team strategies, and past head-to-head matchups. Informed sports betting isn’t a gamble; it’s a calculated endeavour that hinges on the diligent pursuit of knowledge. 

Spotting value bets: Finding the edge

The concept of a “value bet” is central to successful sports betting. Identifying situations where bookmakers’ odds underestimate an outcome’s true probability is crucial for long-term profitability.

• Compare odds across bookmakers : Different bookmakers may offer varying odds for the same event. Comparing prices allows you to spot discrepancies and find the best value.
• Line movements : Track line movements to understand how public sentiment and sharp money influence odds. Quick shifts may indicate valuable opportunities.
• Be selective : Rather than betting on every available market, be selective. Focus on markets where you have expertise and can identify potential value.

Emotional discipline: Mastering the mental game

The psychological aspect of sports betting is often underestimated. Successful bettors maintain emotional discipline, avoiding impulsive decisions driven by excitement, frustration, or fear. Therefore, you should:

• Stick to your strategy : Develop and adhere to a clear betting strategy. Avoid deviating based on emotions or chasing losses.
• Learn from mistakes : Acknowledge that losses are inevitable, but view them as learning opportunities. Analyse errors and refine your approach.
• Avoid “gamblers fallacy” : Each bet is independent of the previous ones. Refrain from falling into the trap of thinking that past outcomes influence future results.

Live betting and in-play opportunities: Seizing the moment

In the evolving landscape of sports betting, live betting has emerged as a transformative force, inviting bettors to engage actively with the unfolding drama of a match. In-play odds undergo continuous adjustments as the game progresses, offering a real-time reflection of the unfolding events. This fluidity in odds presents a canvas of strategic possibilities for bettors who can discern favourable moments to place their wagers. This adaptability aligns with the unpredictability inherent in sports, providing a safety net for strategic wagering.

Specialising in niche markets: Finding your niche

While mainstream sports attract the majority of betting attention, specialising in niche markets or less popular sports can provide unique advantages:

• Less bookmaker attention : Niche markets may receive less attention from bookmakers, potentially leading to softer odds and more value.
• Specialist knowledge : Becoming an expert in a particular niche allows for more informed betting decisions, as you have a deeper understanding of the factors influencing outcomes.
• Reduced public influence : Niche markets may be less susceptible to the influence of public sentiment and media coverage, providing a more objective betting landscape.

In conclusion, the science and strategy of successful sports betting require a multifaceted approach that blends mathematical acumen, strategic decision-making, and emotional discipline. By understanding the intricacies of odds, grasping the science behind the expected value, practising effective bankroll management, conducting thorough research, and staying disciplined in the face of wins and losses, bettors can enhance their chances of long-term success in the dynamic world of sports wagering. Betting may have an element of chance, but with the right approach, it can evolve into a calculated and strategic pursuit beyond mere luck.

While it’s feasible to increase your chances of winning, we’d always remind you that sports betting is gambling and ultimately comes with a sizeable risk of losing your money. Never bet more than you can afford to lose.

Recent Posts

Phil Foden’s rise: generational talent and a feverish passion for football 

Phil Foden is a 23 year old with no fewer than 6 Premier League titles to his name. Let’s take a look at how Foden became Foden.

Hiking in Canada – What You Need to Know

Canada is a huge place with some of the best hiking in the world. While we couldn’t possibly claim to be able to tell you EVERYTHING you need to know in once single sentence, we can certainly offer up some key information for those of you planning a hiking led adventure break in arguably the most beautiful part of North America.

Best Portable Charging Banks for Hiking

Traditionalists may not like it! But personally, I don’t like to risk being stuck on a hill or mountain with a dead phone or head torch battery. So I like to carry a portable charging device (or 2, or 3) with me. Here are the best portable charging banks for hiking that I’ve used. (And I’ve used a lot, by the way).

The Hiker’s Guide to Dovestone

Dovestone is a beautiful place in the Peak District’s Dark peak, nestled amongst the Saddleworth Hills. But there’s far more for hikers here than reservoir walks. Here’s a hiker’s guide to Dovestone Reservoir and the surrounding hills.

Across the website, Our Sporting Life uses affiliate links on some of its content. In a nutshell, it means that if you click one of those links and then go on to make a purchase, we may receive a commission (but you won’t be charged any more by the seller of that product or service). Importantly, though, opinions on products, services and so forth remain our own. We do not accept payment to review anything and we don’t accept payment to place editorial either.

• Online Sports Betting Market Size

Report on Industry Size & Market Share Analysis - Growth Trends & Forecasts (2024 - 2029)

Online Sports Betting is Experiencing Significant Transformations Due To Technological Advancements. Leading Firms are Enhancing Their Platforms, With Cashless Transactions and Increased Female Participation Expected To Boost the Market. The Sector's Growth is Also Driven by Virtual Reality and Blockchain Technologies. Digital Sports Betting is Popular in Events Like the FIFA World Cup and European Championships, and in Sports Like Horse Racing and Tennis. The Market is Segmented by Sports Type, Device, and Location, With Most Bets Placed On Football, Basketball, Horse Racing, and Cricket. The Convenience of Remote Sports Betting and Secure Transaction Methods are Key Market Drivers. The Market is Highly Competitive and Fragmented, With Players Adopting Strategies Like Expansions, Innovations, and New Product Launches To Maintain Competitiveness.

INSTANT ACCESS

Single User License

Team License

Corporate License

Need a report that reflects how COVID-19 has impacted this market and its growth?

Online Sports Betting Market Analysis

The Online Sports Betting Market size is estimated at USD 48.17 billion in 2024, and is expected to reach USD 83.58 billion by 2029, growing at a CAGR of 11.65% during the forecast period (2024-2029).

Sports betting involves gambling by predicting the outcome of a game and placing a wager. It is popular in various sports, including cricket, football, and horse racing. The proliferation of the internet is expected to drive advancements in the sports betting industry. European football generates the most revenue in online betting sports, with baseball trailing closely behind. The increasing popularity of online sports has revolutionized the methods of sports betting. However, numerous countries prohibit sports betting, which significantly hinders the expansion of the sports betting market. However, relaxation in the regulation framework of betting and gambling activities by governments worldwide is anticipated to provide favorable opportunities for betting operators. For instance, in 2023, as per the American Gaming Association in the United States, sports betting was legally permitted in 36 states, an increase from 32 in 2021. Within the first ten months of 2022, sports bettors in the United States legally wagered approximately USD 73 billion.

• Online Sports Betting Market Trends

Increasing Popularity of Online Gambling

The football segment of the international online sports betting market records a high betting volume, with a growing number of bets. It is especially prevalent in European countries such as Italy, France, and Spain, where football is more popular. The online betting segment is predominantly applied in sports, especially in football events like the Fédération Internationale de Football Association (FIFA) World Cup and European Championships. Moreover, the increased penetration of smartphones is leading to an increase in several mobile application-based lottery games. The end-user has the convenience and comfort of gambling within the comfort of their own space, which is one of the primary drivers of the segment. Casino gambling has been one of the rapidly growing gambling categories, owing to the convenience of usage and optimal user experience. Also, the rising adoption of online payment gateways has made payment options convenient for players. Online payment provides a safe and secure mode of transaction, boosting the growth of the online sports betting market.

Europe Dominates the Market

The market records a high demand from European betting consumers, who bet across multiple leagues, pre-match, and in-play. The market generates a significant portion of its revenue from the United Kingdom, Spain, Germany, and other European countries, as most companies are expanding into regulated markets to generate sustainable revenues. As online sports betting is predominantly applied in events such as the FIFA World Cup, the Wimbledon Championship, and the European Championships, the online sports betting market in Europe has grown significantly in the last few years. Companies are expanding their presence in the region based on various factors, including offerings, user experience, brand equity, personalized payoffs, and access to multiple platforms. For instance, in September 2023, global gaming operator Betsson entered the French online betting market. Betsson’s entrance into the French market results from collaborating with a local partner. It strategically aims to bring the company closer to the French sports betting arena. In the fourth quarter of 2023, the official launch of Betsson’s flagship brand took place.

Online Sports Betting Industry Overview

The market studied is fragmented due to the strong presence of regional and global players. Key players dominate the market, including Bet365, 888 Holdings PLC, Flutter Entertainment PLC, Entain PLC, and The Stars Group. Key players compete on various factors, including offerings, user experience, brand equity, personalized payoffs, and access to multiple platforms. The key strategies adopted by the players in the market are expansions, innovations, and product launches to maintain competitiveness in the market. They also focus on mergers to increase their market stake and improve profit margins.

Online Sports Betting Market Leaders

The Stars Group

888 holdings PLC

Flutter Entertainment PLC

*Disclaimer: Major Players sorted in no particular order

Online Sports Betting Market News

• December 2023: Betsson collaborated with Racing Club de Avellaneda for the upcoming 2023/2024 season. Starting May 19, the Swedish brand logo would be featured on the upper back of the men’s and women’s First Division football teams’ shirts for local and international matches.
• September 2023: Betsson was awarded a license to offer online sports betting on the locally regulated market in France. The operations in France would run in collaboration with a local partner. The launch in the country took place during the fourth quarter of 2023 under the Betsson brand.
• March 2023: OpenBet, a multinational company that provides sports betting entertainment, expanded its presence in the United States with a day-one launch in Massachusetts.

Online Sports Betting Market Report - Table of Contents

1. INTRODUCTION

1.1 Study Assumptions and Market Definition

1.2 Scope of the Study

2. RESEARCH METHODOLOGY

3. EXECUTIVE SUMMARY

4. MARKET DYNAMICS

4.1 Market Drivers

4.1.1 Increasing Popularity of Online Gambling

4.1.2 Advancement In Security, Encryption, and Streaming Technology

4.2 Market Restraints

4.2.1 Regulatory Uncertainty And Compliance

4.3 Industry Attractiveness - Porter's Five Forces Analysis

4.3.1 Bargaining Power of Suppliers

4.3.2 Bargaining Power of Buyers/Consumers

4.3.3 Threat of New Entrants

4.3.4 Threat of Substitutes

4.3.5 Degree of Competition

5. MARKET SEGMENTATION

5.1 Sport Type

5.1.1 Football

5.1.2 Basketball

5.1.3 Horse Racing

5.1.4 Baseball

5.1.5 Tennis

5.1.6 Other Sport Types

5.2 End User

5.2.1 Desktop

5.2.2 Mobile

5.3 Geography

5.3.1 North America

5.3.1.1 United States

5.3.1.2 Canada

5.3.1.3 Mexico

5.3.1.4 Rest of North America

5.3.2 Europe

5.3.2.1 Spain

5.3.2.2 United Kingdom

5.3.2.3 Germany

5.3.2.4 France

5.3.2.5 Italy

5.3.2.6 Sweden

5.3.2.7 Rest of Europe

5.3.3 Asia-Pacific

5.3.3.1 Oceanic Countries

5.3.3.2 Rest of Asia-Pacific

5.3.4 Rest Of The World

5.3.4.1 South America

5.3.4.2 Middle East & Africa

6. COMPETITIVE LANDSCAPE

6.1 Most Adopted Strategies

6.2 Market Share Analysis

6.3 Company Profiles

6.3.1 888 holdings PLC

6.3.2 Entain PLC

6.3.3 BET 365

6.3.4 Betsson AB

6.3.5 Flutter Entertainment PLC

6.3.6 Draftkings Inc.

6.3.7 Kindred Group PLC

6.3.8 1XBET

6.3.9 22BET

6.3.10 Sportpesa

• *List Not Exhaustive

7. MARKET OPPORTUNITIES AND FUTURE TRENDS

Online Sports Betting Industry Segmentation

Sports betting is predicting sports results and placing a wager on the outcome. The frequency of sports bets varies by culture, with the vast majority placed on football, basketball, horse racing, cricket, and others. The online sports betting market is segmented by sports type, end user, and geography. Based on sports type, the market is segmented into football, basketball, horse racing, baseball, tennis, and other sports types. Based on end users, the market is segmented into desktop and mobile. Moreover, an analysis has been done based on emerging and established regions, including North America, Europe, Asia-Pacific, and the Rest of the World. The market sizing and forecasts have been done for each segment based on value (USD).

Online Sports Betting Market Research FAQs

How big is the online sports betting market.

The Online Sports Betting Market size is expected to reach USD 48.17 billion in 2024 and grow at a CAGR of 11.65% to reach USD 83.58 billion by 2029.

What is the current Online Sports Betting Market size?

In 2024, the Online Sports Betting Market size is expected to reach USD 48.17 billion.

Who are the key players in Online Sports Betting Market?

Entain PLC, The Stars Group, 888 holdings PLC, BET 365 and Flutter Entertainment PLC are the major companies operating in the Online Sports Betting Market.

Which is the fastest growing region in Online Sports Betting Market?

North America is estimated to grow at the highest CAGR over the forecast period (2024-2029).

Which region has the biggest share in Online Sports Betting Market?

In 2024, the Europe accounts for the largest market share in Online Sports Betting Market.

What years does this Online Sports Betting Market cover, and what was the market size in 2023?

In 2023, the Online Sports Betting Market size was estimated at USD 42.56 billion. The report covers the Online Sports Betting Market historical market size for years: 2019, 2020, 2021, 2022 and 2023. The report also forecasts the Online Sports Betting Market size for years: 2024, 2025, 2026, 2027, 2028 and 2029.

What are the key regional trends shaping Online Sports Betting?

The increasing growth in emerging markets like Asia Pacific and Latin America, alongside regulations evolving in established markets are the key regional trends shaping Online Sports Betting.

What are the emerging technologies shaping the future of Online Sports Betting?

The emerging technologies shaping the future of Online Sports Betting are a) AI-powered personalization b) Virtual reality experiences as potential game-changers

Our Best Selling Reports

• Fly Ash Market
• Bio-Based Succinic Acid Market
• Lubricant Additives Market
• Aliphatic Hydrocarbon Solvents and Thinners Market
• Feed Preservatives Market
• Self-Healing Coatings Market
• China EV Market
• IoT Sensor Market
• Global Industrial Fasteners Market
• PC Based Automation Market

Online Sports Betting Industry Report

The global online sports betting market is experiencing significant growth due to the rapid adoption of internet-connected devices like smartphones and tablets. The development of online betting is expected to cause a significant increase in demand for the global sports betting industry over the projected period. The growth of high-speed internet services and the growing acceptance of online gambling, particularly in football-loving nations like Italy, France, and Spain, are driving this expansion. However, the increased negative effects on mental health and restrictions on online gambling in several countries are anticipated to hamper the market expansion. Despite these challenges, the growth of live e-sports coverage platforms is anticipated to create lucrative opportunities for the global market to expand. The market is divided into various segments based on type and sports type, with football and line-in-play segments owning the highest market shares. Statistics for the Online sports betting market share, size, and revenue growth rate, created by Mordor Intelligence™ Industry Reports. Online sports betting analysis includes a market forecast outlook and historical overview. Get a sample of this industry analysis as a free report PDF download.

Online Sports Betting Market Report Snapshots

• Online Sports Betting Market Share
• Online Sports Betting Companies
• Online Sports Betting News

Please enter a valid email id!

Please enter a valid message!

Online Sports Betting Market Size & Share Analysis - Growth Trends & Forecasts (2024 - 2029)

Online Sports Betting Market Get a free sample of this report

Please enter your name

Business Email

Please enter a valid email

Please enter your phone number

Get this Data in a Free Sample of the Online Sports Betting Market Report

Please enter your requirement

Thank you for your Purchase. Your payment is successful. The Report will be delivered in 2 - 4 hours. Our sales representative will reach you shortly with the details.

Please be sure to check your spam folder too.

Sorry! Payment Failed. Please check with your bank for further details.

Add Citation APA MLA Chicago

➜ Embed Code X

Get Embed Code

Want to use this image? X

Please copy & paste this embed code onto your site:

Images must be attributed to Mordor Intelligence. Learn more

About The Embed Code X

Mordor Intelligence's images may only be used with attribution back to Mordor Intelligence. Using the Mordor Intelligence's embed code renders the image with an attribution line that satisfies this requirement.

In addition, by using the embed code, you reduce the load on your web server, because the image will be hosted on the same worldwide content delivery network Mordor Intelligence uses instead of your web server.

Sports Betting

This article provides insights into the increasing interest in sports betting and its connection to consumer behavior, demographics, and personality traits. Understanding these dynamics is essential for businesses seeking to engage with this growing market.

• The increasing popularity of sports betting, driven by prominent platforms like FanDuel and DraftKings, is explored through data from the 2023 Prosper Media Behavior and Influence (MBI) study, encompassing 17,159 adults aged 18+.
•  Two key questions were asked in the study: “Do you gamble on sports?” and “Do you play fantasy sports?” Responses were categorized as regularly, occasionally, or never.
• 31.7% of respondents engage in sports gambling (regularly or occasionally), while 27.2% participate in fantasy sports, with a substantial overlap of 22.7% doing both.
• Within these categories, 71.7% of sports gamblers also play fantasy sports, and 83.4% of fantasy sports participants also gamble on sports.
• Approximately 40% of the 2023 MBI study population resides in states where online sports betting is illegal, despite substantial legal concerns, especially in large states like California and Texas.
• Historical data reveals a steady increase in sports gambling, particularly in recent years, while fantasy sports participation has remained relatively stable.
• In 2023, 42.1% of males reported gambling on sports compared to 21.5% of females, with both genders showing growth over the past three years, albeit at different rates.
• Demographic characteristics reveal higher sports gambling rates among Gen-Z, Millennials, and males, with connections to various occupations and a slight advantage for those with higher education.
• A classification regression tree (CRT) analysis identifies segments with high sports gambling rates, with age, gender, income, marital status, and ethnicity being significant predictors.
• Viewership patterns indicate that gambling on sports is related to specific sports categories, with minor sports and women’s sports exhibiting greater relative interest among gamblers. Additionally, sports bettors are more likely to engage in sports-related activities, favor alcoholic beverages, and display specific personality traits, including high extraversion and lower grit, demonstrating a distinct consumer profile.

Forward By Dr. Larry DeGaris, Executive Director, Medill Spiegel Research Center, Northwestern University

“Legalized sports gambling is the endgame. One-day fantasy delivers a similar fan experience to gambling, so I expect the current database of customers would provide a good foundation for sports gamblers.” I said this in 2015 when I was interviewed for the Bloomberg article “Daily Fantasy Sites Seen Positioned for Jump to Sports Gambling.”

This interview came on the heels of NBA commissioner Adam Silvers’s 2014 NYT op-ed, “Legalize and Regulate Sports Betting,” advocating legalizing sports betting; I knew it was just a matter of time before sports betting was legalized nationwide and DraftKings and FanDuel were planting the seeds.

Fast forward to 2023, and DraftKings and FanDuel are dominating the sports betting market while big sports brands like Fox and big gaming brands like MGM struggle or fail. I thought it was a good strategy. I didn’t realize how good.

The sports betting market highlights the value of customer data. DraftKings and FanDuel’s market domination is based on their customer data strategy. They bet that fantasy players, especially daily fantasy players, would be the most promising market for sports betting. As sports betting became legal in more states, they were ready. It was a big bet and a big win.

Similarly, brands are placing big bets on Retail Media Networks. The strategy is the same. As we enter the cookie-less world, customer data will become more valuable. That’s no fantasy.

More recently, Fanatics is getting into the sports betting game with its acquisition of Pointsbet, betting that its database of fans will help them compete where others have fallen short. I’m not sure I like the odds. Let’s reverse engineer DraftKings’s and FanDuel’s winning strategy. Martin Block’s nifty little sports betting paper shows us the way.

The relationship between betting on sports and playing fantasy sports is very high. The two groups are almost identical. Unlike the years it took Amazon to build a customer database, DraftKings and FanDuel had theirs built before sports betting became legal. As more states legalized sports betting, they were ready. That initial advantage has been difficult for other entrants to overcome.

Having a database of sports fans doesn’t measure up to a targeted list of fantasy sports players. Similarly, I reckon Fanatics has a great database of fans passionate about their teams, which likewise wouldn’t translate well to betting.

Digging deeper, we see the underlying mechanisms of sports betting and outline a path forward. DraftKings and FanDuel won the first round, but it ain’t over ‘til it’s over.

The women’s market in sports tends to be underserved in general because fan populations are more likely to be male. The difference in degree mistakenly is equated with a categorical difference. Sports bettors are more likely to be male, but there are a lot of women betting on sports.

Sports betting companies have been active as sponsors of sports properties. But sports betting is a fan activity and, as Martin points out, there are many product categories that index high for sports bettors. Creating a brand partnership strategy using sports betting as a platform could bring another revenue stream.

Sports bettors are highly social and competitive. That’s the underlying link between fantasy and betting. Fantasy was a natural connection for sports betting in that respect, but if sportsbooks can innovate ways to connect fans who weren’t playing fantasy, they can gain an entry point. What do you talk about when your team is having a bad year? This week’s parlays. Sports betting is a social experience for fans. Sports betting companies can create a social infrastructure for fans to interact, as they’ve done with fantasy leagues.

The story highlights both the importance and limitations of customer data. Customer data tells you who. Traditional market research can help understand the why.

We should keep this in mind as we watch retail media grow. First-party customer data is tremendously valuable but limited without a deeper understanding of shopper experiences and identities.

By Dr. Martin Block, Professor Emeritus, Northwestern University, Retail Analytics Council

Interest in sports betting is increasing with the publicity that FanDuel and DraftKings have recently enjoyed. To better understand sports betting data from the annual Prosper Media Behavior and Influence (MBI) study collected in January of 2023 (n=17,159) of adults 18+ is analyzed. Two questions were asked: “Do you gamble on sports?” and “Do you play fantasy sports?” Both questions were supplied with answer options of regularly, occasionally, and never. Table 1 shows the proportion of the sample that gamble on sports combining regularly and occasionally at 31.7%, and those playing fantasy sports at 27.2%. There is substantial overlap between gambling and fantasy sports, with 22.7% reporting both. Of those who gamble, 71.7% say they also play fantasy sports. Of those who play fantasy sports, 83.4% say they also gamble. Not all respondents say they gamble and play exclusively online. Of those who say they gamble on sports, 66.0% say they also gamble online. Of those who play fantasy sports, 84.1% say they also play fantasy sports online.

The legality of online sports betting has also been a topic of interest. Using FanDuel’s map of states where the service is legal, Table 2 shows that approximately 40% of the population as represented by the 2023 MBI are in states where it is not legal. Two big states, California and Texas are among the states where it is not legal. It should be noted that in states where it is not legal, the proportion is slightly higher than in states where it is legal. The difference, despite the very large sample sizes, is not statistically different. Legalizing online sports betting appears to have little influence.

The gambling and fantasy sports questions have been asked in the MBI for several years. Table 3 shows the trends over the last eight years for gambling, playing fantasy sports, and playing video games. Playing video games has slowly increased over the years but is showing a slight decline in 2023. Gambling on sports has steadily increased, especially in recent years. Playing fantasy sports has remained almost perfectly flat.

Considering just the last three years, as shown in Table 4 illustrates the recent growth, especially for 2023. It also shows a comparison of gambling by gender. In 2023, 42.1% of males reported gambling on sports compared to 21.5% of females. The average annual growth rate for males over the three years is 3.9%, compared to 4.0% for females.

Table 5 shows some differences by various demographic characteristics and the breakdown by regularly and occasionally gambling on sports. Looking at the characteristic variables by themselves, it appears that sports gambling is highest among Gen-Z, Millennials, and males. There seems to be strong connections to occupation, especially business owners, professionals, salespersons, workers, students, and military. All these occupations have a social component. There is also a slight advantage for higher or more education.

Identifying the Sports Bettor

Using a classification regression tree (CRT) to predict gambling on sports allows for the identification of consumer segments that have high rates of gambling on sports. Table 6 shows the results of such analysis and the important predictor variables. The most important predictor, by far, is age. This is followed by gender, income, family size, marital status, ethnicity, and population density. The model itself is reasonably strong, predicting a bettor 73.7% of the time using 10-fold cross-validation.

The tree itself identifies the segments as terminal nodes, as shown in Table 7. The highest node, those under 44, male, incomes over $106.5K, and married report 81.2% betting on sports and comprise 5.3% of the total population of adults. Slight older, ranging from 44 to 51, report 67.7% betting. Under 51, males, over $1065k   and are single (never married) report 60.9% and comprise 1.6% of adults. Males under the age of 51 with incomes between $53k and $106K living in an urban area report 64.0%, and males in suburban and rural areas report 50.6%. Males under the age of 51 with incomes less than $53K, who are non-white report 51.1% gambling on sports and comprise 6.3% of the total population. These are all the male segments above 40% of sports gambling.

The female side of the tree shows non-white with a family size of four or more at 45.3%, comprising 5.4% of the total sample. Non-white females, with a family size of less than four, and younger than 36 report 40.0% gambling and comprise 4.0% of the total sample. Among the older segments are males earning over $53K, aged 52 to 55, reporting 49.6% gambling, and African American males earning under $53K, aged 52 to 63, reporting 44.9% gambling.

Sports Program Viewing

Respondents were asked if they regularly viewed 12 different sporting events. To establish a pattern, the 12 were factor-analyzed (principal component) as shown in Table 8. The viewing patterns are arranged in three groups labeled “major,” which includes the NFL, college football, and other team sports of broad interest. The analysis shows that those who report viewing the NFL are also likely to watch college football. The second group, labeled “minor,” includes professional tennis and golf and have a narrower level of interest and are not thought of as team sports. The third group is labelled “women’s” for woman’s sports..”

Relating the viewing to gambling on sports is shown in Table 9. Overall, 73.2% of adults report viewing at least one of the categories of sports programming. The NFL is viewed by 38.4% of adults, followed by college football, major league baseball, and the NBA. Sports betting is certainly related to viewing, as in the case of the NFL, where 46.1% of the viewers compared to 34.7% of the viewers do not gamble, providing a gambling index of 120.1. The major sports have an average index of 150.7, minor sports have an average index of 176.9, and women’s sports have the highest average index of 189.8. Minor, less team-oriented sports, and especially women’s sports, are of greater relative interest to gamblers.

Weighting sports betting by regular viewing demonstrates the relative interest in the sports category, as shown in Table 11. The NFL overall leads sports betting activity, followed by the NBA, college football, major league baseball, and college basketball. All these categories are team-oriented and of broad interest.

Those who report gambling on sports are more likely to engage in a variety of sports-related activities, as shown in Table 12. Engaging in online fantasy sports and online gambling index at almost double. Sports on the radio and smartphones are higher than regularly watching sports on TV. Fox Sports indexes higher than ESPN.

Respondents were also asked about 31 different leisure activities, which are compared to reported gambling on sports, as shown in Table 13. The highest indexing activity is playing team sports. Also high are very active categories such as tennis and snow skiing. Social activities, such as tailgating and going to gambling resorts, bars, and sporting events are also high. The image of sports betting is being social, interested in competition, and physically active. On the other hand, those indexing low is generally passive and perhaps a bit more solitary, reporting activities such as watching TV, surfing the internet, crafting hobbies, and reading books. Those activities are indexed at nearly the same (not shown) as going shopping, online communities, and travel.

Product Consumption

Consumption of alcoholic drinks is commonly associated with viewing and gambling on sports. Table 14 shows this to be true as most categories of drinks index at almost double among those that gamble. The lowest index is wine.

Average monthly product spending is collected monthly, with the categories rotating across different months. The data from the monthly surveys (average n=7,500) needs to be integrated with the annual MBI, which is done using decision tree equations from available demographic variables common to both the monthly and MBI studies. Shown in Table 15 are the average reported household monthly spending across a variety of product categories. The months are taken from 2022 prior to the 2023 MBI. The averages include zero if the respondent reports no spending in that category. The highest indexing category is sporting goods, followed by men’s clothing and both fast food and full-service restaurants. It is worth noting the home improvement spending indexes highly note the characteristics of the sports gambler. Smartphone and grocery spending is almost the same. Overall, those who gamble on sports spend more on average than those who do not.

Personality

Several personality measures were also collected in the MBI including the five-factor personality inventory commonly referred to as OCEAN, grit, and impulsiveness. The personality inventory used by Prosper in the MBI is taken from the short-form version developed by the Gosling Laboratory at the University of Texas. The five factors include:

• Extraversion: Extraversion is characterized by excitability, sociability, talkativeness, assertiveness, and high amounts of emotional expressiveness. People who are high in extroversion are outgoing and tend to gain energy in social situations. People who are low in extroversion (or introverted) tend to be more reserved and must expend energy in social settings.
• Agreeableness: This personality dimension includes attributes such as trust, altruism, kindness, affection, and other prosocial behaviors. People who are high in agreeableness tend to be more cooperative, while those low in this trait tend to be more competitive and even manipulative.
• Conscientiousness: Standard features of this dimension include high levels of thoughtfulness, with good impulse control and goal-directed behaviors. Those high on conscientiousness tend to be organized and mindful of details.
• Neuroticism: Neuroticism is a trait characterized by sadness, moodiness, and emotional instability. Individuals who are high in this trait tend to experience mood swings, anxiety, moodiness, irritability, and sadness. Those low in this trait tend to be more stable and emotionally resilient. This trait is sometimes inversely reported as “emotional stability.”  The Prosper database uses the “emotional stability” terminology.
• Openness: This trait features characteristics such as imagination and insight, and those high in this trait also tend to have a broad range of interests. People who are high in this trait tend to be more adventurous and creative. People low in this trait are often much more traditional and may struggle with abstract thinking.

Table 16 shows a comparison of sports gamblers by the five personality traits. All five are statistically different. The biggest differences are gamblers are more competitive and more extravagant with less impulse control. Gamblers also have higher sociability, are more traditional, and are less likely to think abstractly. They are also more emotionally stable and resilient.

A non-cognitive predictor of success in life, such as grit, like having a clear inner compass that guides all your decisions and actions. Grit is defined as the disposition to demonstrate perseverance and passion for long-term goals (Angela Duckworth, Grit: The Power of Passion and Perseverance , Scribner, 2016). Passion is a deep, enduring knowledge of what you want. Perseverance is hard work and resilience. All three predictors are expressed as a five-point scale, with higher values being stronger on the characteristics developed from a series of five-points. Prosper used the short-scale version.

Again, shown in Table 16 is the comparison of sports gamblers. All the differences are statistically different. Gamblers have lower overall grit, meaning they are less oriented toward long-term goals. The biggest difference for gamblers is lower passion, that is knowledge of what they want. Gamblers are higher in perseverance.

Impulsivity, as measured by agreement with the statement “Live for today because tomorrow is so uncertain,” also demonstrates a difference among gamblers, as shown in Table 17. Gamblers are much more likely to strongly agree with the statement and, conversely, strongly disagree. This reinforces the lack of long-term orientation among gamblers.

The Prosper Media Behavior and Influence study from January 2023 unveils the rising interest in sports betting, emphasizing the convergence of sports gambling and fantasy sports. The analysis explores demographic predictors, revealing age, gender, income, and marital status as key factors influencing sports betting engagement. Furthermore, the study uncovers correlations between sports program viewing, personality traits, leisure activities, and product consumption, painting a comprehensive picture of the diverse profile of sports gamblers.

Articles on Sports betting

Displaying 1 - 20 of 59 articles.

How the 18th-century ‘probability revolution’ fueled the casino gambling craze

John Eglin , University of Montana

Professional sport commissioners are fighting to preserve league integrity amid gambling scandals

Craig Greenham , University of Windsor

Sports gambling creates a windfall, but raises questions of integrity – here are three lessons from historic sports-betting scandals

Jared Bahir Browsh , University of Colorado Boulder

March Madness brings unique gambling risks for college students

M. Dolores Cimini , University at Albany, State University of New York

Why March Madness is a special time of year for state budgets

Jay L. Zagorsky , Boston University

Ads, food and gambling galore − 5 essential reads for the Super Bowl

Nick Lehr , The Conversation

The Super Bowl gets the Vegas treatment, with 1 in 4 American adults expected to gamble on the big game

Thomas Oates , University of Iowa

Colleges face gambling addiction among students as sports betting spreads

Jason W. Osborne , Miami University

How AI and AR could increase the risk of problem gambling for online sports betting

Philip Newall , University of Bristol and Jamie Torrance , University of Chester

Australia has a strong hand to tackle gambling harm. Will it go all in or fold?

Charles Livingstone , Monash University

Sport is being used to normalise gambling. We should treat the problem just like smoking

Grattan on Friday: Peter Dutton warns of threat to ‘working poor’ in budget reply lacking a big picture

Michelle Grattan , University of Canberra

Premier League’s front-of -shirt gambling ad ban is a flawed approach. Australia should learn from it

Samantha Thomas , Deakin University ; Hannah Pitt , Deakin University , and Simone McCarthy , Deakin University

As March Madness looms, growth in legalized sports betting may pose a threat to college athletes

How the push to end tobacco advertising in the 1970s could be used to curb gambling ads today

Carolyn Holbrook , Deakin University and Thomas Kehoe

Gambling Act review: how EU countries are tightening restrictions on ads and why the UK should too

Raffaello Rossi , University of Bristol ; Agnes Nairn , University of Bristol ; Ben Ford , University of Bristol , and Jamie Wheaton , University of Bristol

A boon for sports fandom or a looming mental health crisis? 5 essential reads on the effects of legal sports betting

Data from New Jersey is a warning sign for young sports bettors

Lia Nower , Rutgers University

Millions of Americans are problem gamblers – so why do so few people ever seek treatment?

James P. Whelan , University of Memphis

I treat people with gambling disorder – and I’m starting to see more and more young men who are betting on sports

Tori Horn , University of Memphis

Related Topics

• Gambling addiction
• Gambling harm
• Gambling in America
• Gambling reform
• Online gambling
• Problem gambling

Top contributors

Associate Professor, School of Public Health and Preventive Medicine, Monash University

Knight Chair in Sports Journalism and Society, Penn State

Lecturer and Researcher in Psychology, Swansea University

Professor of Public Health, Deakin University

PhD Student in Clinical Psychology, University of Memphis

Principal Research Fellow, CQUniversity Australia

Professor and Director, Center for Gambling Studies, Rutgers University

Clinical Psychologist, University of Sydney

Associate Professor of Markets, Public Policy and Law, Boston University

Researcher, School of Psychology, University of Sydney

Postdoctoral Research Fellow, Deakin University

Research Professor of Clinical Health, University of Memphis

VicHealth Postdoctoral Research Fellow, Deakin University

Assistant Professor of Psychology, East Tennessee State University

Lecturer in the School of Psychological Science, University of Bristol

• X (Twitter)
• Unfollow topic Follow topic

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here .

Loading metrics

Open Access

Peer-reviewed

Research Article

A statistical theory of optimal decision-making in sports betting

Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Visualization, Writing – original draft, Writing – review & editing

* E-mail: [email protected]

Affiliation Dept of Biomedical Engineering, City College of New York, New York, NY, United States of America

• Jacek P. Dmochowski

• Published: June 28, 2023
• https://doi.org/10.1371/journal.pone.0287601
• Peer Review
• Reader Comments

The recent legalization of sports wagering in many regions of North America has renewed attention on the practice of sports betting. Although considerable effort has been previously devoted to the analysis of sportsbook odds setting and public betting trends, the principles governing optimal wagering have received less focus. Here the key decisions facing the sports bettor are cast in terms of the probability distribution of the outcome variable and the sportsbook’s proposition. Knowledge of the median outcome is shown to be a sufficient condition for optimal prediction in a given match, but additional quantiles are necessary to optimally select the subset of matches to wager on (i.e., those in which one of the outcomes yields a positive expected profit). Upper and lower bounds on wagering accuracy are derived, and the conditions required for statistical estimators to attain the upper bound are provided. To relate the theory to a real-world betting market, an empirical analysis of over 5000 matches from the National Football League is conducted. It is found that the point spreads and totals proposed by sportsbooks capture 86% and 79% of the variability in the median outcome, respectively. The data suggests that, in most cases, a sportsbook bias of only a single point from the true median is sufficient to permit a positive expected profit. Collectively, these findings provide a statistical framework that may be utilized by the betting public to guide decision-making.

Citation: Dmochowski JP (2023) A statistical theory of optimal decision-making in sports betting. PLoS ONE 18(6): e0287601. https://doi.org/10.1371/journal.pone.0287601

Editor: Baogui Xin, Shandong University of Science and Technology, CHINA

Received: December 19, 2022; Accepted: June 8, 2023; Published: June 28, 2023

Copyright: © 2023 Jacek P. Dmochowski. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: Data is available at https://github.com/dmochow/optimal_betting_theory .

Funding: The author received no specific funding for this work.

Competing interests: The authors have declared that no competing interests exist.

Introduction

The practice of sports betting dates back to the times of Ancient Greece and Rome [ 1 ]. With the much more recent legalization of online sports wagering in many regions of North America, the global betting market is projected to reach 140 billion USD by 2028 [ 2 ]. Perhaps owing to its ubiquity and market size, sports betting has historically received considerable interest from the scientific community [ 3 ].

A topic of obvious relevance to the betting public, and one that has also been the subject of multiple studies, is the efficiency of sports betting markets [ 4 ]. While multiple studies have reported evidence for market inefficiencies [ 5 – 11 ], others have reached the opposite conclusion [ 12 , 13 ]. The discrepancy may signify that certain, but not all, sports markets exhibit inefficiencies. Research into sports betting has also revealed insights into the utility of the “wisdom of the crowd” [ 14 – 16 ], the predictive power of market prices [ 17 – 20 ], quantitative rating systems [ 21 , 22 ], and the important finding that sportsbooks exploit public biases to maximize their profits [ 13 , 23 ].

Indeed, the decisions made by sportsbooks to set the offered odds and payouts have been previously analyzed [ 13 , 23 , 24 ]. On the other hand, arguably less is known about optimality on the side of the bettor . The classic paper by Kelly [ 25 ] provides the theory for optimizing betsize (as a function of the likelihood of winning the bet) and can readily be applied to sports wagering. The Kelly bet sizing procedure and two heuristic bet sizing strategies are evaluated in the work of Hvattum and Arntzen [ 26 ]. The work of Snowberg and Wolfers [ 27 ] provides evidence that the public’s exaggerated betting on improbable events may be explained by a model of misperceived probabilities. Wunderlich and Memmert [ 28 ] analyze the counterintuitive relationship between the accuracy of a forecasting model and its subsequent profitability, showing that the two are not generally monotonic. Despite these prior works, idealized statistical answers to the critical questions facing the bettor, namely what games to wager on, and on what side to bet, have not been proposed. Similarly, the theoretical limits on wagering accuracy, and under what statistical conditions they may be attained in practice, are unclear.

To that end, the goal of this paper is to provide a statistical framework by which the astute sports bettor may guide their decisions. Wagering is cast in probabilistic terms by modeling the relevant outcome (e.g. margin of victory) as a random variable. Together with the proposed sportsbook odds, the distribution of this random variable is employed to derive a set of propositions that convey the answers to the key questions posed above. This theoretical treatment is complemented with empirical results from the National Football League that instantiate the derived propositions and shed light onto how closely sportsbook prices deviate from their theoretical optima (i.e., those that do not permit positive returns to the bettor).

Importantly, it is not an objective of this paper to propose or analyze the utility of any specific predictors (“features”) or models. Nevertheless, the paper concludes with an attempt to distill the presented theorems into a set of general guidelines to aid the decision making of the bettor.

Problem formulation: “Point spread” betting

For positive s (home team favored), the home team is said to “cover the spread” if m > s , whereas the visiting team has “beat the spread” otherwise. Conversely, for negative s (visiting team favored), the visiting team covers the spread if m < s , and the home team has beat the spread otherwise. The home (visiting) team is said to win “against the spread” if m − s is positive (negative).

Denote the profit (on a unit bet) when correctly wagering on the home and visiting teams by ϕ h and ϕ v , respectively. Assuming a bet size of b placed on the home team, the conventional payout structure is to award the bettor with b (1 + ϕ h ) when m > s . The entire wager is lost otherwise. The total profit π is thus bϕ h when correctly wagering on the home team (− b otherwise). When placing a bet of b on the visiting team, the bettor receives b (1 + ϕ v ) if m < s and 0 otherwise. Typical values of ϕ h and ϕ v are 100/110 ≈ 0.91, corresponding to a commission of 4.5% charged by the sportsbook.

In practice, the event m = s (termed a “push”) may have a non-zero probability and results in all bets being returned. In keeping with the modeling of m by a continuous random variable, here it is assumed that P ( m = s ) = 0. This significantly simplifies the development below. Note also that for fractional spreads (e.g. s = 3.5), the probability of a push is indeed zero.

Wagering to maximize expected profit.

Consider first the question of which team to wager on to maximize the expected profit. As the profit scales linearly with b , a unit bet size is assumed without loss of generality.

Corollary 1 . Assuming equal payouts for home and visiting teams ( ϕ h = ϕ v ), maximization of expected profit is achieved by wagering on the home team if and only if the spread is less than the median margin of victory .

A subtle but important point is that knowledge of which side to bet on for each match is insufficient for maximizing overall profit. The reason is that even if wagering on the side with higher expected profit, it is possible (and in fact quite common, see empirical results below) that the “optimal” wager carries a negative expectation. Thus, an understanding of when wagering should be avoided altogether is required. This is the subject of the theorem below.

It is instructive to consider the conditions above for typical values of ϕ h and ϕ v . When wagering on the home team with ϕ h = 0.91, positive expectation requires the spread to be no larger than the 0.476 quantile of m . When wagering on the visiting team, the spread must exceed the 0.524 quantile. This means that, if the spread is contained within the 0.476-0.524 quantiles of the margin of victory, wagering should be avoided . Practically, it is thus important to obtain estimates of this interval and its proximity to the median score in units of points .

The result of Theorem 2 is reminiscent of the “area of no profitable bet” scenario described in [ 28 ]. Whereas the latter result is presented in terms of outcome probabilities estimated by the bettor and the sportsbook, Theorem 2 here delineates the conditions under which the sportsbook’s point spread assures a negative expectation on the bettor’s side.

Optimal estimation of the margin of victory.

Theorem 3 . Define an “error” as a wager that is placed on the team that loses against the spread. The probability of error is bounded according to : min{ F m ( s ), 1 − F m ( s )} ≤ p (error) ≤ max{ F m ( s ), 1 − F m ( s )}.

The result of Theorem ( 8 ) provides both the best- and worst-case scenario of a given wager. When F m ( s ) is close to 1/2, both the minimum and maximum error rates are near 50%, and wagering is reduced to an event akin to a coin flip. On the other hand, when the true median is far from the spread (i.e., F m ( s ) deviates from 1/2), the minimum and maximum error rates diverge, increasing the highest achievable accuracy of the wager.

Optimality in “moneyline” wagering

Corollary 4 . Define an “error” as a wager that is placed on the team that loses the match outright. The probability of error in moneyline wagering is bounded according to : min{ F m (0), 1 − F m (0)} ≤ p (error) ≤ max{ F m (0), 1 − F m (0)}.

Notice that optimal decision-making in moneyline wagers requires knowledge of quantiles that may be near 0 (if ϕ v ≫ ϕ h ) or near 1 (if ϕ h ≫ ϕ v ). More subtly, the required quantiles will differ for matches that exhibit different payout ratios. For example, a match with two even sides will require knowledge of central quantiles, while a match with a 4:1 favorite will require knowledge of the 80th and 20th percentiles. The implications of this property on quantitative modeling are described in the Discussion .

The moneyline wagering considered in this section is a two-alternative bet that is popular in North American sports. In European betting markets, the most common type of wager is the three-alternative “Home-Draw-Away” bet where there is no point spread and the task of the bettor is to forecast one of the three potential outcomes: m > 0, m = 0, or m < 0, each of which are endowed with a payout (see, for example, [ 26 , 29 , 30 ]). Clearly the the probability p ( m = 0) will be non-zero in this context. As a result, the methodology here, which models m by a continuous random variable, cannot be straightforwardly applied to the case of the Home-Draw-Away bet. The extension of the present findings to the case of multi-way bets with discrete m is a potential topic of future research.

Optimality in “over-under” betting

The following two results may be proven by replacing m with τ , ϕ h with ϕ o , and ϕ v with ϕ u in the Proofs of Theorems 1 and 2, respectively.

In the special case of ϕ o = ϕ u , one should bet on the over only if and only if the sportsbook total τ falls below the median of t .

Define F t ( τ ) as the CDF of the true point total evaluated at the sportsbook’s proposed total. The following corollary may be proven by following the Proof of Theorem 3.

Corollary 8 . Define an “error” in over-under betting as a wager that is placed on the “over” when t < τ or on the “under” when t > τ. The probability of error is bounded according to : min{ F t ( τ ), 1 − F t ( τ )} ≤ p (error) ≤ max{ F t ( τ ), 1 − F t ( τ )}.

Empirical results from the National Football League

In order to connect the theory to a real-world betting market, empirical analyses utilizing historical data from the National Football League (NFL) were conducted. The margins of victory, point totals, sportsbook point spreads, and sportsbook point totals were obtained for all regular season matches occurring between the 2002 and 2022 seasons ( n = 5412). The mean margin of victory was 2.19 ± 14.68, while the mean point spread was 2.21 ± 5.97. The mean point total was 44.43 ± 14.13, while the mean sportsbook total was 43.80 ± 4.80. The standard deviation of the margin of victory is nearly 7x the mean, indicating a high level of dispersion in the margin of victory, perhaps due to the presence of outliers. Note that the standard deviation of a random variable provides an upper bound on the distance between its mean and median [ 31 ], which is relevant to the problem at hand.

To estimate the distribution of the margin of victory for individual matches, the point spread s was employed as a surrogate for θ . The underlying assumption is that matches with an identical point spread exhibit margins of victory drawn from the same distribution. Observations were stratified into 21 groups ranging from s o = −7 to s o = 10. This procedure was repeated for the analysis of point totals, where observations were stratified into 24 groups ranging from t o = 37 to t o = 49.

How accurately do sportsbooks capture the median outcome?

It is important to gain insight into how accurately the point spreads proposed by sportsbooks capture the median margin of victory. For each stratified sample of matches, the median margin of victory was computed and compared to the sample’s point spread. The distribution of margin of victory for matches with a point spread s o = 6 is shown in Fig 1a , where the sample median of 4.34 (95% confidence interval [2.41,6.33]; median computed with kernel density estimation to overcome the discreteness of the margin of victory; confidence interval computed with the bootstrap) is lower than the sportsbook point spread. However, the sportsbook value is contained within the 95% confidence interval.

• PPT PowerPoint slide
• PNG larger image
• TIFF original image

( a ) The distribution of margin of victory for National Football League matches with a consensus sportsbook point spread of s = 6. The median outcome of 4.26 (dashed orange line, computed with kernel density estimation) fell below the sportsbook point spread (dashed blue line). However, the 95% confidence interval of the sample median (2.27-6.38) contained the sportsbook proposition of 6. ( b ) Same as (a), but now showing the distribution of point total for all matches with a sportsbook point total of 46. Although the sportsbook total exceeded the median outcome by approximately 1.5 points, the confidence interval of the sample median (42.25-46.81) contained the sportsbook’s proposition. ( c ) Combining all stratified samples, the sportsbook’s point spread explained 86% of the variability in the median margin of victory. The confidence intervals of the regression line’s slope and intercept included their respective null hypothesis values of 1 and 0, respectively. ( d ) The sportsbook point total explained 79% of the variability in the median total. Although the data hints at an overestimation of high totals and underestimation of low totals, the confidence intervals of the slope and intercept contained the null hypothesis values.

https://doi.org/10.1371/journal.pone.0287601.g001

Aggregating across stratified samples, the sportsbook point spread explained 86% of the variability in the true median margin of victory ( r 2 = 0.86, n = 21; Fig 1c ). Both the slope (0.93, 95% confidence interval [0.81,1.04]) and intercept (-0.41, 95% confidence interval [-1.03,0.16]) of the ordinary least squares (OLS) line of best fit (dashed blue line) indicate a slight overestimation of the margin of victory by the point spread. This is most apparent for positive spreads (i.e., a home favorite). Nevertheless, the confidence intervals of both the slope and intercept did include the null hypothesis values of 1 and 0, respectively. The data for all sportsbook point spreads with at least 100 matches is provided in Table 1 .

Regular season matches from the National Football League occurring between 2002-2022 were stratified according to their sportsbook point spread. Each set of 3 grouped rows corresponds to a subsample of matches with a common sportsbook point spread. The “level” column indicates whether the row pertains to the 95% confidence interval (0.025 and 0.975 quantiles) or the mean value across bootstrap resamples. The dependent variables include the 0.476, 0.5, and 0.524 quantiles, as well as the expected profit of wagering on the side with higher likelihood of winning the bet for hypothetical point spreads that deviate from the median outcome by 1, 2, and 3 points, respectively.

https://doi.org/10.1371/journal.pone.0287601.t001

The distribution of observed point totals for matches with a sportsbook total of τ = 46 is shown in Fig 1b , where the computed median of 44.45 (95% confidence interval [42.25,46.81]) is suggestive of a slight overestimation of the true total. Combining data from all samples, the sportsbook point total explained 79% of the variability in the median point total ( r 2 = 0.79, n = 24; Fig 1d ).

Interestingly, the data hints at the sportsbook’s proposed point total underestimating the true total for relatively low totals (i.e., black line is below the blue for sportsbook totals below 43), while overestimating the total for those matches expected to exhibit high scoring (i.e., black line is above the blue line for sportsbook totals above 43). Note, however, that the confidence intervals of the regression line (slope: [0.72,1.02], intercept: [-1.14, 12.05]) did contain the null hypothesis values. The data for all sportsbook point total with at least 100 samples is provided in Table 2 .

Matches were stratified into 24 subsamples defined by the value of the sportsbook total. The dependent variables are the 0.476, 0.5, and 0.524 quantiles of the true point total, as well as the expected profit of wagering conditioned on the amount of bias in the sportsbook’s total.

https://doi.org/10.1371/journal.pone.0287601.t002

Do sportsbook estimates deviate from the 0.476-0.524 interval?

In the common case of ϕ = 0.91, a positive expected profit is only feasible if the point spread (or point total) is either below the 0.476 or above the 0.524 quantiles of the outcome’s distribution. It is thus interesting to consider how often this may occur in a large betting market such as the NFL. To that end, the 0.476 and 0.524 quantiles of the margin of victory were estimated in each stratified sample (horizontal bars in Fig 2 ; the point spread is indicated with an orange marker; all quantiles are listed in Table 1 ).

With a standard payout of ϕ = 0.91, achieving a positive expected profit is only feasible if the sportsbook point spread falls outside of the 0.476-0.524 quantiles of the margin of victory. The 0.476 and 0.524 quantiles were thus estimated for each stratified sample of NFL matches. Light (dark) black bars indicate the 95% confidence intervals of the 0.476 (0.524) quantiles. Orange markers indicate the sportsbook point spread, which fell within the quantile confidence intervals for the large majority of stratifications. An exception was s = 5, where the sportsbook appeared to overestimate the margin of victory. For two other stratifications ( s = 3 and s = 10), the 0.524 quantile tended to underestimate the sportsbook spread, with the 95% confidence intervals extending to just above the spread.

https://doi.org/10.1371/journal.pone.0287601.g002

For the majority of samples, the confidence intervals of the 0.476 and 0.524 quantiles contained the sportsbook spread. One exception was the spread s = 5, where the margin of victory fell below the sportsbook value (95% confidence interval of the 0.524 quantile: [0.87,4.85]). The margin of victory for s = 3 (95% confidence interval of the 0.524 quantile: [0.78,3.08]) and s = 10 (95% confidence interval of the 0.524 quantile: [6.42,10.06]) also tended to underestimate the sportsbook spread, with the confidence intervals just containing the sportsbook value.

The analysis was repeated for point totals ( Fig 3 , all quantiles listed in Table 2 ). All but one stratified sample exhibited 0.476 and 0.524 quantiles whose confidence intervals contained the sportsbook total ( t = 47, [41.59, 45.42]). Examination of the sample quantiles suggests that NFL sportsbooks are very adept at proposing point totals that fall within 2.4 percentiles of the median outcome.

The 0.476 and 0.524 quantiles of the true point total were estimated for each stratified sample of NFL matches. For all but one stratification ( t = 47, 95% confidence interval [41.59-45.42], sportsbook overestimates the total), the confidence intervals of the sample quantiles contained the sportsbook proposition. Visual inspection of the data suggests that, in the NFL betting market at least, sportsbooks are very adept at proposing totals that fall within the critical 0.476-0.524 quantiles.

https://doi.org/10.1371/journal.pone.0287601.g003

How large of a discrepancy from the median is required for profit?

In practice, it is desirable to have an understanding of how large of a sportsbook bias, in units of points, is required to permit a positive expected profit. To address this, the value of the empirically measured CDF of the margin of victory was evaluated at offsets of 1, 2, and 3 points from the true median in each direction. The resulting value was then converted into the expected value of profit (see Materials and Methods ). The computation was performed separately within each stratified sample, and the height of each bar in Fig 4 indicates the hypothetical expected profit of a unit bet when wagering on the team with the higher probability of winning against the spread . For the sake of clarity, only the four largest samples ( s ∈ {−3, 2.5, 3, 7}) are shown in the Figure, with data for all samples listed in Table 1 .

In order to estimate the magnitude of the deviation between sportsbook point spread and median margin of victory that is required to permit a positive profit to the bettor, the hypothetical expected profit was computed for point spreads that differ from the true median by 1, 2, and 3 points in each direction. The analysis was performed separately within each stratified sample, and the figure shows the results of the four largest samples. For 3 of the 4 stratifications, a sportsbook bias of only a single point is required to permit a positive expected return (height of the bar indicates the expected profit of a unit bet assuming that the bettor wagers on the side with the higher probability of winning; error bars indicate the 95% confidence intervals as computed with the bootstrap). For a sportsbook spread of s = 3 (dark black bars), the expected profit on a unit bet is 0.021 [0.008-0.035], 0.094 [0.067-0.119], and 0.166 [0.13-0.2] when the sportsbook’s bias is +1, +2, and +3 points, respectively (mean and confidence interval over 500 bootstrap resamples).

https://doi.org/10.1371/journal.pone.0287601.g004

The expected profit is negative (i.e., ( ϕ − 1)/2 = −0.045) when the spread equals the median (center column). Interestingly however, for 3 of the 4 largest stratified samples, a positive profit is achievable with only a single point deviation from the median in either direction (the confidence intervals indicated by error bars do not extend into negative values). Averaged across all n = 21 stratifications, the expected profit of a unit bet is 0.022 ± 0.011, 0.090 ± 0.021, and 0.15 ± 0.030 when the spread exceeds the median by 1, 2, and 3 points, respectively (mean ± standard deviation over n = 21 stratifications, each of which is an average over 1000 bootstrap ensembles). Similarly, the expected return is 0.023 ± 0.013, 0.089 ± 0.026, and 0.15 ± 0.037 when the spread undershoots the median by 1, 2, and 3 points respectively. This indicates that sportsbooks must estimate the median outcome with high precision in order to prevent the possibility of positive returns.

The analysis was repeated on the data of point totals. A deviation from the true median of only 1 point was sufficient to permit a positive expected profit in all four of the largest stratifications ( Fig 5 ; t ∈ {41, 43, 44, 45}; error bars indicate 95% confidence intervals; data for all samples is provided in Table 2 ). When the sportsbook overestimates the median total by 1, 2, and 3 points, the expected profit on a unit bet is 0.014 ± 0.0071, 0.073 ± 0.014, and 0.13 ± 0.020, respectively (mean ± standard deviation over n = 24 samples, each of which is a average over 1000 bootstrap resamples). When the sportsbook underestimates the median, the expected profit on a unit bet is 0.015±0.0071, 0.076± 0.014, and 0.14± 0.020, for deviations of 1, 2, and 3 points, respectively. Note that despite the dependent variable having a larger magnitude (compared to margin of victory), the required sportsbook error to permit positive profit is the same as shown by the analysis of point spreads.

Vertical axis depicts the expected profit of an over-under wager, conditioned on the sportsbook’s posted total deviating from the true margin by a value of 1, 2, or 3 points (horizontal axis). The analysis was performed separately for each unique sportsbook total, and the figure displays the results for the four largest samples. A deviation from the true median of a single point permits a positive expected profit in all four of the depicted groups. For a sportsbook total of t = 44 (green bars), the expected profit on a unit bet is 0.015 [0.004-0.028], 0.075 [0.053-0.10], and 0.13 [0.10-0.17] when the sportsbook’s bias is +1, +2, and +3 points, respectively (mean and confidence interval over 500 bootstrap resamples).

https://doi.org/10.1371/journal.pone.0287601.g005

The theoretical results presented here, despite seemingly straightforward, have eluded explication in the literature. The central message is that optimal wagering on sports requires accurate estimation of the outcome variable’s quantiles. For the two most common types of bets—point spread and point total—estimation of the 0.476, 0.5 (median), and 0.524 quantiles constitutes the primary task of the bettor (assuming a standard commission of 4.5%). For a given match, the bettor must compare the estimated quantiles to the sportsbook’s proposed value, and first decide whether or not to wager ( Theorem 2 ), and if so, on which side ( Theorem 1 ).

The sportsbook’s proposed spread (or point total) effectively delineates the potential outcomes for the bettor ( Theorem 3 ). For a standard commission of 4.5%, the result is that if the sportsbook produces an estimate within 2.4 percentiles of the true median outcome, wagering always yields a negative expected profit— even if consistently wagering on the side with the higher probability of winning the bet . This finding underscores the importance of not wagering on matches in which the sportsbook has accurately captured the median outcome with their proposition. In such matches, the minimum error rate is lower bounded by 47.6%, the maximum error rate is upper bounded by 52.4%, and the excess error rate ( Theorem 4 ) is upper bounded by 4.8%.

The seminal findings of Kuypers [ 13 ] and Levitt [ 23 ], however, imply that sportsbooks may sometimes deliberately propose values that deviate from their estimated median to entice a preponderance of bets on the side that maximizes excess error. For example, by proposing a point spread that exaggerates the median margin of victory of a home favorite, the minimum error rate may become, for example, 45% (when wagering on the road team), and the excess error rate when wagering on the home team is 10%. In this hypothetical scenario, the sportsbook may predict that, due to the public’s bias for home favorites, a majority of the bets will be placed on the home team. The empirical data presented here hint at this phenomenon, and are in alignment with previous reports of market inefficiencies in the NFL betting market [ 5 , 32 – 35 ]. Namely, the sportsbook point spread was found to slightly overestimate the median margin of victory for some subsets of the data ( Fig 2 ). Indeed, the stratifications showing this trend were home favorites, agreeing with the idea that the sportsbooks are exploiting the public’s bias for wagering on the favorite [ 23 ].

The analysis of sportsbook point spreads performed here indicates that only a single point deviation from the true median is sufficient to allow one of the betting options to yield a positive expectation. On the other hand, realization of this potential profit requires that the bettor correctly, and systematically, identify the side with the higher probability of winning against the spread. Forecasting the outcomes of sports matches against the spread has been elusive for both experts and models [ 6 , 36 ]. Due to the abundance of historical data and user-friendly statistical software packages, the employment of quantitative modeling to aid decision-making in sports wagering [ 37 ] is strongly encouraged. The following suggestions are aimed at guiding model-driven efforts to forecast sports outcomes.

The argument against binary classification for sports wagering

The minimum error and minimum excess error rates defined in Theorems 3 and 4, respectively, are analogous to the Bayes’ minimum risk and Bayes’ excess risk in binary classification [ 38 ]. Indeed, one can cast the estimation of margin of victory in sports wagering as a binary classification problem, aiming to predict the event of “the home team winning against the spread”. Here this approach is not advocated. In conventional binary classification, the target variable (or “class label”) is static and assumed to represent some phenomenon (e.g. presence or absence of an object). In the context of sports wagering, however, the event m > s need not be uniform for different matches. For example, the event of a large home favorite winning against the spread may differ qualitatively from that of a small home “underdog” winning against the spread. Moreover, the sportsbook’s proposed point spread is a dynamic quantity. To illustrate the potential difficulty of utilizing classifiers in sports wagering, consider the case of a match with a posted spread of s = 4, where the goal is to predict the sign of m − 4. But now imagine that the the spread moves to s = 3. The resulting binary classification problem is now to predict the sign of m − 3, and it is not straightforward to adapt the previously constructed classifier to this new problem setting. One may be tempted to modify the bias term of the classifier, but it is unclear by how much it should be adjusted, and also whether a threshold adjustment is in fact the optimal approach in this scenario. On the other hand, by posing the problem as a regression, it is trivial to adapt one’s optimal decision: the output of the regression can simply be compared to the new spread.

The case for quantile regression

Conventional ordinary least-squares (OLS) regression yields estimates of the mean of a random variable, conditioned on the predictors. This is achieved by minimizing the mean squared error between the predicted and target variable.

The findings presented here suggest that conventional regression may be a sub-optimal approach to guiding wagering decisions, whose optimality relies on knowledge of the median and other quantiles. The presence of outliers and multi-modal distributions, as may be expected in sports outcomes, increases the deviation between the mean and median of a random variable. In this case, the dependent variable of conventional regression is distinct from the median and thus less relevant to the decision-making of the sports bettor. The significance of this may be exacerbated by the high noise level on the target random variable, and the low ceiling on model accuracy that this imposes.

Therefore, a more suitable approach to quantitative modeling in sports wagering is to employ quantile regression, which estimates a random variable’s quantiles by minimizing the quantile loss function [ 39 ]. Any features that are expected to forecast sports outcomes could be provided as the predictors in a quantile regression to produce estimates that are aligned with the bettor’s objectives: to avoid wagering on matches with negative expectation for both outcomes, and to wager on the side with zero excess error.

Potential challenges in moneyline wagering

Bias-variance in sports wagering

The view that low variance implies “simple” models has recently been challenged in the context of artificial neural networks [ 41 ]. Nevertheless, the desire for low-variance, high-bias modeling in sports wagering does suggest the preference for simpler models. Thus, it is advocated to employ a limited set of predictors and a limited capacity of the model architecture. This is expected to translate to improved generalization to future data.

Sport-specific considerations

The three types of wagers considered in this work—point spread, moneyline, and over-under—are the most popular bet types in North American sports. The empirical analysis employed data from the National Football League (NFL). One unique aspect of American football is its scoring system, in which the points accumulated by each team increase primarily in increments of 3 or 7 points. The structure of the scoring imposes constraints on the distribution of the margin of victory m . For example, in American Football, the distribution of the margin of victory is expected to exhibit local maxima near values such as: ±3, ±7, ±10. In the case of games in the National Basketball Association (NBA), the most common margins of victory tend to occur in the 5-10 interval, reflecting the overall higher point totals in basketball and its most common point increments (2 and 3). As a result, the shape and quantiles of the distribution of m may vary qualitatively between the NBA and NFL.

As a final illustrative example of the importance of the quantiles of m , consider the hypothetical scenario of two American football teams playing a match whose parameters θ have been exactly matched three times previously. In those past matches, the outcomes were m = 3, m = 7, and m = 35. In this fictitious example, the median is 7 but the mean is 15. Now imagine that the point spread for the next match has been set to s = 10 (home team favored to win the match by 10 points). Assuming that one has committed to wagering on the match, the optimal decision is to bet on the visiting team, despite that fact that the home team has won the previous matches by an average of 15 points.

Materials and methods

All analysis was performed with custom Python code compiled into a Jupyter Notebook (available at https://github.com/dmochow/optimal_betting_theory ). The figures and tables in this manuscript may be reproduced by executing the notebook.

Empirical data

Historical data from the National Football League (NFL) was obtained from bettingdata.com , who has courteously permitted the data to be shared on the repository listed above. All regular season matches from 2002 to 2022 were included in the analysis ( n = 5412). The data set includes point spreads and point totals (with associated payouts) from a variety of sportsbooks, as well as a “consensus” value. The latter was utilized for all analysis.

Data stratification

In order to estimate quantiles of the distributions of margin of victory and point totals from heterogeneous data (i.e., matches with disparate relative strengths of the home and visiting teams), the sportsbook point spread and sportsbook point total were used as a surrogate for the parameter vector defining the identity of each individual match ( θ in the text). This permitted the estimation of the 0.476, 0.5, and 0.524 quantiles over subsets of congruent matches.

Only spreads or totals with at least 100 matches in the dataset were included, such that estimation of the median would be sufficiently reliable. To that end, data was stratified into 21 samples for the analysis of margin of victory: {-7, -6, -3.5, -3, -2.5, -2, -1, 1, 2, 2.5, 3, 3.5, 4, 4.5, 5, 5.5, 6, 6.5, 7, 7.5, 10) and 24 samples for the analysis of point totals (37, 37.5, 38, 39, 39.5, 40, 40.5, 41, 41.5, 42, 42.5, 43, 43.5, 44, 44.5, 45, 45.5, 46, 46.5, 47, 47.5, 48, 48.5, 49 }. This resulted in the employment of n = 3843 matches in the analysis of point spreads and n = 4300 matches in the analysis of point totals.

Note that the stratification process did not account for varying payouts, for example −110 versus −105 in the American odds system, as this would greatly increase the number of stratified samples while decreasing the number of matches in each sample. It is likely that the resulting error is negligible, however, due to the likelihood of the payout discrepancy being fairly balanced across the home and visiting teams.

Median estimation

In order to overcome the discrete nature of the margins of victory and point totals, kernel density estimation was employed to produce continuous quantile estimates. The KernelDensity function from the scikit-learn software library was employed with a Gaussian kernel and a bandwidth parameter of 2. For the margin of victory, the density was estimated over 4000 points ranging from -40 to 40. For the analysis of point totals, the density was estimated over 4000 points ranging from 10 to 90. The regression analysis relating median outcome to sportsbook estimates ( Fig 1 ) was performed with ordinary least squares (OLS).

Confidence interval estimation

In order to generate variability estimates for the 0.476, 0.5, and 0.524 quantiles of the margin of victory and point total, the bootstrap [ 42 ] technique was employed. 1000 resamples of the same size as the original sample were generated in each case. The confidence intervals were then constructed as the interval between the 2.5 and 97.5 percentiles of the relevant quantity. Bootstrap resampling was also employed to derive confidence intervals on the regression parameters relating the median outcomes to sportsbook spreads or totals ( Fig 1 ), as well as the confidence intervals on the expected profit of wagering conditioned on a fixed sportsbook bias (Figs 4 and 5 ).

Expected profit estimation

To model the idealized case of always placing the wager on the side with the higher probability of winning against the spread, the reported expected profit was taken as the maximum of the two expected values in ( 15 ). The analogous procedure was conducted for the analysis of point totals.

Acknowledgments

The author would like to thank Ed Miller and Mark Broadie for fruitful discussions during the preparation of the manuscript. The author would also like to acknowledge the effort of the reviewers, in particular Fabian Wunderlich, for providing many helpful comments and critiques throughout peer review.

• View Article
• PubMed/NCBI
• Google Scholar
• 2. Bloomberg Media. Sports Betting Market Size Worth $140.26 Billion By 2028: Grand View Research, Inc.; 2021. Available from: https://www.bloomberg.com/press-releases/2021-10-19/sports-betting-market-size-worth-140-26-billion-by-2028-grand-view-research-inc .
• 21. Glickman ME, Stern HS. A state-space model for National Football League scores. In: Anthology of statistics in sports. SIAM; 2005. p. 23–33.
• 38. Devroye L, Györfi L, Lugosi G. A probabilistic theory of pattern recognition. vol. 31. Springer Science & Business Media; 2013.
• 39. Koenker R, Chernozhukov V, He X, Peng L. Handbook of quantile regression. 2017;.
• 41. Neal B, Mittal S, Baratin A, Tantia V, Scicluna M, Lacoste-Julien S, et al. A modern take on the bias-variance tradeoff in neural networks. arXiv preprint arXiv:181008591. 2018;.

• Follow us on Facebook
• Follow us on Twitter
• Criminal Justice
• Environment
• Politics & Government
• Race & Gender

Expert Commentary

Sports betting in the US: A research roundup and explainer

We look at the landscape of legal sports betting in America, explain what the research says about how legalization affects tax revenues, and provide a brief history of the activity.

Republish this article

This work is licensed under a Creative Commons Attribution-NoDerivatives 4.0 International License .

by Clark Merrefield, The Journalist's Resource October 25, 2022

This <a target="_blank" href="https://journalistsresource.org/economics/sports-betting-research-roundup-explainer/">article</a> first appeared on <a target="_blank" href="https://journalistsresource.org">The Journalist's Resource</a> and is republished here under a Creative Commons license.<img src="https://journalistsresource.org/wp-content/uploads/2020/11/cropped-jr-favicon-150x150.png" style="width:1em;height:1em;margin-left:10px;">

On Nov. 8, Californians will vote on two ballot measures that would allow for different forms of sports betting in the state.

Proposition 26 would allow sports betting at licensed casinos and horse tracks on tribal lands and run by federally recognized Native American tribes.

Prop. 27 would allow tribes licensed to offer gambling and major gaming companies to offer online sports betting. These companies include FanDuel and DraftKings, which together make up roughly two-thirds of the U.S. online sports betting market.

“If both pass, they might both go into effect or the result could be decided in court, depending on which one gets more yes votes,” writes CalMatters economics reporter Grace Gedye in an article from June.

Although California is the only state with sports betting on the midterm ballot , it’s not the only state where sports betting is a topic of political discussion — and relevant for journalists across beats to understand. For example, Georgia gubernatorial candidate Stacey Abrams recently expressed support for legalized sports betting in her state. Abrams’ opponent, Gov. Brian Kemp, is opposed.

In Missouri, state lawmakers from both parties support legalizing sports betting , but Gov. Mike Parson is hesitant. Vermont lawmakers are considering taking up a sports betting bill during the next legislative session. Gubernatorial candidates in South Carolina and Texas support legal sports betting. In Florida, there is an ongoing lawsuit over whether the state should be allowed to give the Seminole Tribe the exclusive right to run online sports betting there.

Legal sports wagering in the U.S. has grown vertically in recent years — from less than $5 billion worth of bets placed in 2018 to $57 billion in 2021 — despite sports betting remaining illegal in nearly half of states. Sportsbooks, the entities that take sports bets, bring in about $4 billion yearly after wagers are settled.

The reason for this growth: a May 2018 Supreme Court ruling. Justice Samuel Alito, in delivering the 6-3 decision , reasoned that 1992 federal legislation banning states from allowing sports betting was unconstitutional.

Under the 1992 law, the Professional and Amateur Sports Protection Act, the federal government did not “make sports gambling itself federal crime,” Alito writes in the 2018 decision. “Instead, it allows the [U.S.] Attorney General, as well as professional and amateur sports organizations, to bring civil actions to enjoin violations.” Other than legislative powers the Constitution grants Congress, the federal government cannot “issue direct orders to state legislatures,” Alito writes. The majority interpreted the Professional and Amateur Sports Protection Act as doing so.

Here is how John Holden , an assistant professor of business at Oklahoma State University who has written extensively on sports gambling, explains the 2018 ruling:

“If the federal government wants to make sports betting illegal, they’re free to do so, but they can use the [Federal Bureau of Investigation] and the [Department of Justice] to enforce that,” Holden says. “They can’t tell a state legislature that you need to keep that law on the books and use your state police to go out and bust up gambling rings.”

A fundraising breakdown from the Los Angeles Times shows about $132 million has been raised to support Prop. 26, with about $43 million in opposition funding. Top backers include the Federated Indians of Graton Rancheria, the Pechanga Band of Indians and the Yocha Dehe Wintun Nation. Non-Native American casino and gaming interests are largely opposed — conversely, they have backed Prop. 27, which would open up sports betting to all gambling interests, not just Native American-run casinos.

Tribal gaming brings in nearly $40 billion a year across all tribes that operate gambling enterprises, according to the National Indian Gaming Commission . “Gaming operations have had a far-reaching and transformative effect on American Indian reservations and their economies,” write the authors of a 2015 paper about how the act affected tribal economic development. “Specifically, Indian gaming has allowed marked improvements in several important dimensions of reservation life.”

The landscape of legal sports betting

If California legalizes sports betting, it would represent a major coup for gaming interests in the state. In California, the most populous state, horse racing is the only legal form of sports betting.

Sports wagering is legal in 28 states plus the District of Columbia, according to a recent Washington Post analysis. Seven states prohibit online sports betting and only allow in-person wagers at licensed locations, such as casinos: Delaware, Maryland, Mississippi, Montana, New Mexico, North Carolina and North Dakota. Sports betting is legal but pending rollout in four states: Maine, Massachusetts, Nebraska and Ohio. Kansas is the most recent state to implement legal in-person and online sports gambling, as of Sept. 1.

States place a range of licensing fees on operators and tax rates on sports betting revenue, from a low of 6.75% in Nevada and Iowa to a high of 51% in New Hampshire and New York. States use tax revenues for a variety of purposes . Some, like Delaware, put sports wagering taxes toward their general fund. Colorado uses sports betting taxes to pay for its statewide water plan, Illinois funds transportation infrastructure and New York funds education programs.

In states where sports betting is legal, bettors can wager on nearly any major sporting event, both professional and amateur. For example, bettors can wager on the outcome of a baseball game, as well as events within the game, such as whether a particular player will hit a home run.

Polling indicates California may be unlikely to join the legal betting club. CalMatters reported earlier this month that despite various campaigns raising more than $440 million in marketing related to Props. 26 and 27, each measure is garnering support from less than a third of likely voters, according to October polling from the Institute of Governmental Studies at the University of California, Berkeley.

Below, we explore recent research on sports betting. Among the findings of the seven studies featured here:

• Sports bettors are more likely to be white, male, and exhibit psychological traits consistent with narcissism.
• Tax revenue from sports betting may appear substantial in raw numbers, but the impact on tax coffers is muted when compared with income and sales taxes, or tax revenue from other gambling offerings.
• Evidence is mixed as to whether introducing sports betting cannibalizes — eats away at — revenue from other types of gambling.
• Some college football referees may more heavily penalize betting favorites.

The nonprofit National Council on Problem Gambling estimates as many as 8 million adults in the U.S. may have a mild, moderate or severe gambling problem. However, there is a lack of comprehensive, recent academic research on the extent of gambling addiction in the U.S., and the societal costs.

If you feel you may have a problem with gambling you can get help from the National Council on Problem Gambling by call or text at 1-800-522-4700, or online chat at ncpgambling.org/chat .

Research roundup

The Income Elasticity of Gross Sports Betting Revenues in Nevada: Short-Run and Long-Run Estimates Ege Can and Mark Nichols. Journal of Sports Economics, October 2021.

The study: The authors analyze quarterly sports betting data from Nevada covering 1990 to 2019, to explore whether sports betting might be a viable tax revenue stream for other states. Sports betting has been legal in Nevada for decades, so it is the only state with long-run data that can potentially provide insight on the tax base future in states that have legalized sports betting since 2018. The authors note that Nevada is a “mature” market for sports betting, meaning industry growth is relatively stable year to year. A state that newly legalizes sports gambling is likely to see an immediate jump in sports betting revenue, with industry growth levelling off over time.

The findings: In the short-run, quarter-to-quarter, the rise and fall of sports betting revenue in Nevada is most closely tied to changing sports seasons. The authors suggest this is due to differences in how much bettors wager on various sports — the NFL, for example, is “the most popular sport to place wagers on,” with revenues rising and falling as an NFL season begins and ends. In the long run, taxable income in the state and sports betting revenues tend to grow at similar rates. Sports betting revenue in Nevada is a small fraction of revenues from other sources.

The authors write: “Total sports betting revenue in Nevada, the amount kept by the casinos, was $329 million in 2019, implying $22.2 million in tax revenue for the state. In contrast, casino gambling in Nevada in 2019 was $12 billion, generating $810 million in tax revenue. Sports betting is a gambling activity where the amount retained by the casino, and consequently retained by the state, is relatively small as most of the money from losing bets is transferred to those with winning bets. Therefore, sports gambling is a smaller contributor to tax coffers compared to more traditional tax sources such as income and taxable sales or, if applicable, casino revenue.”

A Comparative Analysis of Sports Gambling in the United States Brendan Dwyer, Ted Hayduk III and Joris Drayer. International Journal of Sports Marketing and Sponsorship, August 2022.

The study: The authors explore whether there are psychological differences between bettors and those who do not bet, as well as differences in how closely bettors identify with social institutions, such as religious organizations and far-right or far-left politics.

The authors surveyed 377 bettors and 402 non-betting sports fans from 47 states and explored differences between bettors and non-bettors in states with legal gambling and states where gambling is banned. They also asked about narcissism, which past research has found “is associated to gambling behavior especially as it relates to risky behavior such as participating in illegal gambling,” the authors write. Bettors in the sample were 81% male, compared with 69% of non-bettors. Among bettors, 64% were white and 27% were Black, while 77% of non-bettors were white and 17% were Black.

The findings: In legal gambling states, bettors felt more self-worth than non-bettors, though in states where gambling is illegal the difference in self-worth was almost nil. In legal gambling states, bettors reported a stronger personal identity, “or the importance with which an individual identifies with their relationship and career,” than non-bettors. This relationship flipped in illegal gambling states, with non-bettors showing a stronger personal identity than bettors. In both illegal and legal gambling states, bettors reported slightly higher levels of social uselessness — “an individual’s perceived lack of worth related to social institutions” — than non-bettors, though the gap was wider in illegal gambling states.

The authors write: “Bettors look different and come from different backgrounds and locations. Psychographically, they were clearly more narcissistic. They also indicated a higher social identity and self-worth, yet perceived themselves as less worthy members of important social institutions. In general, sports bettors out consumed non-bettors as it relates sports spectatorship.”

Game Changing Innovation or Bad Beat? How Sports Betting Can Reduce Fan Engagement Ashley Stadler Blank, Katherine Loveland, David Houghton. Journal of Business Research, June 2021.

The study: Legal sports betting means more than $4 billion in additional yearly revenue across the four major sports leagues, according to research the authors cite from the American Gaming Association. At the same time, there may be drawbacks that come with the financial windfall. The authors conduct two studies to explore how sports betting affects fan engagement — the emotional connection fans have with their favorite teams.

The first study included 325 people recruited from Mechanical Turk and focused on betting on a team to win, also called moneyline betting. The second was among 167 Mechanical Turk-recruited participants and focused on prop, or “proposition” bets. Prop bets are bets made on the outcome of some action during the game — whether the next foul ball is caught, missed or goes into the stands, for example. The study is among the first to explore whether there are negative emotional responses from fans related to sports betting.

Participants read a scenario — they were to imagine watching a Major League Baseball game, then randomly they were told they placed either no bet on the game or one of several types of bets. These bets included a $20 bet for the home team to win, along with prop bets. Gaming experts, according to the authors, contend that prop bets can potentially keep fans engaged even if the outcome of the game is obvious — if a team is up by 10 runs by the middle innings, for example. In each study, the participants were asked questions to gauge their emotional investment before and after being told the outcome of the game and their bets. Questions broadly asked about team loyalty, feelings of connectedness to the team and the likelihood participants would watch the team or attend a game, along with other measures of fan engagement.

The findings: In the short run, immediately after a game, the study indicates that betting and losing can decrease fan engagement. Participants who placed no bet were more likely to exhibit loyalty and purchase team-branded merchandise when the team lost, compared with those who placed a moneyline bet. Those who won a prop bet were slightly more likely to be engaged with the team than those who did not bet — but those who lost a prop bet were much less engaged than those who did not bet.

The authors write: “Although industry experts expect sports betting to increase fan engagement, results from two studies do not support this expectation. Instead, we find that when fans lose a bet, positive emotions and subsequent fan engagement decrease.”

College Football Referee Bias and Sports Betting Impact Rhett Brymer, Ryan Rodenberg, Huimiao Zheng and Tim Holcomb. Eastern Economic Journal, January 2021.

The study: The authors explore whether betting lines are related to bias in officiating in the six major Division I college football conferences across 6,598 games from 2005 to 2012. Betting lines indicate whether a sportsbook thinks a game will be close, will favor one team or the other, or be a blowout. The authors note that “college football and basketball are the only major U.S. sports in which conferences have primary managerial responsibility for officials.” If there is a game late in the season with an undefeated team playing a middling team, the conference will benefit financially if the undefeated team wins and goes on to play in a high-profile bowl game. “Referees, as employees of the conferences, are theoretically more likely to have implicit bias favoring the team with higher revenue potential,” the authors write. They use penalty yards per game as a proxy for whether an officiating crew exhibits bias toward one team or the other.

The findings: The authors find signs of bias during in-conference games in two conferences: the Atlantic Coast Conference and the Big East, which reorganized in 2013 and no longer sponsors football. In-conference games are those where two teams from the same conference play each other. In those ACC games where there one team was favored to win by three touchdowns or more, the authors find officials call 6.5 more penalty yards per game against the favorite. In the Big East, the penalty yards increase to 5.7 for the heavy favorite. Further, ACC officials appeared to flag fewer penalty yards against teams that had been in the league longer and enjoyed historic success, rather than newer teams enjoying more recent success. The authors found no officiating bias when an out-of-conference opponent was heavily favored.

The authors write: “… with increasing state regulation, there will likely be more scrutiny of officiating given that a wider spectrum of consumers will have a financial interest in game outcomes. Increased fan engagement via legal sports wagering highlights the importance of pinpointing evidence of bias and undertaking measures to ensure unbiased officiating and game integrity.”

Legalized Sports Betting, VLT Gambling, and State Gambling Revenues: Evidence from West Virginia Brad Humphreys. Eastern Economic Journal, January 2021.

The study: In one of the only studies to examine state-level sports betting revenue after the 2018 Supreme Court ruling, Humphreys looks at sports betting tax revenues in West Virginia and whether gamblers shifted their wagering from video lottery terminal games in casinos to sports betting.

The findings: From September to December 2018, casinos in West Virginia introduced five new sportsbooks, one at each of its licensed casinos. The year before, the state saw a windfall of $253 million in tax revenue from video lottery games. In the year after sports betting was introduced, the author estimates $45.4 million in lost video lottery revenue, with new sports betting revenue pegged at only $2.6 million. The state taxes video lottery revenues at 53.5%, while sports betting revenues are taxed at 10%.

The author writes: “These results should give state policy makers considering legalization of sports betting pause. While new revenue streams from legalized sports betting appear attractive on the surface, states already generate substantial tax revenues from gambling, and the introduction of sports betting to this mix does not leave spending on other forms of gambling untouched.”

Sports Betting’s Impact on Casino Gambling: Cannibalization or Expansion? Ernest Goss and Peyton Miller. University of Illinois Law Review, October 2021.

The study: Another one of few papers to examine how tax revenues and the games bettors played changed after the 2018 ruling, the authors analyze what happened after Iowa allowed sports gambling after August 2019. Iowa casinos that offer sportsbooks pay 6.75% of their sports betting revenue to the state, “a rate tied with Nevada for the lowest nationally,” the authors write. Like in West Virginia, taxes on all other forms of casino gambling are much higher — 22% on revenue over $3 million. The authors do not look at the specific effects of sports betting on other types of gambling, but rather whether there were any changes in overall revenues after August 2019.

The findings: Mobile sports betting and sports betting in casinos did not affect statewide gambling revenues from August 2019 to March 2020. After March, casinos shuttered due to the COVID-19 pandemic.

The authors write: “While these results do not indicate cannibalization within the Iowa gambling market, there are relevant implications for both casinos and the state. Conditions within the state of Iowa may limit the applicability to other states. For example, the varying tax brackets across gambling forms differ from casino taxing in other states.”

Frameworks of Gambling Harms: A Comparative Review and Synthesis Virve Marionneau, Michael Egerer and Susanna Raisamo. Addiction Research and Theory, August 2022.

The study: The authors gather and analyze “harm frameworks” related to problem gambling. A framework in this context refers to a way of categorizing and thinking about an issue with an ultimate goal of understanding the issue in a comprehensive way and finding solutions. A harm is simply an outcome that most of society would classify as negative — losing one’s house, for example, because of gambling-related losses.

While not specifically related to sports betting, the frameworks explored in the paper are useful for those who want to better understand what can happen to individuals and families affected by problem gambling. After searching major academic research databases, the authors settled on seven papers published between 2000 and 2021 that developed an original harm framework related to problem gambling — four of the papers focused on developing the same framework, leaving four frameworks total. The authors, while applauding the research that has already been done, note that further research is needed.

The findings: Two of the frameworks discussed problem gambling harms related to the workplace and personal relationships. One framework separated psychological and cultural harms, and harms related to crime. Another framework mostly focused on risk factors related to problem gambling, risks which “also occur on the individual, familial, community, and societal levels,” the authors write. They note none of the models explore the degree to which problem gambling harms individuals, families, communities and society — all the harms or risks were “treated as somewhat equal,” they write. Financial harms, they argue, might be a relatively worse harm since they “can be seen to precede or even cause many of the other harms, including criminal acts or emotional suffering.” The authors argue for more research on social harms, where, on the whole, the existence of high levels of problem gambling, “can cause harms irrespective of individual participation, including corruption, economic substitution, match fixing, environmental damage related to tourism, or animal suffering.”

The authors write: “We have found that while existing conceptualizations include a wide definition of harms, most harm items are still seen to stem from individual engagement with gambling. Further incorporation of social and societal harms is still needed to conceptualize and operationalize gambling as a public health issue. This includes the development of societal-level harm measurement and harm minimization.”

A brief history of U.S. sports betting

Americans have a long tradition of gambling. College of the Holy Cross economist Victor Matheson recounts in a January 2021 article in the Eastern Economics Journal:

“Lotteries funded activities such as the original European settlement at Jamestown, the operations of prestigious universities such as Harvard and Princeton, and construction of historic Faneuil Hall in Boston … In the sports realm, by 1900 betting on horse races was made illegal except in Kentucky and Maryland, states that to this day host two of the three Triple Crown events in American horseracing, the Kentucky Derby and the Preakness Stakes. States began to relegalize gambling on horse racing in the 1930s as a method of economic stimulus during the Great Depression.”

By the early 1960s, illegal gambling enterprises run by organized crime groups were worth a combined $7 billion . For more than 30 years, the Wire Act , enacted in September 1961, was the only federal law that addressed sports gambling. The law prohibits the use of a wire — a phone, or, more recently, the internet — to transmit information about placing sports bets across state lines.

The Indian Gaming Regulatory Act, which became federal law in 1988, allowed federally recognized Native American tribes to operate casinos on their land. Sports betting in tribal-run casinos, however, was not allowed unless a tribal-state compact was signed. This is the root of the current legal dispute in Florida. Such compacts were in effect in 22 states as of June 2021, according to the International Center for Gaming Regulation at the University of Nevada, Las Vegas.

By the early 1990s, federal legislators were expressing moral panic over the possibility of states allowing sports betting within their borders, to take advantage of billions being wagered illegally.

Illegal transactions are, by nature, difficult to track. People who bet illegally and their bookies do not typically share receipts with the government or trade groups, so it is difficult to say with precision how big the illegal gambling market was before 2018.

Noting that caveat, the American Gaming Association estimated illegal sports betting as a $150 billion-a-year business before the 2018 Supreme Court ruling. It is an oft-repeated figure in news stories and on websites devoted to sports betting.

In 1991, when overall illegal sports bets were estimated in the tens of billions, a Senate report declared sports gambling a “national problem.” The report continued:

“The harms it inflicts are felt beyond the borders of those states that sanction it. The moral erosion it produces cannot be limited geographically. Once a state legalizes sports gambling, it will be extremely difficult for other States to resist the lure. The current pressures in such places as New Jersey and Florida to institute casino-style sports gambling illustrate the point. Without federal legislation, sports gambling is likely to spread on a piecemeal basis and ultimately develop an irreversible momentum.”

Professional sports league commissioners and former athletes were publicly and adamantly against legal sports betting. They objected that sports integrity would be irreparably harmed, including the possibility of fixed games.

Gary Bettman, the top lawyer for the National Basketball Association, made clear to federal lawmakers in 1990 that sports betting was at odds with the league’s profit motive: “Bettors do not care about the win of their team,” Bettman said during a Senate committee hearing. “They only care about the spread being covered and winning their bets. That is not our product. That is not the product we are selling.”

President George H. Bush signed the Professional and Amateur Sports Protection Act on Oct. 28, 1992. It went into effect on Jan. 1, 1993. Bettman became commissioner of the National Hockey League a month later — and, eventually, a fan of sports betting.

“What we’ve learned is that [sports gambling] is another point of engagement for the fans,” Bettman said during a 2019 American Gaming Association conference. “Ultimately, I think if you’re interested in sports betting, you’re going to have an increased opportunity to engage with the game.”

After 1992, some limited sports betting was grandfathered in Delaware, Oregon and Montana. Delaware, for example, allowed a certain type of bets on National Football League games. States were given one year to legalize casino sports betting after the federal law went into effect, but none did. Nevada was the only grandfathered state that fully allowed sports betting. For nearly three decades, the 1992 federal legislation enshrined Las Vegas as the U.S. sports betting hub.

“[Legal] betting was pretty much happening in Nevada, in a regulated market, and you’d have to be at casino to do it,” says Holden, who wrote a comprehensive overview of the rise of legal sports betting published February 2019 in the Georgia State University Law Review. “That gave us sort of the image of sportsbook-style betting that you would see in the movies, you would see on TV. You go to the counter, you place a bet and you watch the game on 50 different screens.”

Constitutional cracks emerged in the 2010s. In 2014, New Jersey legislators voted to reverse their law banning sports betting there. The National Collegiate Athletic Association brought the state to court. This was the case the Supreme Court heard years later, leading to the fall of the Professional and Amateur Sports Protection Act.

Major sports leagues today are on board with gambling. It is impossible to watch professional sports without encountering advertisements encouraging betting. Online sportsbooks spend $154 million yearly in local TV spots, according to Nielsen. Aside from accepting ad dollars from sportsbooks, every major sports league and numerous individual teams have lucrative partnership deals with sportsbooks.

About The Author

Clark Merrefield

Why Do Individuals Engage in In-Play Sports Betting? A Qualitative Interview Study

• Original Paper
• Open access
• Published: 13 August 2020
• Volume 37 , pages 221–240, ( 2021 )

Cite this article

You have full access to this open access article

• Elizabeth A. Killick   ORCID: orcid.org/0000-0002-0576-8081 1 &
• Mark D. Griffiths 1  

27k Accesses

14 Citations

45 Altmetric

Explore all metrics

Increasing technological advancements and changing consumer behavior has resulted in individuals having access to a wider range of online gambling markets and sporting events than ever before. Sports betting in real time has been aided by the accessibility of smartphone devices. Consequently, the popularity of live sports betting (i.e., ‘in-play’ betting) has spread across Europe and around the rest of world. The aim of the present exploratory study was to examine attitudes and opinions towards online sports betting. Qualitative interviews were conducted with 17 males and 2 females aged between 21 and 32 years. Participants were asked a range of semi-structured interview questions based on pre-determined topic areas. Socio-demographic data were collected and the Problem Gambling Severity Index (PGSI) was used to assess problem gambling. The data were analyzed using thematic analysis in order to identify themes. Analysis of the transcripts identified several notable areas including the ease of engaging in in-play sports betting, motivations for engaging in in-play sports betting (including increased excitement, demonstrating knowledge/skill and response to live odds), and different reasons for using the ‘cash-out’ feature. The findings will contribute to the design of future research investigating in-play sports betting behaviours.

Similar content being viewed by others

The Psychology of Mukbang Watching: A Scoping Review of the Academic and Non-academic Literature

The impact of sports participation on mental health and social outcomes in adults: a systematic review and the ‘Mental Health through Sport’ conceptual model

Online Gambling Addiction: the Relationship Between Internet Gambling and Disordered Gambling

Avoid common mistakes on your manuscript.

Introduction

In recent years there have been many changes in the way that consumers behave and interact with gambling products. There is a continuous stream of technological development and new features being introduced to the gambling market, especially in the UK sports betting market. Mobile technology has been paramount in contributing to the rise in popularity of online sports betting due to that fact that it provides an easy and accessible method of placing sports bets. Traditionally, sports betting took place inside of bookmakers. Now, due to technological change, sports betting can take place online via smartphones, laptops, and tablets in real time and has altered how individuals can place their bets.

In addition, the number of smartphone users has been increasing over the last few years in the UK and over 85% of adults now own a smartphone (Lee & Paul, 2018 ). According to the UK Gambling Commission, almost 30% of online gamblers are using a mobile device to place their bets, and there has been a 10% increase in mobile usage between 2016 and 2017 (Gambling Commission, 2018a , b , c ). Mobile betting allows individuals the capability to bet from almost any location and can also enable individuals to place a range of live bets on different sporting markets. These bets can be made from numerous locations (e.g., work, home, bars, restaurants) with friends or alone.

There has been an increasing conversion of sports betting into an online activity and this increase has been mirrored by a rise in in-play betting. In-play sports betting comprises the wagering of money on something within a sporting event once it has started and up to its conclusion (e.g., who will score next in a soccer match). It is also known as ‘live action’ betting and ‘in-running’ betting. Although it varies from sport to sport, live betting odds are essentially extracted from pre-match odds with (in the case of soccer) the current score, time remaining, and other elements all combined (e.g., the awarding of red and yellow cards, predicting next team or person to score, correct score, the total number of goals, etc.). Over one-quarter of all online gamblers in the UK have placed a bet in-play, with the largest proportion of those by those aged 25–34 years (Gambling Commission, 2018a , b , c ). In-play betting is largely an online activity. Bet365 (the most profitable online British bookmaker) reported that over three-quarters of their sports betting revenue is derived from in-play betting (Barber, 2018 ) and that the most popular sport to bet on is football (soccer).

As well as increased use of mobile device technology, there has been an escalation in the coverage of live soccer matches and other sporting fixtures from around the world. This has resulted in an expansion in the online betting market and an increase in the opportunities to bet on these in-play markets. This is set to continue to rise. The English soccer Premier League showed 200 out of 380 of its matches during 2018–2019, 42 more than 2017–2018 as a result of a new UK broadcast deal. These recent deals were the first time a full round of soccer matches were streamed live in the UK (BBC Sport, 2018 ). Subsequently, there will be an increase in matches for betting consumers to engage with.

Also of note is the ‘cash out’ facility that has been introduced alongside the ability to bet in-play. The ‘cash out’ feature is now offered to sports bettors by many online gambling operators. It enables sports bettors to settle an open bet for a value offered at the time of ‘cashing out’ (Lopez-Gonzalez & Griffiths, 2017 ). This figure is based on the current status of the bet and the statistical likelihood of the bet winning. This figure can also be higher or lower than the initial stake amount.

In recent years, increased attention has been given to researching in-play sports betting. A Gambling Commission ( 2016 ) prevalence survey reported that individuals who bet in-play were more likely to be categorized as problem gamblers. A recent scoping study identified 16 academic papers that had referenced in-play sports betting (Killick & Griffiths, 2019 ) and concluded that in-play sports betting has the potential to be more harmful than more traditional ways of gambling. The review also noted that different research methods had been used to explore this area. One method is the use of behavioral tracking data provided to researchers by online gambling operators. Such research has found that heavily involved gamblers are more likely to bet on in-play events (LaBrie, Laplante, Nelson, Schumann, & Shaffer, 2007 ) and that they increased the frequency of the number of in-play bets being placed after a three-month period (LaPlante, Schumann, LaBrie, & Shaffer, 2008 ). Secondly, some researchers have used self-report methodologies and reported an association between in-play sports betting and risk of problem gambling (Hing, Russell, Vitartas, & Lamont, 2015 ; Lopez-Gonzalez, Estévez, & Griffiths, 2019 ). The review also identified theoretical papers which had discussed the role of the structural characteristics of in-play sports betting. These papers argued that in-play betting had changed traditional sports betting from a discontinuous form of gambling into a more continuous one, and that the increased event frequency of in-play betting would be more likely to have an association with problem gambling than discontinuous (i.e., low event frequency) forms of gambling (Griffiths & Auer, 2013 ; Lopez-Gonzalez & Griffiths, 2017 ).

Two structural characteristics relevant to in-play sports betting and potential problem gambling are bet frequency (the number of bets placed in a particular time frame) and event frequency (how many games/matches are available to bet on in a certain period of time; Griffiths, 2012 ). It has also been argued that problem gambling is related to the structural characteristics that reinforce and facilitate gambling behaviour once it has started (e.g., bet frequency, event frequency, event duration, and pay-out interval; Griffiths & Auer, 2013 ). Lopez-Gonzalez and Griffiths ( 2017 ) suggested that the ‘cash out’ feature might be utilized during a time where emotions run high and the structural characteristics of this feature might facilitate sports bettors to lose control when they are placing their bets.

Lopez-Gonzalez et al. ( 2019 ) carried out a study of 659 Spanish sports bettors and examined the association between structural characteristics of online sports betting and gambling severity. The results demonstrated that sports bettors with high problem gambling scores were more likely to use in-play betting and the ‘cash out’ feature. More recently, Parke and Parke ( 2019 ) carried out in-depth interviews with 19 online problem gamblers. The core theme to emerge was labelled the ‘online sports betting loop’, which comprised the new structural features of the online sports betting market, and included in-play sports betting, cash out, and instant depositing. They noted that online sports betting offered features that allow gamblers to almost immediately re-engage with the sports betting activity. Some of their participants found it a challenge to maintain their self-control and others admitted chasing their losses. The authors suggested that attention should be directed towards increasing enforced breaks in this type of gambling.

Other studies on in-play sports betting have supported the idea that in-play sports betting may possess a number of features that encourage individuals to bet more, and there could be an association between in-play sports betting and a risk of harm from gambling (Lopez-Gonzalez et al., 2019 ; Lopez-Gonzalez, Griffiths, & Estévez, 2020 ; Parke & Parke, 2019 ). Lopez-Gonzalez et al. ( 2020 ) reported that within a sample of 659 Spanish sports bettors, those who engaged in-play sports betting (compared to those who did not) reported significantly greater (1) problem gambling severity, (2) sport watching consumption, (3) consumption of junk food, (4) alcohol consumption when watching sport, and (5) watching sport to escape from everyday preoccupations. They concluded that in-play betting was associated with impulsivity which occurred under circumstances where there was a high level of emotional involvement (i.e., watching live sport and betting on it).

A few studies have attempted to delineate the relationship between in-play sports betting and increased harm amongst problem gamblers. Previous research has found that impulse sports bettors prefer to bet in-play rather than on overall match outcomes (Hing, Russell, Li, & Vitartas, 2018 ). However, it is believed that trait impulsivity is not a unitary construct, but encompasses four individual traits: sensation seeking, lack of planning, lack of perseverance, and negative urgency (acting impulsively in the context of strong emotions; Sharma, Markon & Clake, 2014 ). Hing et al. ( 2018 ) suggested that research into contextual factors that contribute to urges to bet impulsively would help the field gain a better understanding of problematic gambling behaviour.

Another explanation that has been provided as to why sports betting may be associated with problem gambling is that betting features within live sporting events such as in-play betting and ‘cash out’ might make sports bettors more susceptible to experiencing cognitive biases (Lopez-Gonzalez, Estévez, & Griffiths, 2017 ; Lopez-Gonzalez & Griffiths 2017 ). Furthermore, technological advancements along with narratives found within sports betting adverts that enhance control could lead to an increase in perceived skill causing bettors to place their wagers more uncontrollably (Lopez-Gonzalez et al., 2017 ).

One cognitive heuristic related to gambling behavior is the illusion of control (Langer, 1975 ). The illusion of control is the inclination for individuals to overestimate the control they have over the outcome of events. It has been suggested that the illusion of control may be heightened because sports bettors can choose the amount to stake, the number of bets, and the speed in which they place them, which may result in sports bettors overestimating their control over uncountable events (Lopez-Gonzalez et al., 2017 ). The availability heuristic refers to the placing of more weight on information that is easier to recall. Information that is easier to recall is judged to be more common (Tversky & Kahneman, 1973 ) which leads to an overestimation of the probability of similar things happening in the future. Gamblers often utilize heuristics to process information more quickly such as representativeness heuristics. The use of these mental shortcuts could lead to biased decisions and/or distorted perceptions (Griffiths, 1994 ).

D’Astous and Gaspero ( 2015 ) reported that when there is a limited timeframe for bet placement, sports bettors (n = 161) used heuristic processing. Sports bettors were more likely to use heuristic (intuitive and fast) processing, rather than analytic processing (slow and deliberate). This form of processing was found to result in a lower gambling return on investment. Furthermore, this study reported more experienced gamblers were more likely to use analytic processing and their bets were more favorable (D’Astous & Gaspero, 2015 ). The authors suggested that these heuristic and analytic processes act as mediators in the relationship between previous experience and betting performance. It has also been argued that features such as in-play betting and “cash-out” betting may result in sports bettors having a higher likelihood of experiencing cognitive biases (Lopez-Gonzalez et al., 2017 ; Lopez-Gonzalez & Griffiths, 2017 ) and as a result place less planned bets.

Although the potential impact of emerging online sport betting features has been raised as a possible concern in relation to risk of problem gambling, to date, there has been minimal research carried out on the underlying mechanisms and attitudes towards specific online sports betting features including in-play sports betting and the use of the “cash out” feature. Consequently, the present study explored the opinions towards in-play sports betting behaviours. More specifically, it explored sports bettors’ perceived motivation and opinions towards online sports betting features. The specific objectives were to explore participants’ opinions and attitudes to: (1) in-play sports betting, and (2) towards the ‘cash out’ feature use within online sports betting.

Participants

The participants in the present study (n = 19) were recruited from across the UK including Nottingham, London, Bristol, Birmingham, Derby, York, Leeds, Sheffield, Oxford and Dundee (see Table 1 ). Participants comprised no-risk gamblers (n = 4), low risk gamblers (n = 7), moderate risk gamblers (n = 7), and one problem gambler (n = 1). Their ages ranged from 21 to 32 years (mean = 25.5 years; SD = 3.25). Of these gamblers, most were male (n = 17), and white ethnicity (n = 16). Two participants identified as mixed race and one was of non-white ethnicity. The education level ranged from general certificate of secondary education (GCSE) to bachelor degree level (12 participants had a degree). Over two-thirds of participants did not identify with a religion (n = 13), 13 participants worked full-time (68.4%), four were university students (21%), one worked part-time, and one was in the military. Seven participants were married or lived with their partners (36.8%) and one participant had children (see Table 1 for an overview of demographic characteristics).

Qualitative interviews were conducted among a convenience sample of adults (n = 19) who had placed an in-play sports bet online within 6 months prior to the interview taking place. This inclusion criterion was selected to ensure that participants were able to freely discuss their experiences with in-play sports betting and it aimed to provide rich data. In order to recruit participants, members of the general public were approached outside of bookmakers in the research team’s home town, posters were put up around research team’s university premises, and via adverts on social media.

A semi-structured interview schedule was created using a number of open-ended questions which encompassed a set of key areas, previously identified within the literature. The interviews schedules included the following topic areas: (1) initial experience with sports betting, (2) experience of other gambling activities, (3) current sports betting behaviour (including in-play betting and use of the ‘cash out’ feature), (4) sports betting advertising, and (5) responsible gambling (full interview schedule available on request to the first author). Data were collected between August 2019 and December 2019. Participants were interviewed face-to-face at various locations including the research team’s university and the participant’s homes, apart from four interviews that were conducted via telephone. All interviews were recorded using a digital voice recorder, with consent from participants. Each participant was interviewed once (all of which lasted 25 min to 1 h). Demographic information in the form of a questionnaire was collected, including age, sex, occupation, marital status, highest level of education, marital status and city of residence (see Table 1 ).

Data Analyses and Theoretical Approach

Interviews were recorded and analysed using NVivo, a qualitative data analysis computer software package. Thematic analysis was used in the present study because it is a flexible approach that delivers a rich and detailed, yet complex account of data (Braun & Clarke, 2006 ; King 2004 ). Six stages of thematic analyses were implemented, as outlined by Braun and Clarke ( 2006 ). These were: (1) familiarization of the data, (2) generating initial codes, (3) searching for themes, (4) reviewing themes, (5) defining and naming themes, and (6) producing the report. The themes emerged through careful reading and re-reading of the data (Rice & Ezzy, 1999 ). A general inductive approach to thematic analysis was adopted, whereby transcripts were read, re-read and coded line-by-line in order to identify key themes relating to the research aims. The researchers then met regularly to discuss the emergence of major themes. New prompts and areas for investigation were added to the interview schedule as they emerged. Themes were refined and any differences in interpretation were discussed until agreement was reached by the authors. These categories were organized into themes and sub-themes and findings are supported by direct quotations from the interviews in the “ Results ” section. Expressions are used to indicate approximate endorsement: ‘most’ (16 or more participants), ‘many’ (10–15 participants), ‘some’ (4–9 participants), and a ‘few’ (three or fewer participants).

The Problem Gambling Severity Index (PGSI) was used to assess problem gambling behaviour. Participants were asked to self-assess their gambling behavior and gambling consequences during the past 12 months across nine items. The results categorize the participant into one of the following groups: non-problem gambler, low-risk gambler, moderate-risk gambler, or problem gambler depending upon the score. The PGSI was used because it was specifically designed for the general population and has been found to be valid in calculating the degrees of problem gambling severity in a non-clinical context (Holtgraves, 2009 ). However, the PGSI groupings must be treated with some caution as they cannot be seen to sufficiently explain broader gambling behaviors for the participants. In the present study, two participants scored in the ‘moderate risk’ PGSI group, but also described patterns of excessive sports betting that were not picked up by the PGSI. Additionally, two participants said they had had gambled infrequently in the previous 12 months and scored in the ‘moderate risk’ PGSI group, but also described themselves as problem gamblers because of a previous ‘addiction’ to sports betting.

The research team obtained ethical approval from their university research committee. The participants signed a consent form, in which they were reassured that all their responses were confidential and anonymous and they had the right to withdraw from the study at any time. Additionally, participants gave their consent to be audio recorded.

Based on the analysis, the themes that emerged were categorized under the broad concepts of (1) accessibility of betting via a smartphone, (2) in-play betting motivating factors to participate, (3) in-play vs. pre-match betting engagement, and (4) beliefs and attitudes towards the ‘cash out’ feature.

Accessibility of Betting via A Smartphone

The three sub-themes for betting via a smartphone were the: (1) transition from betting at a bookmaker’s shop, (2) ease in placing a bet, and (3) ability to place a bet anywhere.

Transition from Betting at A Bookmaker’s Shop

Many participants described how they initially began betting at a high street bookmakers’ shop, then transitioned to online gambling once it had become more popular. The factors that influenced sports bettors to gamble online included an increase in the number of online bookmakers, ‘welcome offers’ and other inducements offered by online operators, and the convenience of accessing online betting websites. Using a smartphone to place sports bets was the primary method of bet placement by everyone in the study sample. For example:

“I guess I moved over from the bookies as soon as I had a [smart]phone for the first time” (Participant 6). “We all went down on our lunch break to go and put football bets on. So I just started doing it then…in the shop on the coupon” (Participant 17) “If I had a spare couple of quid I’d go and get bet slips down the bookies. I don’t go into bookies anymore, I do it all online; on apps and on the internet and stuff” (Participant 12)

One participant commented that they preferred to place bets online because it offered a cash-out feature, whereas the high-street bookmaker did not:

“Online betting gives you the option to cash out, and you don’t really get that ability in the betting shop” (Participant 6)

Ease in Placing A Bet

Many participants commented on how easy it is to place a bet via a smartphone or tablet, compared to other methods (e.g., a laptop, high-street bookmakers). For example:

“It’s obviously very convenient to do it on your [smart]phone or your tablet rather than getting a laptop out and logging in” (Participant 5) “It’s just convenience, isn’t it? It’s in your pocket, turn it on [smartphone]. The apps are really easy to use” (Participant 1)

Other participants commented that they bet on a mobile device due because it has the advantage of saving time. For example:

“It’s easy to do – so it’s in front of you and it’s on your [smart]phone. There’s no going down to the [bookmaker] shops” (Participant 12). “You could be out and about and think ‘there’s a couple of games later, I’ll just have a quick bet on it’ and then you’re away. You don’t have to scout around for a bookies or anything like that” (Participant 18) “On an app, it literally is just the case of pressing buttons and pressing place bet…so it’s as quick as your thumb could move” (Participant 17).

As well as gambling apps, there were other apps mentioned that were accessed on smartphones that assisted participants with bet placement. These were Flash Scores (a website that allows sports bettors to see live match updates) and Odds Checker (so sports bettors can compare odds against different online bookmakers). One participant discussed how using a smartphone allowed him to compare different inducements across gambling sites:

“It’s normally on a mobile. It’s just easy isn’t it? You’ve got it to hand and you tend to get better offers online and you can see what offers they are straight away and compare them to other betting sites” (Participant 2).

Ability to Place A Bet Anywhere

As previously mentioned, the most popular method for placing bets was on a smartphone. Some participants discussed how it was now possible for them to access the gambling apps in any location, at any time, and they did not have to rely on a laptop or computer in order to place a bet. Therefore, one of the main advantages of betting on a smartphone was the flexibility of location that it allowed. For example:

“It’s really convenient being on a mobile [phone]. My phone’s in my hand the majority of the day anyway and the gambling apps are on my phone anyway” (Participant 1).

Some participants discussed how they placed bets in multiple locations using their smartphones. Frequently mentioned betting locations included at the participant’s home, the pub, at friends’ houses, and at work.

“It’s just easy to use. Use can use it when you’re at the pub, or in different environments. It’s handy for the in-play ones ” (Participant 19) “I liked having the convenience of being able to do it anywhere, anytime, not having to be at home sat in a specific place to do it. Which is probably part of the problem as well because I could literally do it anywhere. You know, I could do it in the car, out shopping, at work, anywhere like that and no-one would know what I was doing” (Participant 5)

One participant commented on how he used his smartphone to check the status of his bet:

“On a Saturday I would watch the scores live but if not I’m always checking my [smart]phone constantly. Last time I went out for a meal I had a bet running and everyone was like ‘why are you on your phone and I was like ‘the football’s on’” (Participant 6).

In-Play Betting Motivating Factors to Participate

The three main sub-themes as to reasons why participants engaged in live in-play betting were that it: (1) increases excitement, (2) makes the game more intense, and (3) allows gamblers to use their betting skill and knowledge.

Increases Excitement

Betting on a sporting event provided increased interest and excitement while watching it. Many participants commented that they took part in in-play sports betting because it increased their engagement with the game. This is because it made the game more exciting to watch because there was an opportunity for monetary gain. For example:

“It increases excitement of that game and your attention and enjoyment” (Participant 2). “It makes the game more interesting…and more exciting” (Participant 11). “It’s quite fun trying to predict what’s going to happen” (Participant 19).

One participant described online sports betting as a “buzz”. A feeling of excitement has come from placing a bet, and this is amplified if the bet is a winning bet, particularly if he is with friends and they are sharing the experience.

“ Well it’s just a bit of a buzz really. Like…if you’ve got a bet on something…the bets I tend to place are bets that like carry on going on for most of the game. Like…if I’m betting on an individual match, let’s say I’m betting on somebody to score at a particular time, you’ve got the whole game that could actually come through. So it’s the whole buzz and expectation thing. When you win it is actually a buzz, especially if you’re with your mates and they’ve got it on as well” (Participant 8)

Makes The Game More Intense

Some participants discussed how in-play betting increased feeling of intensity when watching and betting on a match simultaneously. Sports betting on a match whilst watching it allowed for the game to be more psychologically interesting. For example:

“It makes the game a more enjoyable and adds a bit of tension” (Participant 19). “With in-play, you’re more invested in it. You can place sports bets that are in the future and if you do that, I don’t know about other people, but I can place a bet over a span of a few days and then forget about it and come back to it and think ‘oh, it lost’. Whereas in-play you’re more invested in what you’re watching anyway. So it’s not very often you would place an in-play sports bet and not be watching the play happen. There’s a bit more of a thrill to it I guess” (Participant 3).

Allows Gamblers to Use Their Betting Skill and Knowledge

Some individuals engaged in in-play sports betting because they believed that they possessed skills which would influence the outcome of their bets, and thus providing them with a level of control. Many participants commented that they watched the game whilst betting on it, because this allowed them to assess the status of their match before placing a bet. For example:

“If a team scores then they get momentum and they turn it over, especially if you’re expecting them to win anyway. So I suppose watching the game and thinking ‘yeah I know what’s going to happen’” (Participant 6). “I guess in-play betting you feel more confident that you know. Everyone thinks they’re an expert. You’re watching it and you think ‘well there you go, this is actually quite accurate’…You’ve been watching the game and that would inform your decision” (Participant 15). “If I’m not watching it because I can’t see what’s going on, I won’t [bet in-play]” (Participant 7).

A few participants then discussed how a game developed and what they had observed during the match influenced them to place a bet. For example:

“It might have been Newcastle [United] at Stoke [City] and I was watching it and I got the impression that Stoke were going to score before half time and they got a penalty in injury time in the first half and they scored and it felt great” (Participant 7). “The ultimate goal is to beat the bookies, isn’t it? So I guess when you play in-play you think ‘I’ve analysed this and there’s loads of corners coming’, or God knows what. And you can make more of an educated guess. It makes you feel better about the gambling” (Participant 11).

In-Play Betting Vs. Pre-Match Engagement

Participants also made comparisons between fixed odds sports betting and in-play sports betting. Some of the participants commented on the dynamic odds being offered by online bookmakers during a game as a motivation for engaging in in-play sports betting. One of the benefits of this was possible monetary gains which were seen as an advantage. For example:

“You can easily make more money in-play betting rather than pre-match betting because you know, with pre-match the odds are set at a certain price and that’s what the bookmakers offer. But in-play the price changes and that’s what a lot of people will look for. They’ll see whether they can get value. They'll see where they can maybe make as much money as they can” (Participant 14). “Obviously the odds change as the match is going, so you can get quite lucrative winnings back depending on how much you put down” (Participant 18). “I’ll only in-play bet if I’m actually watching the football because the idea, or the one good thing about in-play betting is that you can put a bet on before the match but you realize that ten minutes in that it’s not going the way that I thought it would and the team that I expected to win are actually not playing particularly well at all and it looks like the other teams are going to win and put on another bet” (Participant 6).

Most participants discussed this in relation to football (soccer) because that was the event that they were betting on. However, one participant discussed this idea in the context of betting on tennis and darts:

“You look at something that’s going against what should be happening and you try and hit it at the point when it’s furthest away from where it should be. I suppose tennis is a good example of that. The odds change so dramatically. [In] darts, the odds change so dramatically, that when you do it prior to the game you’ll never get odds on the favourite. But during the game, you can get great odds on someone like [tennis player Novak] Djokovic to beat someone outside of the top ten [tennis players] providing that Djokovic is already a set and a break down in the second [set] or something” (Participant 12).

A few participants did view the odds changing as beneficial to their sports betting outcomes and preferred to place bets before the event started. This was because they had more time to think about the bet before they placed it. This was most popular for football (soccer) accumulators placed on Saturday fixtures. The following example is of a participant who wanted to take their time before placing the bet:

“You have little time really, so you’ve got to rush yourself a little bit. I don’t like to rush myself, I like to think about it a little bit” (Participant 10).

The timing of when the match was on was also discussed as a reason for placing a pre-match bet as opposed to an in-play sports bet. For example:

“If it’s a Saturday, I’m more likely to do a pre-match bet because there is a full set of fixtures. If it’s on a weeknight for [the European soccer] Champions League or something like that, it’s more likely to be on TV, and I would be more likely to do an in-play bet when something’s happened in the game which triggers me thinking about betting” (Participant seventeen). “But if I had 10 minutes, rather than….because bets are a bit more long term…like you can do kind of action…kinda sort of stuff…like with the in-play bets, you are like involved with it more often than not because you’re more invested in the game and whatever it is you’re sort of in-play betting with but if you start placing a bet on something like an accumulator and it’s got like a bet that’s on right now or something or starts like five hours later, you don’t really always keep track of it as much. But you can kind of place it and leave it ” (Participant 3)

The following participant discussed how the odds were something that changed rapidly:

“The odds can change pretty fast. Obviously it just depends on what’s happening. So, if you’re betting on a football games that’s in-play and one team is obviously doing a lot better than the other, you can just start to see the odds getting shorter and shorter and shorter. If you place a bet sort of, fast, if you hesitate a little bit maybe, you could end up, if it wins, obviously if you place money on that bet you’d get less money than you thought you would with the bet had you been a little bit quicker placing it. That’s just in-play. It’s just the environment with it I guess” (Participant 3).

A few participants reported that in-play sports betting allowed them to continue their betting and allowed them an attempt to recoup their losses or place multiple bets in a game. For example:

“It’s almost like an instant win depending on what you’re in-play betting on. You don’t have to wait until the end of the game to win so you could potentially have a few in-play bets on the same game” (Participant 7). “It does make it very easy or very quick to put bets on in a short space of time, and I always think that that’s kind of dangerous if you do get in that mindset” (Participant 9).

Beliefs and Attitudes Towards The ‘Cash Out’ Feature

There were three sub-themes related to the ‘cash out’ feature: (1) recouping a losing bet, (2) the ‘cash out’ monetary value being high, and (3) regret after cashing out.

Recouping A Losing Bet

The analysis showed that participants had different motivations for cashing out, including minimizing losses when the bets were losing, and acquiring more funds to allow the placement of additional bets. Bets can be withdrawn whilst a sporting event is still in play, to guarantee at least some profit and/or to minimize losses. This was dependent upon the cash out value being of what the participants perceived to be an acceptable amount. Some participants chose not to cash out their bets at all. All participants had cashed out a bet at some point in their life. The most popular sport where the cashing out of bets was during football (soccer) matches. Reasons for this included the length of the sport (i.e., being a 90-min game), and there were more likely to be surprises or changes within the game which resulted in the participant potentially cashing out their bet. One participant talked about how once the game started and they cashed out, the newly acquired funds allowed them to gain momentum and continue betting to reach an expected target that they originally had in mind. For example:

“I think it encourages quite a bit of repetition betting in a sense that you might cash out and use that money straight away [to re-bet]. So there’s a bit more of a momentum type thing. If you cash out, and say you’ve got a bet with returns of £420 and you cash out at £70 and you’ve only placed £10 down to get that, you’re still £60 up. But you want that, or you have an idea of £420 in your head at some point. So you think ‘I’ve got a bit of a bigger sum to reach here-so you’ll probably just invest your money back into the site’. I think at that point, once you’ve got something, I’m not playing with the original money that I invested with anymore” (Participant 3).

Another participant discussed the emotions that came into play when deciding whether or not to cash out their bet. They described different emotions they have experienced, with one way resulting in cashing out the bet due to “nerves” and the other letting the bet ride because they felt more confident. For example:

“I would cash out because of the nerves. I’d be sweating it, thinking ‘you know I’ve made the money, let’s not be greedy, have it over and done with’. But on the other hand I haven’t cashed out because I’m confident, I’m risking it a little bit” (Participant 16). “I’ve put on an unlikely bet and my team has scored but they’re still losing so they’ve offered me a cash out which was more than my stake and I’ve changed my mind and thought the team I bet on aren’t going to win, so it’s worth taking the extra or doubling my stake instead of ten times my stake. Or just cutting my losses essentially” (Participant 17).

The ‘Cash Out’ Monetary Value Being High

The cash out value has to be enough to be deemed worth cashing out by the sports bettor. For example, in the instance that the bet has made a profit on the initial stake. The performance of the team may influence whether individuals cash out their bet because they feel the team are not performing well and the bet may lose. For example:

“If I thought that the bet was going to lose then you want to try and recover as much of the original stake as possible” (Participant 19). “If the accumulator has got to a good amount where I’m making at least more money than I expected or if they look like…I don’t think the team is going to win or the bet isn’t going to come in then I’ll try and cash out early but usually I’ll end up just leaving [the bet] on” (Participant 11).

Other participants would only cash out if they were betting for a profit, or alternatively if they thought that the cash out value was at an amount that was worth taking. For example:

“It will always be for a profit. If I cashed out and lost money, I’d think well I might as well have let it run its course” (Participant 7). “I probably would cash out if my winnings were say £250 to £300, because that’s quite a lot of money. Say I’m on for £15/£20 I’d probably just see [the bet] through until the end” (Participant 18).

Regret After Cashing Out

A few participants expressed that they had not always made the correct cash out decision and then came to regret it after. This went both ways with participants either affecting their profits by taking a risk that turned out to be cashing out too early, rather than letting the bet ride or have the bet lose and not cash out. For example:

“I’ve won a few bets but cashed out too early so I just don’t bother anymore” (Participant 19) “There’s been a lot of times that I’ve waited on it and thought ‘no, I’m going to ride it out’ and it’s lost and I should have taken the cash out” (Participant 6). “There’s been a few times where I haven’t cashed out and I’ve regretted it because I’ve been close to winning money and I’ve been offered quite a good amount of money to cash out and I’ve not taken that option” (Participant 10). “A couple of weeks ago I placed an in-play accumulative. It was like five teams and within about 20 minutes in the second half they were all winning…I was getting offered £90 but had I let it run. I was being offered around £400 but I took it anyway. I took the £90 quid and of course 90 minutes came and if I'd let it play, it would have won, but I still see it as making money anyway. I still made £70 but I was a bit hacked off. Had I just let it run…I could have had a bit more…I'm still not been able to decide whether cash out is a good thing or not because it can be beneficial at some point and I guess that's the risk you take” (Participant 14).

Problem Gambling Behavior and In-Play Betting Features

Whilst most participants discussed the advantages of in-play betting on a mobile app, there were some aspects of mobile betting that appeared to encourage problematic gambling behavior. Online sports betting removes the social context where people who have problems with their gambling behavior might experience guilt, self-consciousness, fear of stigma, and friend or family intervention due to repeated losses and high expenditure. For example, the following participants who had experienced gambling problems discussed how they were more likely to remain in control of their in-play betting expenditure when they were in a social environment:

“There’s been times when I’m with friends, I’ll make bets that are always a bit lower but when I’m on my own I have a moment of ‘you know what, I’m going to put a big bet on this’ just because, like, no-one’s around to be like ‘don’t do that’” (Participant 1)

One participant described how sports betting was initially a social activity that then developed into a more compulsive behavior, associated with secrecy. Betting on a mobile allowed this individual to hide their gambling whilst in the same room as their partner:

“It kind of got more of a problem when I started sports betting and I would do it on my own. Or we might be in the same room on the sofa but I’d be on my phone and [my partner] wouldn’t know what I was doing. I was very secretive about it once it got past that initial gambling for fun stage” (Participant 5)

Mobile betting provides a solitary environment and appears to facilitate riskier gambling in these cases compared to in-person betting at a high street bookmaker. This was reflected in the some of the quotes:

“I liked having the convenience of being able to do it anywhere, anytime, not having to be at home sat in a specific place to do it. Which is probably part of the problem as well because I could literally do it anywhere. You know, I could do it in the car, out shopping, at work, anywhere like that and no-one would know what I was doing. It helped me to keep it secret…even if I was at the gym I’d have my phone with me and I’d be able to place bets at the gym and follow them whilst I was there. Anywhere really, anywhere that I could get a bit of privacy so no one could see what I was doing” (Participant 5) “I've been gambling for so long and you know, addiction has cropped up quite often and I kind of keep it private now. Well, as much as I can. So, like no-one really knows that I do it anymore. Well, they do and they don’t. Sometimes I just can’t hide it, especially if I had like a big loss, people know. I do it on my tablet and I do it in private because, one, I don’t want anyone complaining at me, and two, I don't want anyone getting worried. Um, three, it's a personal thing. I want to enjoy it myself. It sounds a bit morbid actually, you know, now that I'm talking about it, but yeah, I do it on my tablet and I do it alone. I never gamble with friends” (Participant 14)

In most cases, participants described online sports betting as a gambling activity involving skill, analysis, and engagement with the sporting event. However, one participant described how as their gambling behavior became increasingly problematic, they transitioned into placing bets without much thought of the outcome and without prior analysis, but instead on the type of bet that would get them the highest monetary return:

“On a roulette table or blackjack there is a house edge and will lose eventually no matter how good of a run you go on because they’re designed that way. But with sports betting, I felt like I could analyze the form and look at the game and get a feel for it and bet according to that, which is what I first started betting on. But as it got later on, I wasn’t betting based on any data, or form, or feeling. I was literally just looking at the odds and placing a bet on it. I was betting on anything…that I maybe had no idea about” (Participant 5)

There is a constant stream of sports betting opportunities available for in-play betting. One participant described how they temporarily banned themselves from gambling online after they began in-play betting on sporting events that they would not ordinarily be interested in:

“There was probably a little bit of an addictive sort of temptation, like, looking at your phone and placing bets on matches that you didn’t really care about” (Participant 19)

The purpose of the present study was to contribute to knowledge concerning online sports betting features, specifically in-play sports betting and the ‘cash out’ feature. Based on the analysis, four broad themes emerged (accessibility of betting via a smartphone, in-play betting motivating factors to participate, in-play vs. pre-match betting engagement, and beliefs and attitudes towards the ‘cash out feature) comprising nine subthemes (see Table 2 ).

To date, there has been a small amount of research carried out on in-play sports betting and the findings from the present study will be discussed in relation to these. One theme was ease of access to in-play sports betting using smartphones. Sports bettors now have immediate access to sports betting websites in most locations and situations. The findings here suggest that smartphone betting allows immediate access to gambling, supporting previous research that online gambling is easy to access via mobile devices (Deans, Thomas, Daube, & Derevensky, 2016 ). It has been previously suggested that this increased accessibility to online gambling websites and the ease of being able to use online platforms, may speed up maladaptive learned behaviours, including problem gambling (James, O'Malley, & Tunney, 2016 ; McCormack & Griffiths, 2013 ). The sports bettors within the present sample had a preference for placing bets on mobile devices, which supports previous research that sports bettors (83.4%) prefer to use a remote device to place a bet rather than going to a betting shop (Lopez-Gonzalez et al., 2019 ). Additionally, the same study found that problem gamblers were more likely to prefer to use a mobile device.

Sports bettors in the present study would often bet on the match in order to make the game appear more exciting and intense. Previous research has suggested that one way in which the structural characteristics of in-play sports betting that may contribute to problem gambling is that they make the event more interesting and/or exciting (Parke & Griffiths, 2007 ). Gambling games that involve speed and excitement have been previously associated with problem gambling (Parke & Griffiths, 2007 ). Whilst the findings here concur with previous research, it may be of value to investigate to clarify which features of in-play sports betting add to the excitement and if these are more specifically related to problem gambling.

One area that was prominent within the interviews was participants’ awareness of the odds that were being offered during in-play betting by the bookmakers. Lamont, Hing and Vitartas ( 2016 ) reported that live odds updates during sports events may prompt bettors into placing impulse bets. These impulse bets were more likely to be placed if the odds were perceived as good and related to their favourite team. Some participants believed that it was easier to make money in in-play sports betting as opposed to betting before a match. Some gamblers perceive sports betting as a skill-based form of gambling (Cantinotti, Ladouceur & Jacques, 2004 ). Previous research on motivations to engage in sports betting in Tasmania (Australia) was related to the sports bettor’s perceived amount of knowledge or experience with the sports that they were betting on (Palmer, 2014 ). In a study of 258 individuals, Khazaal et al. ( 2012 ) reported that experts (i.e., professional soccer players, coaches, or journalists) were no more successful at predicting soccer match outcomes than the non-professionals. In systematic review carried out on the role of chance and skill in sports bettors (and focusing on cognitions in the behaviour), Mercier et al. ( 2018 ) reported that sports bettors overestimated the importance of skill on the overall match outcome. Ladouceur, Giroux and Jacques ( 1998 ) found that gamblers on horse races who were classed as ‘experts’ picked more winning selections but did not have better monetary outcomes than random selection. It was concluded that the experts were more thoughtful, careful, and likely to place safer bets.

It is possible that some in-play sports gamblers may experience higher levels of perceived skill in the activity due to cognitive distortions. It is possible that cognitive distortions could lead to the development and maintenance of a gambling disorder. Cantinotti, Ladouceur and Jacques ( 2004 ) examined whether the idea of having betting skill was illusion or chance. Compared to bets made by expert sports bettors versus randomly selected wagers, they found that sports bettors demonstrated a higher accuracy for correctly predicting game outcomes compared to chance (i.e. randomly selected bets), although, the overall amount of money won was not higher than chance. The researchers concluded that the notion of skill when betting is the result of cognitive distortions.

Theoretical papers that have focused on the structural characteristics of in-play sports bettors have specifically noted that the nature of the gambling activities has changed from what was previously a discontinuous (low-risk) form of gambling to a continuous (high-risk) form of gambling (Griffiths & Auer, 2013 ; Lopez-Gonzalez & Griffiths, 2017 ). In the present study, participants noted that getting a bet credited in in-play sports betting felt like “an instant win” and that multiple bets could be placed within a small window. Therefore, the shortening of bets being paid out has reduced delays in receiving rewards from gambling, and allowing the gambler the potential of placing multiple bets per match.

The present study found that reasons for using the ‘cash out’ feature varied between individuals. Some individuals cashed out to cut their losses, whilst others cashed out when they were betting a profit, while other preferred not to cash out and let the game run to completion. Lopez-Gonzalez et al. ( 2019 ) found that problem gamblers were more likely to use the ‘cash out’ feature than non-problem gamblers. Further research should investigate what types of individuals (in terms of demographics and personality traits) use the ‘cash out’ feature and their motivations for doing so. Comparisons have previously been made between the ‘cash out’ feature and stock market trading (Lopez-Gonzalez & Griffiths, 2017 ). For example, cashing out is similar to a stop-loss order within financial trading, which is an order to sell an existing shareholding which is triggered if the bid price falls to, or below the stop price set by a trader. This might be used when somebody buys a share to give them some protection and help minimize loss should a share price fall. With the ‘cash out’ feature, individuals can decide whether they are going to cash out when it receives a specific level of profit or cash out when the bet is losing a specific amount (or alternatively let the bet run until the end). Lopez-Gonzalez and Griffiths ( 2017 ) claimed that in-play online sports betting may benefit from regulations that are currently applied within the stock market industry.

Limitations

The present study has a number of limitations to take into account when interpreting the findings. Firstly, the sample mainly consisted of non-problem male sports bettors, despite efforts by the research team to recruit female gamblers and more problem gamblers. Future research should attempt to recruit greater numbers of females and problem gamblers in their samples. Secondly, the present study specifically targeted individuals who had placed at least one sports bet within the past six months. For this reason, participants may have had varied levels of engagement with sports betting and although they were assumed to qualify and meet the aims of the study, they were not representative of all online sports bettors or the wider betting population. Thirdly, the study relied on self-report data which can be affected by a number of well-known biases (such as social desirability and recall biases). Finally, it is important to highlight that the study was exploratory which allows for a preliminary understanding of in-play sports betting behavior, rather than allowing for definitive conclusions, especially because of the sample size.

Overall the sports bettors in the present study viewed in-play sports betting favorably and readily accessible. However, the results demonstrated that this is a way of gambling that can be played without interruption and which may lead to repetitive (i.e., continuous) gambling and/or unwarranted feelings of control. Given that this was an exploratory study, further research is required in order to draw more definitive conclusions. Future research could focus on the following areas: (1) qualitative and quantitative studies examining the motivation and perceptions of in-play sports betting use with females and/or samples of vulnerable individuals; (2) empirical studies on how factors such as the marketing and advertising of sports betting products influence sports betting behavior; and (3) longitudinal studies to track the game-play of in-play sports bettors. Further research into this area is required in order to provide direction for policymakers to develop responsible gambling measures for this relatively new way of gambling.

Barber, B. (2018). Bet365 cash in as sports betting and in-play boost profits by 33 per cent. Racing Post, November 20. Retrieved April 6, 2020, from: https://www.racingpost.com/news/boost-for-bet365-as-profits-soar-33-per-cent-to-682-million/354551 .

BBC Sport (2018). Premier League TV rights: Amazon to show 20 matches a season from 2019–2022, June 7. Retrieved April 6, 2020, from: https://www.bbc.co.uk/sport/football/44396151 .

Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology , 3 (2), 77–101. https://doi.org/10.1191/1478088706qp063oa .

Article   Google Scholar  

Cantinotti, M., Ladouceur, R., & Jacques, C. (2004). Sports betting: Can gamblers beat randomness? Psychology of Addictive Behaviors, 18 (2), 143–147. https://doi.org/10.1037/0893-164x.18.2.143 .

Article   PubMed   Google Scholar  

D’Astous, A., & Gaspero, M. D. (2015). Heuristic and analytic processing in online sports betting. Journal of Gambling Studies, 31 (2), 455–470. https://doi.org/10.1007/s10899-013-9438-x .

Deans, E. G., Thomas, S. L., Daube, M., & Derevensky, J. (2016). “I can sit on the beach and punt through my mobile phone”: The influence of physical and online environments on the gambling risk behaviours of young men. Social Science and Medicine, 166 , 110–119. https://doi.org/10.1016/j.socscimed.2016.08.017 .

Gambling Commission. (2016). In-play (in-running) betting: Position paper. Retrieved March 10, 2020, from: https://www.gamblingcommission.gov.uk/pdf/In-running-betting-position-paper.pdf

Gambling Commission. (2018a). Gambling participation in 2017: Behaviour, awareness and attitudes. Annual report. Birmingham: Gambling Commission. Retrieved March 10, 2020, from: https://www.gamblingcommission.gov.uk/pdf/survey-data/gambling-participation-in-2017-behaviour-awareness-and-attitudes.pdf

Gambling Commission. (2018b). Review of online gambling. Birmingham: Gambling Commission. Retrieved March 10, 2020, from: https://www.gamblingcommission.gov.uk/PDF/Online-review-March-2018.pdf

Gambling Commission. (2018c). Strategy 2018–2021. Birmingham: Gambling Commission. Retrieved March 10, 2020, from: https://www.gamblingcommission.gov.uk/PDF/Strategy-2018-2021.pdf

Griffiths, M. D. (1994). The role of cognitive bias and skill in fruit machine gambling. British Journal of Psychology, 85 (3), 351–369. https://doi.org/10.1111/j.2044-8295.1994.tb02529.x .

Griffiths, M. D. (2012). The 'In' crowd: Is there a relationship between 'in-play' betting and problem gambling? June 20. Retrieved March 10, 2020, from: https://drmarkgriffiths.wordpress.com/2012/04/29/the-in-crowd-is-there-a-relationship-between-in-play-betting-and-problem-gambling/

Griffiths, M. D., & Auer, M. (2013). The irrelevancy of game-type in the acquisition, development and maintenance of problem gambling. Frontiers in Psychology , 3 , 621. https://doi.org/10.3389/fpsyg.2012.00621 .

Article   PubMed   PubMed Central   Google Scholar  

Hing, N., Russell, A. M., Li, E., & Vitartas, P. (2018). Does the uptake of wagering inducements predict impulse betting on sport? Journal of Behavioral Addictions, 7 (1), 146–157. https://doi.org/10.1556/2006.7.2018.17 .

Hing, N., Russell, A. M., Vitartas, P., & Lamont, M. (2015). Demographic, behavioural and normative risk factors for gambling problems amongst sports bettors. Journal of Gambling Studies, 32 (2), 625–641. https://doi.org/10.1007/s10899-015-9571-9 .

Article   PubMed Central   Google Scholar  

Holtgraves, T. (2009). Evaluating the Problem Gambling Severity Index. Journal of Gambling Studies, 25 (1), 105–120. https://doi.org/10.1007/s10899-008-9107-7 .

James, R., O'Malley, C., & Tunney, R. (2016). Understanding the psychology of mobile gambling: A behavioural synthesis. British Journal of Psychology, 108 (3), 608–625. https://doi.org/10.1111/bjop.12226 .

Khazaal, Y., Chatton, A., Billieux, J., Bizzini, L., Monney, G., et al. (2012). Effects of expertise on football betting. Substance Abuse Treatment, Prevention, and Policy, 7 (1), 18. https://doi.org/10.1186/1747-597x-7-18 .

Killick, E. A., & Griffiths, M. D. (2019). In-play sports betting: A scoping study. International Journal of Mental Health and Addiction, 17 , 1456–1495. https://doi.org/10.1007/s11469-018-9896-6 .

King, N. (2004). Using templates in the thematic analysis of text. Essential guide to qualitative methods in organizational research (pp. 256–270). London: Sage. https://doi.org/10.4135/9781446280119.n21 .

Chapter   Google Scholar  

LaBrie, R. A., Laplante, D. A., Nelson, S. E., Schumann, A., & Shaffer, H. J. (2007). Assessing the playing field: A prospective longitudinal study of internet sports gambling behavior. Journal of Gambling Studies, 23 (3), 347–362. https://doi.org/10.1007/s10899-007-9067-3 .

Ladouceur, R., Giroux, I., & Jacques, C. (1998). Winning on the horses: How much strategy and knowledge are needed? Journal of Psychology: Interdisciplinary and Applied, 132 (2), 133–142. https://doi.org/10.1080/00223989809599154 .

Lamont, M., Hing, N., & Vitartas, P. (2016). Affective response to gambling promotions during televised sport: A qualitative analysis. Sport Management Review, 19 (3), 319–331. https://doi.org/10.1016/j.smr.2015.06.002 .

Langer, E. J. (1975). The illusion of control. Journal of Personality and Social Psychology, 31 (2), 311–328. https://doi.org/10.1037/0022-3514.32.2.311 .

LaPlante, D. A., Schumann, A., LaBrie, R. A., & Shaffer, H. J. (2008). Population trends in internet sports gambling. Computers in Human Behavior, 24 , 2399–2414. https://doi.org/10.1016/j.chb.2008.02.015 .

Lee, P., & Paul. (2018). Discover the latest trends via our mobile consumer survey. October 5. Retrieved March 10, 2020, from: https://www2.deloitte.com/uk/en/pages/technology-media-and-telecommunications/articles/mobile-consumer-survey.html

Lopez-Gonzalez, H., Estévez, A., & Griffiths, M. D. (2017). Controlling the illusion of control: A grounded theory of sports betting advertising in the UK. International Gambling Studies, 18 (1), 39–55. https://doi.org/10.1080/14459795.2017.1377747 .

Lopez-Gonzalez, H., Estévez, A., & Griffiths, M. D. (2019). Internet-based structural characteristics of sports betting and problem gambling severity: Is there a relationship? International Journal of Mental Health and Addiction, 17 , 1360–1373. https://doi.org/10.1007/s11469-018-9876-x .

Lopez-Gonzalez, H., & Griffiths, M. D. (2017). “Cashing out” in sports betting: Implications for problem gambling and regulation. Gaming Law Review, 21 (4), 323–326. https://doi.org/10.1089/glr2.2017.2144 .

Lopez-Gonzalez, H., Griffiths, M. D., & Estévez, A. (2020). In-play betting, sport broadcasts, and gambling severity: A survey study of Spanish sports bettors on the risks of betting on sport while watching it. Communication and Sport, 8 , 50–71. https://doi.org/10.1177/2167479518816338 .

McCormack, A., & Griffiths, M. D. (2013). A scoping study of the structural and situational characteristics of internet gambling. International Journal of Cyber Behavior, Psychology and Learning, 3 (1), 29–49. https://doi.org/10.4018/ijcbpl.2013010104 .

Mercier, J., Sévigny, S., Jacques, C., Goulet, A., Cantinotti, M., & Giroux, I. (2018). Sports bettors: A systematic review. Journal of Gambling Issues . https://doi.org/10.4309//jgi.2018.38.11 .

Palmer, C. (2014). Sports betting research literature review. Retrieved March 10, 2020, from: https://www.dhhs.tas.gov.au/__data/assets/pdf_file/0006/170772/Sports_Betting_Literature_Review.pdf .

Parke, J., & Griffiths, M. D. (2007). The role of structural characteristics in gambling. In G. Smith, D. Hodgins, & R. Williams (Eds.), Research and measurement issues in gambling studies (pp. 211–243). New York: Elsevier.

Google Scholar  

Parke, A., & Parke, J. (2019). Transformation of sports betting into a rapid and continuous gambling activity: A grounded theoretical investigation of problem sports betting in online settings. International Journal of Mental Health and Addiction, 17 , 1340–1359. https://doi.org/10.1007/s11469-018-0049-8 .

Rice, P., & Ezzy, D. (1999). Qualitative research methods: A health focus . Melbourne: Oxford University Press.

Sharma, L., Markon, K. E., & Clark, L. A. (2014). Toward a theory of distinct types of “impulsive” behaviors: A meta-analysis of self-report and behavioral measures. Psychological Bulletin, 140 (2), 374–408. https://doi.org/10.1037/a0034418 .

Tversky, A., & Kahneman, D. (1973). Availability: A heuristic for judging frequency and probability. Cognitive Psychology, 5 , 207–232. https://doi.org/10.1037/e301722005-001 .

Download references

No financial support was received for this study.

Author information

Authors and affiliations.

International Gaming Research Unit, Psychology Department, Nottingham Trent University, Nottingham, UK

Elizabeth A. Killick & Mark D. Griffiths

You can also search for this author in PubMed   Google Scholar

Contributions

EAK and MDG wrote the paper.

Corresponding author

Correspondence to Elizabeth A. Killick .

Ethics declarations

Conflict of interest.

The first author declares that she has no conflict of interest. The second author has received funding for a number of research projects in the area of gambling education for young people, social responsibility in gambling and gambling treatment from Gamble Aware (formerly the Responsibility in Gambling Trust), a charitable body which funds its research program based on donations from the gambling industry. The second author also undertakes consultancy for various gaming companies in the area of social responsibility in gambling.

Ethical Approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Killick, E.A., Griffiths, M.D. Why Do Individuals Engage in In-Play Sports Betting? A Qualitative Interview Study. J Gambl Stud 37 , 221–240 (2021). https://doi.org/10.1007/s10899-020-09968-9

Download citation

Published : 13 August 2020

Issue Date : March 2021

DOI : https://doi.org/10.1007/s10899-020-09968-9

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Sports betting
• In-play betting
• Thematic analysis
• Live action betting
• Find a journal
• Publish with us
• Track your research

Numbers, Facts and Trends Shaping Your World

Read our research on:

Full Topic List

Regions & Countries

• Publications
• Our Methods
• Short Reads
• Tools & Resources

Read Our Research On:

As more states legalize the practice, 19% of U.S. adults say they have bet money on sports in the past year

Around one-in-five U.S. adults (19%) say they have personally bet money on sports in some way in the last 12 months, whether with friends or family, in person at a casino or other gambling venue, or online with a betting app, according to a new Pew Research Center survey.

The survey comes more than four years after the Supreme Court effectively legalized commercial sports betting in the United States . As of this month, 35 states and the District of Columbia have authorized the practice in some form, with Massachusetts becoming the latest state to do so in August.

Despite the growth of commercial sports betting in the wake of the Supreme Court’s 2018 ruling, the most common way that Americans bet on sports is with friends or family, according to the Center’s survey, which was fielded July 5-17 among 6,034 adults. Some 15% of adults say they have bet money on sports with friends or family in the last 12 months, such as in a private betting pool, fantasy league or casual bet. Smaller shares say they have bet money on sports in person at a casino, racetrack or betting kiosk in the past year (8%) or that they have done so online with a betting app, sportsbook or casino (6%). All told, 19% of adults have bet money on sports in at least one of these ways in the past year.

The U.S. Supreme Court decided in 2018 to effectively legalize commercial sports betting in the United States . Pew Research Center conducted this analysis about four years later to find out what share of U.S. adults have personally bet on sports in the past year and to understand Americans’ broader views about the impact of sports betting on society and sports.

The analysis is based on a survey conducted July 5-17, 2022, among 6,034 U.S. adults. The poll included 4,996 respondents from the Center’s American Trends Panel (ATP), an online survey panel that is recruited through national, random sampling of residential addresses. This way nearly all U.S. adults have a chance of selection. It also included an oversample of 1,038 respondents from Ipsos’ KnowledgePanel. The survey is weighted to be representative of the U.S. adult population by gender, race, ethnicity, partisan affiliation, education and other categories. Read more about the ATP’s methodology . Here are the questions used for this analysis, along with responses, and its methodology .

This survey includes a total sample size of 234 Asian adults. The sample primarily includes English-speaking Asian Americans and, therefore, may not be representative of the overall Asian adult population. Despite this limitation, it is important to report the views of Asian Americans on the topics in this study. As always, Asian adults’ responses are incorporated into the general population figures throughout this analysis.

Men are more likely than women (24% vs. 15%) to say they have bet on sports in some form in the past year, as are adults under the age of 50 when compared with those 50 and older (22% vs. 17%). There are also differences by race and ethnicity: Black (27%) and Hispanic adults (24%) are more likely than White (18%) and Asian American adults (10%) to report doing so.

There are no significant differences in self-reported sports betting by educational attainment or household income level. For example, 18% of college graduates say they have bet on sports in some way in the past year, as have 20% of non-college graduates. Similar shares of adults in upper-income (22%), middle-income (19%) and lower-income households (19%) say they have bet on sports in the past year.

There is also no significant difference by party affiliation: 21% of Democrats and Democratic-leaning independents say they have bet on sports in some way in the last 12 months, as have 19% of Republicans and Republican-leaning independents.

Public awareness of legal sports betting

Overall, 56% of adults say they have read or heard a lot (12%) or a little (44%) about the fact that sports betting is now legal in much of the country, while 44% say they have read or heard nothing at all about it.

Awareness of legalized sports betting varies by demographic group. Men are far more likely than women to say they have read or heard at least a little about it (69% vs. 44%). Americans ages 50 and older, those with a college degree and those in upper-income households are also among the groups who are more likely to have read or heard about it.

Perhaps not surprisingly, Americans who have read or heard a lot about the widespread legalization of sports betting in the U.S. are far more likely than other Americans to say they personally have bet on sports in the past year. Among this group of Americans, nearly half (46%) report betting on sports in some way in the past year, compared with 23% of those who have read or heard only a little and just 9% of those who have read or heard nothing at all about the fact that sports betting is now legal in much of the country.

Perceptions about legal sports betting’s effect on society, sports

The widespread legalization of sports betting has created a new revenue stream for many state governments , but it has also raised concerns about gambling addiction and other societal harms.

So how do Americans feel about the fact that sports betting is now legal in much of the U.S.? Regardless of how much they have heard or read about it, a majority of adults (57%) say it is neither a good nor bad thing for society, while about a third (34%) say it is a bad thing. Only 8% say it is a good thing for society.

The public is slightly more divided on a separate question about whether the widespread legalization of sports betting is a good or bad thing for sports . Around half of Americans (49%) say the fact that sports betting is now legal in much of the country is neither a good nor bad thing for sports, while a third say it is a bad thing for sports and 16% say it is a good thing.

On these questions, too, there are some demographic differences. Americans 50 and older are more likely than adults under 50 to say that legalized sports betting is a bad thing for both society (41% vs. 27%) and sports (38% vs. 29%). College graduates and those in upper-income households are also more likely to see the widespread legalization of sports betting as a bad thing for both society and sports.

While there are no significant partisan differences on most of the Center’s questions related to sports betting, one such divide does appear on the question of whether legalized sports betting is a good or bad thing for society. Republicans are slightly more likely than Democrats to say this is a bad thing for society (38% vs. 31%). Still, more than half in both groups (54% of Republicans and 59% of Democrats) see it as neither a good nor bad thing for society.

Note: Here are the questions used for this analysis, along with responses, and its methodology .

• Personal Finances

John Gramlich is an associate director at Pew Research Center .

Most Americans Favor Legalizing Marijuana for Medical, Recreational Use

5 facts about americans and sports, by a wide margin, americans say football – not baseball – is ‘america’s sport’, among black adults, those with higher incomes are most likely to say they are happy, about 1 in 10 restaurants in the u.s. serve mexican food, most popular.

1615 L St. NW, Suite 800 Washington, DC 20036 USA (+1) 202-419-4300 | Main (+1) 202-857-8562 | Fax (+1) 202-419-4372 |  Media Inquiries

Research Topics

• Email Newsletters

ABOUT PEW RESEARCH CENTER  Pew Research Center is a nonpartisan fact tank that informs the public about the issues, attitudes and trends shaping the world. It conducts public opinion polling, demographic research, media content analysis and other empirical social science research. Pew Research Center does not take policy positions. It is a subsidiary of  The Pew Charitable Trusts .

Copyright 2024 Pew Research Center

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List

The Gambling Behaviour and Attitudes to Sports Betting of Sports Fans

1 Social and Global Studies Centre, RMIT University, 360 Swanston Street, Melbourne, VIC 3000 Australia

Buly A. Cardak

2 School of Business, La Trobe University, Melbourne, VIC Australia

Matthew Nicholson

3 Monash University Malaysia, Kuala Lumpur, Malaysia

4 Centre for Sport and Social Impact, La Trobe University, Melbourne, VIC Australia

Alex Donaldson

Paul o’halloran, erica randle, kiera staley, associated data.

Survey responses from a sample of nearly 15,000 Australian sports fans were used to study the determinants of: (i) gambling behaviour, including if a person does gamble and the type of gambling engaged with; (ii) the number of sports and non-sports bets made over a 12-month period; and (iii) attitudes towards betting on sports. The probability of betting on sports decreased with increasing age and was lower for women and people with a university education. This gender difference varied with age, with the greatest difference found among the young. Similar effects were observed for the number of sports bets made, which declined with age. The gender difference in the number of sports bets also varied with age with the greatest difference found among the young arising from the high propensity of young men to bet on sports. Attitudes to sports betting were also analysed, with a key finding that, within friendship circles, the views that sports betting is perceived as harmless, common and very much a part of enjoying sports were stronger among young men. These permissive attitudes were stronger among people who bet on sports and those who bet on sports more frequently. The analysis of sports fans provides insights into the characteristics of the target market most likely to bet on sports, which can be used to inform public health initiatives and harm reduction campaigns.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10899-021-10101-7.

Introduction

Harm from gambling is a significant global public health issue, with negative impacts on the health and wellbeing of individuals, families and communities (Gainsbury et al., 2014 ). Researchers have argued the harm to health and wellbeing caused by gambling is equivalent to that associated with major depressive disorders, and substance misuse and dependence (Browne et al., 2016 ). There is an array of research linking harmful gambling to health and social issues, including an individual’s health and wellbeing (Rockloff et al., 2020 ; Suomi et al., 2014 ), impacts on families and relationships (Dowling, 2014 ), and an association with intimate partner and family violence (Dowling et al., 2019 ). Generally, harms related to gambling reflect social and health inequalities, with negative effects unequally skewed towards economically and socially disadvantaged groups (Cowlishaw et al., 2016 ; Raybould et al., 2021 ; Wardle et al., 2018 ). Further, Deans et al., ( 2017a , 2017b ) argued that older adults, young men, and children are most vulnerable to harm from gambling.

In this paper, we explore the gambling choices of a diverse group of sports fans from Victoria, Australia, aged 18 and over, based on data from a survey of almost 15,000 members and fans of elite sporting clubs. In doing so, we investigate the relationship between individual demographic characteristics, the gambling behaviour of these sports fans and differences in attitudes to sports betting. Australia is recognised globally as having one of the most accessible and liberalised gambling environments, with policy and regulation, online platforms and the diversification of gambling products all increasing the availability and uptake of different gambling opportunities (Deans et al., 2016a ; Hing et al., 2017 ; Pitt et al., 2017a ). However, this trend is reflected elsewhere, with similar issues reported in the United Kingdom (McGee, 2020 ), Spain (Lopez-Gonzalez, et al., 2020 ) and Ireland (Fulton, 2017 ), implying our analysis is of international importance for those seeking to understand gambling choices and attitudes, and mitigate harm through appropriate policies and programs.

Recent evidence from the Household, Income and Labour Dynamics in Australia survey (a nationally representative longitudinal survey) demonstrated that there were 6.8 million regular gamblers in 2015, of whom an estimated 1.1 million were at risk of harm from gambling-related problems (Armstrong & Carroll, 2017 ). The National Australian Gambling Statistics Report highlighted that total gambling losses rose 5% between 2017 and 2018 to $24.89 billion. These statistics demonstrate gambling is an ongoing and increasing threat to individual and public health. Not only are individuals at risk of harm from gambling, for one person with problematic behaviour, an estimated five to ten people are adversely affected (Productivity Commission, 1999 ), implying widespread economic and social costs of gambling (Wardle et al., 2018 ).

Rise and Normalisation of Sports Betting

This research focuses on sports betting, a rapidly emerging sector of the gambling industry. Its impact on normalising gambling, especially among the young, has been of increasing concern over the last decade in countries like Australia and the United Kingdom (Purves et al., 2020 ). Sports betting is one of the few forms of gambling that has shown a substantial increase in participation in recent years (Hare, 2015 ). In Australia, sports betting resulted in the largest year-on-year percentage increase (16.3%) in gambling losses during 2017–2018. The relationship between sports and gambling is increasingly symbiotic, with teams from Australia’s two major professional sports, the Australian Football League (AFL) and National Rugby League (NRL), significantly involved in the ownership and promotion of gambling products and services. Activities include formal sports partnerships, uniform naming rights, stadium signage and the promotion of odds during televised broadcasts. This general trend has been termed the ‘gamblification’ of sports by McGee ( 2020 ) and has become ubiquitous across a variety of sports settings, from elite to community level.

As a consequence of the pervasiveness of sports betting, researchers have increasingly sought to identify and describe the ‘normalisation’ effect of sports betting and its acceptance as part of peer-based socialisation and general sports fandom (Bunn et al., 2019 ; Raymen and Smith, 2017 ). A growing body of evidence has started to address the factors that lead to sports betting being perceived as an everyday part of sports, fostering its uptake. This is aligned with an increasing focus within broader gambling research on the influence of the environment and social determinants on people’s behaviour, as opposed to concentrating on a problem or pathology within the individual (Johnston and Regan, 2020 ).

There has been a strong research focus on the rise and prominence of sports betting marketing, quantifying how prevalent gambling promotions are during sports broadcasting (Milner et al., 2013 ), on social media platforms (Thomas et al., 2018 ), in live events within stadia (Thomas et al., 2012 ), and exploring how online platforms have been harnessed by wagering companies to encourage consumption (Deans et al., 2016a ). Thomas et al. ( 2012 ) highlighted there were very few visible or audible messages to counter overwhelmingly positive messages about sports betting during matches. Their research also addressed how sports betting advertising and associated strategies affect the attitudes of specific community sub-groups, including young people, parents, and young males. Pitt et al. ( 2016a ) found children could recall sports betting brand names, places they had seen betting advertising and associated plot details of advertisements. Deans et al. ( 2017b ) conducted similar work with young men and demonstrated sports betting marketing influences their betting behaviour.

Research has also focused on people’s attitudes to sports betting advertising, to improve understanding of community sentiment. Generally, this has shown that both parents and young people disagree with the increase in sports betting advertising and have concerns about how these messages promote a seemingly natural affinity between gambling and sports (Nyemcsok et al., 2021 ; Pitt et al., 2016b ). However, Pitt et al. ( 2016b ) reported that young people’s discourses about sports increasingly involve discussions about gambling ‘odds’ and that some young people believe that gambling is a usual and valued consumption activity during sports. Alternative evidence suggests young men feel particularly overwhelmed and bombarded by sports betting advertising (Thomas et al., 2012 ). This is unsurprising because this group is the target market for most Australian wagering operators. Hing et al. ( 2016 ) argued that such operators deliberately position sports betting as an activity engaged in by single, professional, upwardly mobile young men.

Other environmental or ‘normalisation’ issues investigated in the research literature include the availability and convenience of sports betting on mobile phone apps or online, with ease of access facilitating gambling (McGee, 2020 ), the socio-cultural alignment between sports betting and sports (Deans et al., 2017a ; Thomas, 2014 ), and how physical and online environmental factors influence the gambling risk behaviours of young men (Deans et al., 2016b ). For example, Deans et al. ( 2017a ) conducted semi-structured interviews with a convenience sample of 50 Australian men, aged 20–37, who were fans of and had bet on either NRL or AFL matches (games). These young male sports bettors reported their betting was normal and socially accepted, especially among sports fans, and ‘gambling-related language had become embedded in peer discussions about sport’ (p. 112). As such, Deans et al. concluded an exaggerated normalisation of wagering might exist in male sports fans’ peer groups. The young men in their study had established rituals (e.g. punters’ clubs) that reinforced their social connection to sports betting, but also enhanced the peer pressure to bet—an outcome that is perhaps inevitable given the role of social interaction in normalising behaviour (Russell et al., 2018 ).

Understanding ‘Sports Bettors’

A separate but connected branch of research literature has concentrated on profiling groups most at risk of experiencing harm from sports betting, particularly describing their attitudes and characteristics. As indicated previously, two groups of major concern are men and youth in general (both male and increasingly female). Studies and reviews have consistently found that young adult males are at greater risk of problem gambling (Hing et al., 2015 ; Williams et al., 2012 ). Recently, such research has also explored sports betting specifically. For example, Hing et al. ( 2016 ) in a quantitative study with a purposive sample of 639 Australian adults, identified key demographic risk factors for problem sports bettors included being male, younger, never married, and living either alone, in a one-parent family with children, or in a group household. Other risk factors included having a higher level of education and working or studying full-time. Numerous, frequent, and larger bets appeared to characterise high-risk sports bettors, as opposed to those deemed at low risk of experiencing harm. This is supported by recent research by Ayandele, Popoola and Obos ( 2019 ), who surveyed 749 Nigerian tertiary students aged 16–30 years to explore how socio-demographic factors, peer-based gambling and sports betting knowledge interact to shape young adults’ attitudes to sports betting. They found a favourable attitude towards sports betting was associated with being older and male, having a knowledge of sports betting and was positively related to the betting attitudes and behaviours of friends.

Whilst understanding normalisation factors and processes is important, alongside the attitudes and characteristics of different sub-groups, what is missing from current research is a large-scale examination of the attitudes of sports fans specifically, and the key demographic factors that are associated with their sports betting behaviour. This study addresses the gap. To the best of our knowledge, this research, conducted with almost 15,000 unique respondents, is the largest quantitative study operating at the nexus of sports fans and gambling behaviour. Arguably, the issues described above are more impactful in this cohort because they are highly engaged with sports and very exposed to marketing and the gambling economy. It is imperative to understand how such environmental and socio-cultural processes influence sports fans’ betting behaviour and to identify the sub-groups where these behaviours are most apparent.

Using quantitative research with a broad demographic group of sports fans (aged 18 and over), we aimed to compare attitudes between sports bettor, non-sports bettor and non-bettor cohorts, and examine the factors that would make it more likely for a sports fan to be a sports bettor. Broadly, we focused on attitudes to betting in sports, the risks associated with sports betting and perceptions about how much of a social norm this activity is. The research was guided by the following questions:

• What demographic factors make it more or less likely for a sports fan to bet on sports? How does this correlate with the number of sports bets a person makes?
• What is the impact of described normalisation processes on the attitudes of sports fans that bet on sports, compared with those that do not bet at all, and those that bet but not on sports?

Understanding the demographic profiles and risk factors for sports bettors and their attitudes is an increasingly important area of research. The results from this research could inform public health interventions and policy to help ensure they appropriately address areas of concern.

The research questions above are addressed as part of a broader project undertaken in partnership with the Victorian Responsible Gambling Foundation (VRGF). The project involved a survey that was distributed in collaboration with 17 professional sporting clubs from Australian Football, Basketball, Cricket, Soccer, Netball and Rugby Union in the state of Victoria, Australia. The survey was targeted at members, fans, and supporters of these clubs. In each instance, the survey was shared via social media channels (including Facebook and Twitter) and electronic direct marketing using email to each club’s membership base. Depending on the sporting code and relevant season (summer or winter), the survey was shared either between 30th October–3rd December 2020 or 25th February–18th March 2021. Data were collected in the context of the COVID-19 pandemic and research has demonstrated an increase in sports betting and a decrease in other types of gambling (e.g., casino, horse racing, pokies, etc.) during this period (Jenkinson et al., 2020 ). Whilst this could have impacted the behaviour and attitudes of respondents in this survey, the results still highlight the groups most at risk from engaging in sports betting and their associated attitudes. The survey took on average 15 min to complete and elicited a total of 17,228 responses. However, due to incomplete survey responses, the estimating sample is restricted to at most 14,950 observations in the analysis below. Three components of the survey were used in this study. First was data on gambling behaviour comprising responses about (i) whether an individual gambles or not and whether any gambling is sports betting, non-sports betting or both; and (ii) the number of bets in a given period.

Participants were asked about gambling activity in general first. They were advised: “Gambling includes activities in venues such as casino table games, pokies, TAB, Keno etc ., as well as raffles, lotteries and scratchies. It also includes gambling online or via apps such as sports betting, race betting and online pokies and casino games, where you bet with money.”

This was followed with the questions:

• Thinking of all these types of gambling, in the past 12 months, have you spent any money on these gambling activities?
• In the past 12 months, how often have you gambled? (with a number of times per week, month or year options available)

Participants were then asked about sports betting activity. They were advised: “ Sports betting refers to legal wagering with bookmakers on approved types of local, national or international sporting activities, (other than horse or greyhound racing both on or off the course) in person, via the telephone or via an app or online.”

• Thinking of all these types of sports betting, in the past 12 months, have you spent any money on these sports betting activities?
• In the past 12 months, how often have you taken part in sports betting? (with a number of times per week, month or year options available)

This data are summarised in the first six rows of Table ​ Table1 1 under the sub-headings ‘betting category’ and ‘number of bets’. For clarity, when referring to the specific participant groups involved in the research, we will use the terms sports bettors, non-sports bettors, or non-bettors to avoid referring to gambling and betting interchangeably. Table ​ Table1 1 shows that about 35% of the sample are non-bettors while another 35% are non-sports bettors (i.e. they bet on lotteries, raffles, poker or slot machines and casino gambling). The remaining 30% of the sample are sports bettors divided evenly between those that engage in sports betting only and those that engage in both sports and non-sports betting. The average number of bets in a year for the full sample is 19.6 for non-sports and 14.5 for sports bettors. 1 This data are also presented by betting category, showing that sports bettors, on average bet many more times per year than non-sports bettors. However, those that bet on both have a much higher betting frequency again, betting nearly twice as often as those who bet only on sports. The much higher standard deviation among sports bettors is also notable, suggesting there is much greater variation in betting frequency among sports bettors than non-sports bettors.

Descriptive statistics of survey respondents for the full sample and by the different types of bettors. Full sample includes people who responded that they did not gamble at all. Standard deviations provided in parentheses only for variables that are not dichotomous

Observations for Non-Sports Bets and Sports Bets variable are 14,293 and 14,686 respectively for the full sample. For Non-Sports bettors only, observations for Non-Sports Bets are 5102. For Sports bettors only, observations for Sports Bets are 2285. For Sports + Non-Sports bettors, observations for Non-Sports Bets and Sports Bets are 1658 and 2051 respectively

The second component of the data used in this study comprises demographic characteristics. This includes gender, age, location (metropolitan or regional), marital status, education, employment status, income categories, country of birth, parents’ country of birth, Aboriginal Torres Strait islander origin and health status. The data are summarised for the full sample and the different betting categories in Table ​ Table1. 1 . Some key features of the data are that about 70% of the sample is male but nearly 80% of sports bettors are male and the average age of the sample is 50 years, but sports bettors have an average age of 43 years for sports bettors and 46 years for those who both sports and non-sports bet. Two other notable features are the much higher proportion of non-sports bettors who are retirees (0.24) compared to the proportion of sports bettors (0.07) and the high proportion of responders who did not report their income (0.20).

The third component of the data used here comprises responses to questions about attitudes to gambling. The questions are grouped into two categories including (i) general attitudes to gambling and sports betting; and (ii) perceptions of the attitudes and behaviours of others. Responses were elicited on a scale from 0 (totally disagree) to 10 (totally agree). The questions are presented in Table ​ Table2, 2 , where sample means and standard deviations are presented for the full sample and the different betting categories.

Summary statistics of responses to questions about gambling for the full sample and by the different types of bettors. Standard deviations in parentheses

Means of each gambling group are compared to non-gamblers using a t -test with a Sidak correction to account for repeated testing, with the level of statistical significance denoted by * (10%), ** (5%) and *** (1%). P -values of an F -test of whether mean responses of each group are statistically significantly different from each other presented in Full Sample column in braces

An important research question here is whether people in different betting categories respond to each question differently. We conducted a one-way analysis of variance (ANOVA) for each question to test whether the mean responses of each group are statistically significantly different. The p -value of this joint F -test for each question is presented in the ‘Full Sample’ column in braces; all tests have p -value of [0.000] implying we reject the null hypothesis of equality between the mean response of each group to each question. We also test the differences between means for each group using t -tests with a Sidak correction to account for the possibility of a false positive finding given the large number of t- tests. The table reports results of tests of differences between the mean response of non-bettors and each type of betting group with statistically significant differences at the 10%, 5% and 1% levels denoted by *, ** and *** respectively. The results show average responses of almost all betting groups differ from those of non-betters at the 1% level of significance. Differences are most stark in relation to the notion that ‘sports betting should not be part of experiencing sport’ and items related to the social aspects of sports betting, particularly the place of sports betting in the person’s family and friendship groups. However, responses to (i) ‘most people in society think betting on sport is harmless’ were not statistically different between sports bettors and non-bettors; and (ii) ‘odds talk is common in discussions about sport with my friends and peers’ were not statistically different between non-bettors and non-sports bettors.

An important feature of the survey is that it reached beyond people who identify as gamblers, providing insights into a more diverse sample than many previous studies. However, a limitation is that the design is focused on members, fans or supporters of a group of elite clubs or teams, implying to some degree that respondents are likely to be more engaged with sport than the average member of the Victorian population. Therefore, the differences between bettors and non-bettors identified here are potentially lower bounds and analysis of a more representative sample might uncover even greater differences. Another important benefit of the sampling frame is that it is likely the target audience for sports betting advertisers. Therefore, the analysis offers insights into a group that is likely most targeted and affected by sports betting advertising and understanding this group provides valuable insights to harm minimization policies with respect to sports betting.

Empirical Methods

The empirical analysis can be divided into two broad approaches. The first is to analyse the determinants of the betting choices of survey respondents. The second is to analyse the responses to gambling attitude questions. The approaches to these analyses are described in turn below.

Who Bets and How Often?

Each survey respondent is assumed to choose between four types of betting activity: no betting, non-sports betting, sports betting or both sports and non-sports betting. We define the gambling choice of each survey respondent i as G i = j ∈ 1 , 2 , 3 , 4 . We want to understand the relationships between different demographic characteristics’ and gambling choices. Given these four possible gambling choices or outcomes are unordered, we used a multinomial Probit specification to estimate the relationships. The probability that individual i makes gambling choice G i = j is given by

where X i is a vector of personal characteristics (including gender, age, location, marital status, education, employment status, income categories, country of birth, parents’ country of birth, Aboriginal Torres Strait islander origin and health status, as listed in Table ​ Table1), 1 ), Φ . is the cumulative density function of the Standard Normal distribution and β j provides the parameter estimates on X i for gambling choice j ; see Greene ( 2018 ) Chapter 18 for more details. This model allows us to understand the influence of each characteristic on the four possible gambling choices from a model that jointly estimates the probabilities of each alternative gambling choice.

As parameter estimates do not have a clear intuitive interpretation in such models, we compute marginal effects for each X i which are given by

where β ¯ is the probability weighted average of the parameter estimate across the four different possible gambling choices. The multinomial Probit results reported in Table ​ Table3 3 below are average marginal effects which are computed as the average of for δ ij across all i individuals. The interpretation of these marginal effects is that they tell us the impact of a unit increase in X i (for example female versus male or a one-year increase in age) on the probability of making gambling choice j ; that is, four marginal effects will be reported for each variable, one for each of no betting, non-sports betting, sports betting or both sports and non-sports betting.

Results of estimation of multinomial Probit model of betting behaviour. Average marginal effects on each betting behaviour are presented

Model includes linear, second and third order age terms. All age terms interacted with gender indicator. Standard errors in parentheses. Level of statistical significance denoted by * (10%), ** (5%) and *** (1%). Sample size N  = 14,813

Along with the choice of gambling type, individuals choose how many bets to place, and the factors that influence this number of bets are also of interest. Since 35% of survey respondents are non-bettors, the data on the number of bets comprises a large number of zeros. To accommodate this feature of the data, a Cragg hurdle model is adopted (Cragg, 1971 ). This model involves two parts: the first is a model of the decision to gamble (selection model), while the second is a model of the number of bets. As we have data on the number of sports bets and non-sports bets, we estimate this two-part model separately for each of these choices. The first part of the model is the selection decision, given by

where C i is individual i ’s choice of whether to bet on sports (1) or not (0) or alternatively to make non-sports bets (1) or not (0). The control variables X i are as defined above for the multinomial Probit model in Eq. ( 1 ), while α is a vector of parameters capturing the influence of each control variable on the decision to place bets or not and ϵ i is a mean zero, constant variance normally distributed disturbance term. The second part of the model estimates the number of observed bets. This continuous variable is given by

an exponential specification of the Cragg hurdle model where B i is the number of bets made per year by individual i , θ is the set of parameters reflecting the impact of variables X i , again as defined above and u i is a mean zero, constant variance disturbance term.

The key idea of this model is that the decision to gamble is modelled separately from the decision of how many bets to place. Estimates of α are important determinants of the decision not to gamble and therefore choose zero bets, whereas α and θ together determine the number of bets if a person chooses to gamble. The overall marginal effect of control variables X i , which in our specification are common to the selection and number of bets equations, are computed using the margins command in STATA and presented in Table ​ Table4 4 below. The detailed expressions for these marginal effects can be found in Burke ( 2009 ). The interpretation of these marginal effects is that they reflect the impact of a unit change in the value of a control variable X i (i.e. gender or age) on the average number of bets in a given period of time (a year in this instance).

Models of number of bets (intensity of gambling). Estimates are based on Cragg Hurdle regression. Average marginal effects are presented

Standard errors in parentheses. Level of statistical significance denoted by * (10%), ** (5%) and *** (1%)

Factors Affecting Attitudes Towards Sport Betting

Our analysis of responses to the 12 different questions about attitudes to sports betting listed in Table ​ Table2 2 comprises two key objectives. First, we are interested in the relationship between individual gambling choices and attitudes to sports betting—this involves the type of betting and the number of bets. It is anticipated that sports bettors will hold more positive attitudes towards sports betting than non-bettors or non-sports bettors. Second, we are also interested in the relationship between demographic characteristics and attitudes to sports betting. Responses to each question range on a scale from 0 to 10 but are standardized to have mean zero and standard deviation of one, to enable comparisons of effects of different variables across survey questions. These standardized responses are modelled using OLS. The model estimated is given by

where A i is the response of individual i to one of the 12 questions on their attitude to gambling listed in Table ​ Table2. 2 . For each question, two versions of the model are estimated where Z i is a vector of control variables, including all variables in X i which is as defined above, along with either (i) gambling choice, G i , (no betting, non-sports betting, sports betting or both sports and non-sports betting); or (ii) the number of sports and non-sports bets, B i , included. Model parameters are given by γ while ε i is a disturbance term with zero mean and constant variance. As the dependant variable, A i , is a standardized measure of responses to attitude questions, γ should be interpreted as the average number of standard deviations of change in A i per unit change in Z i .

In all models, we allow for non-linear age effects by including quadratic and higher order age terms, along with a linear age term, testing their significance using a likelihood ratio (LR) test. We also allow for gender effects to vary with age by including age and gender interactions and testing for their significance, again using a LR test.

Factors Affecting Gambling Behaviour

The results of estimation of Eq. ( 1 ) are presented in Table ​ Table3. 3 . The average marginal effects of the listed control variables on different gambling choices of no betting, non-sports betting, sports betting and both sports and non-sports betting are presented in first, second, third and fourth columns respectively. The model includes linear, quadratic and third order age terms. 2 Focusing on the cases of sports betting only (third column) and both types of betting (fourth column) and on results that are significant at the 1% level (denoted by ***), we found that relative to males, females are 9.6% less likely to bet on sport and 6.2% less likely to bet on both. To test the hypothesis that gender effects vary with age, all the included age terms are interacted with the gender indictor, with these interactions supported by a LR test = 62.63 ( p value = 0.00). The marginal effect for females relative to males is plotted for each betting category in panels (a)–(d) of Fig.  1 . The difference between men and women sports betting is greatest among the youngest in the sample and the difference decreases with age. The result suggests that young men are up to 25% more likely than women of the same age to bet on sports; the difference is less than 10% for people over 50 years. In addition, the average marginal effect of a 1-year increase in age is reported in Table ​ Table3. 3 . Comparing otherwise identical individuals with a 10-year age difference, the older person is 5.0% less likely to sports bet only and 2.0% less likely to make both sports and non-sports bets relative to the younger person.

Marginal effect of gender on the probability of each betting category plotted over age. Panels ( a )–( d ) are based on estimates presented in columns (1)–(4) of Table ​ Table3 3 respectively. Point estimates are denoted by dots with 95% confidence intervals for each point estimate included

Relative to self-employed, which is the base category, students are 6.7% less likely to bet on sports, while those on home duties are 7.0% less likely to gamble on both sports and non-sports. Income results are all relative to the base range of “less than $10,000”, with little difference between different income ranges in the probability of sports betting only. However, respondents above $60,000 are between 5.6 and 8.0% more likely to bet on both sports and non-sports than those in the base income range. This suggests there is little effect on gambling probability of additional income as the marginal effects are similar for each category above $60,000. Finally, respondents whose parents were both born overseas were 2.7% less likely than the base category (both parents born in Australia) to bet on both sports and non-sports.

In Table ​ Table4, 4 , we present estimates of the model of the number of bets specified in Eqs. ( 3 ) and ( 4 ). The marginal effects of the full set of control variables on the number of non-sports bets and sports bets are presented in first and second columns of Table ​ Table4 4 respectively. Age is included in these models through linear, quadratic and cubic terms. 3 As the number of bets is a continuous variable, the marginal effects are interpreted as the impact of a unit change in the control variable on the number of bets per year.

Focusing first on results significant at the 1% level (denoted by ***) for the number of non-sports bets, females make on average 9 fewer bets per year than males. Betting increases with age, with a 10-year older person making on average 5.29 more bets. The model also includes an interaction between all age terms and gender to test the hypothesis that gender effects vary with age — this interaction is supported relative to the model without the interactions, LR test = 28.41 ( p value = 0.00). The marginal effect of gender on the number of non-sports bets increases with age with women over 58 placing around 20 fewer bets than men of the same age: panel (a), Fig.  2 . Further results include that someone who is married makes 4.5 fewer bets than a single person and parents place 5.0 fewer bets than people with no children. Education reduces betting with those holding trade qualifications betting 6.3 times less than someone who did not complete high school while those with university qualifications betting 15.3 fewer times than someone who did not complete high school. Employment status and income are uncorrelated with the number of non-sports bets. Respondents whose parents were both born overseas place on average 4.4 fewer bets than people whose parents were both born in Australia.

Marginal effect of gender on the ( a ) number of non-sports bets and ( b ) number of sports bets, plotted over age. Panels ( a ) and ( b ) are based on estimates presented in columns (1) and (2) of Table ​ Table4 4 respectively. Point estimates are denoted by dots with 95% confidence intervals for each point estimate included

Again, focusing on results significant at the 1% level, results for the number of sports bets show that women place on average 18.2 fewer bets than men per year. In contrast to non-sports bets, the number of sports bets placed decreases with age; a person 10-years older bets on average 5.0 fewer times per year. The interaction between age and gender confirms that the gender effect on the number of sports bets does vary with age—supported relative to the model without the interactions, LR test = 37.52 ( p value = 0.00). However, the marginal effect of gender on the number of sports bets decreases with age, with 18-year-old women placing 50 fewer sports bets than 18-year-old men. This difference is as low as 10 fewer sports bets per year when comparing women and men aged over 60 years: panel (b), Fig.  2 . Education effects are similar to those for non-sports bets. People with trade qualifications bet on average 6.7 (10.0) times less than someone who did not complete high school. The only employment status category that is related to the number of sports bets is being a student, who place on average 10.3 fewer bets than the self-employed. The number of sports bets is related to income with most annual income categories above $40,000 betting on average between 8 and 11 more times per year, though some differences are significant only at the 5% level and others are smaller with 5 more bets per year and significant at the 10% level. Finally, respondents whose parents were both born overseas place 4.2 fewer sport bets on average than people whose parents were both born in Australia.

Attitudes to Sports Betting

Selected results of estimating the model presented in Eq. ( 5 ) using the responses to the first set of questions on general attitudes about sports betting summarized in Table ​ Table2 2 are presented in Tables ​ Tables5 5 and ​ and6. 6 . Full results of these models are available in Online Resource 1 and Online Resource 2 of the Supplementary Materials, where estimates for all variables included in the models are presented. In all these models, an interaction between gender and all age terms (up to third order age terms are included in all models) is considered with the result of a LR test of the restriction that the coefficients on the interactions are zero presented in the last row of each column. In cases where the restriction is rejected and the interaction is non-zero ( p value < 0.05), the model presented includes the interactions.

Selected results of general attitudes about gambling and sports betting with focus on type of gambling behaviour included as explanatory variables

Dependant variable modelled in each column is response to the respectively labelled statement below. Responses to each question are standardised to have a mean of zero and standard deviation of one. Age is included as a third order polynomial with average marginal effects presented. P value of likelihood ratio ( LR ) test of interaction between age and gender presented in last row with interaction included if p value < 0.05. Standard errors in parentheses. Level of statistical significance denoted by *(10%), **(5%) and ***(1%). All models also include controls for marital status, being a parent, employment status, income bands, country of birth, parent’s country of birth, ATSI and health condition or disability over past 6 months – full results available in Supplementary Materials

(1): Sports betting should not be part of experiencing sport

(2): People who bet regularly on sport are at risk of harm from gambling

(3): Sports betting can place people at higher risk of relationship problems, mental health and wellbeing issues and money worries

(4): Regular discussion of the ‘odds’ when talking about sport can lead to gambling problems in individuals

(5): It’s easy for people with sports betting issues to stop gambling

Selected results of general attitudes about gambling and sports betting with focus on the number of non-sport and sport bets per year as explanatory variables

Dependant variable modelled in each column is response to the respectively labelled statement below. Responses to each question are standardised to have a mean of zero and standard deviation of one. Age is included as a third order polynomial with average marginal effects presented. P -value of likelihood ratio ( LR ) test of interaction between age and gender presented in last row with interaction included if p -value < 0.05. Standard errors in parentheses. Level of statistical significance denoted by *(10%), **(5%) and ***(1%). All models also include controls for marital status, being a parent, employment status, income bands, country of birth, parent’s country of birth, ATSI and health condition or disability over past 6 months—full results available in Supplementary Materials

Results in Table ​ Table5 5 are for the model estimated with all the demographic control variables in X i together with a set of indicators for each type of betting behaviour, G i , including non-sports betting, sports betting and both sports and non-sports betting, with no betting the omitted base category. Each column presents estimates for a model of standardized responses (mean zero and standard deviation of one) to a separate statement about sports betting. The statements upon which each dependent variable is based are listed in the table notes. Focusing on results significant at the 1% level (denoted by ***), the results show that after controlling for a large set of individual demographic characteristics, people who bet are on average less concerned about sport betting issues than non-bettors. This is evident for all 5 statements modelled. We can see in column (1), for example, responses to the statement ‘sports betting should not be part of experiencing sport’, relative to non-bettors, the average response of people who bet on non-sport only is 0.20 standard deviations lower, while people who bet on sport (only or both sport and non-sport) have an average response that is 0.71 standard deviations lower. Other key results from these models are that females are more concerned about sports betting than males, except for in their responses to the statement ‘people who bet regularly on sport are at risk of harm from gambling’, where there is no difference between men and women. The relationship with age is statistically significant for the statements ‘sports betting should not be part of experiencing sport’ and ‘people who bet regularly on sport are at risk of harm from gambling’ but the effects are small, with a 10-year older person having on average a 0.06 standard deviation higher response to the former question and a 0.03 standard deviation lower response to the latter. People from regional locations are on average more concerned about sports betting than people from metropolitan locations; however, this concern is not evident in response to ‘sports betting should not be part of experiencing sport’. The results on education show that relative to the base case of ‘did not complete high school’, those with trade qualifications are on average more concerned about sports betting and in turn, people with university education are even more concerned with even greater differences evident than for those with trade qualifications.

The results presented in Table ​ Table6 6 are for models of the same questions with the same demographic controls included but with betting behaviour replaced by the number of sports and non-sports bets, B i / 100 . Once again, each column presents estimates for a model of standardized responses (mean zero and standard deviation of one) to a separate statement about sports betting. The statements upon which each dependent variable is based are listed in the notes to the table. The impacts of the demographic characteristics in Table ​ Table6 6 are qualitatively similar to those found in Table ​ Table5. 5 . The key difference between Tables ​ Tables5 5 and ​ and6 6 is that gambling categories are replaced with the number of sports bets and the number of non-sports bets. On average, people who bet more often are less concerned about sports betting. However, 100 more sports bets per year (approximately 2 bets per week) has nearly double the impact on responses of 100 more non-sports bets. For example, responses to ‘sports betting should not be part of experiencing sport’ are on average 0.14 standard deviations lower for every additional 100 non-sports bets but are 0.30 standard deviations lower for every additional 100 sports bets.

The above analysis is repeated for the second set of statements summarized in Table ​ Table2 2 which focus on perceptions of the sports betting attitudes and behaviours of others. Selected results of this analysis with betting categories included are presented in Table ​ Table7 7 and with the number of bets included are presented in Table ​ Table8. 8 . Each column presents estimates for a model of responses to a separate statement about sports betting. Survey responses used to estimate each model have been standardized to have mean zero and standard deviation of one. The statements upon which each dependent variable is based are listed in the notes to each table. Full results of these models are available in Online Resource 3 and Online Resource 4 of the Supplementary Materials.

Selected results of perceptions of other’s attitudes and behaviours with respect to sports betting with focus on type of gambling behaviour included as explanatory variables

Dependant variable modelled in each column is response to the respectively labelled statement below. Responses to each question are standardised to have a mean of zero and standard deviation of one. Age is included as a third order polynomial with average marginal effects presented. P value of likelihood ratio ( LR ) test of interaction between age and gender presented in last row with interaction included if p value < 0.05. Standard errors in parentheses. Level of statistical significance denoted by *(10%), **(5%) and ***(1%). All models also include controls for marital status, being a parent, employment status, income bands, country of birth, parent’s country of birth, ATSI and health condition or disability over past 6 months—full results available in Supplementary Materials

(1): Most people in society think betting on sport is harmless

(2): Most people in society bet on sport

(3): Most people in my family think betting on sport is harmless

(4): Most people in my family bet on sport

(5): Most people in my friendship group think betting on sport is harmless

(6): Most people in my friendship group bet on sport

(7): Odds talk is common in discussions about sport with my friends and peers

Selected results of models of perceptions of other’s attitudes and behaviours with respect to sports betting with focus on the number of non-sport and sport bets per year as explanatory variables

Dependant variable modelled in each column is response to the respectively labelled statement below. Responses to each question are standardised to have a mean of zero and standard deviation of one. P value of likelihood ratio ( LR ) test of interaction between age and gender presented in last row with interaction included if p value < 0.05. Standard errors in parentheses. Level of statistical significance denoted by *(10%), **(5%) and ***(1%). All models also include controls for marital status, being a parent, employment status, income bands, Country of birth, parent’s country of birth, ATSI and health condition or disability over past 6 months—full results available in Supplementary Materials

Focusing on significance at the 1% level (denoted by ***), Table ​ Table7 7 shows, except for the first two statements that focus on attitudes in society, bettors have families and friendship groups where gambling is common and perceived as harmless. However, the difference between sports bettors and non-sports bettors is stark. For example, in response to the question ‘most people in my friendship group bet on sport’, the average response of non-sports bettors is 0.16 standard deviations higher than non-bettors whereas the average response of sports bettors is up to 0.50 standard deviations higher, with both significant at 1%. The difference between non-bettors and sports bettors (0.49 standard deviations) is nearly 10 times as large as the difference between non-bettors and non-sports bettors (0.05 standard deviations) in response to the question ‘odds talk is common in discussions about sport with my friends and peers’. The largest impact of age is on the peer and friendship group statements, columns (5)–(7), with a response of a person 10-years younger on average 0.15 standard deviations higher. The interaction between gender and age is illustrated for the results reported in column (6) which is based on the statement “most people in my friendship group bet on sport” in panel (a) of Fig.  3 . The difference between men and women is greatest at younger ages with women in the 18–45 year range responding on average 0.5 standard deviations lower than men, suggesting young men have a much stronger belief than young women that their friends are involved in sports betting.

Marginal effect of gender on response to the question “Most people in my friendship group bet on sport”, plotted over age. Panel ( a ) is based on model in column (6) from Table ​ Table7 7 which includes betting categories and panel ( b ) is based on model in column (6) from Table ​ Table8 8 which includes number of bets. Point estimates are denoted by dots with 95% confidence intervals for each point estimate included

The results presented in Table ​ Table8 8 are for models of the same questions analyzed in Table ​ Table7, 7 , with the same demographic controls included but with betting behaviour replaced by the number of sports and non-sports bets, B i / 100 . The relationships with the demographic characteristics are qualitatively similar to the results presented in Table ​ Table7 7 and discussed above. The more a respondent bets, the greater their agreement with all but the first two statements that focus on attitudes in society. The relationship between responses and number of sports bets is up to 4 times as large as the relationship with the number of non-sports bets. For example, responses to the question ‘most people in my friendship group bet on sport’ on average increase by 0.10 standard deviations for people who place 100 more non-sports bets per year (2 more bets per week) but for an otherwise identical person who places an additional 100 sports bets, their response is on average 0.38 standard deviations higher. These sorts of differences are evident across all questions about family and friendship groups, with a greater number of bets associated with responses that show sports betting is believed to be more common and perceived as less harmful in these circles. It is also found the more sports bets a person places, the stronger their agreement with the statement ‘most people in society bet on sport’, though the relationship with the number of non-sports bets is statistically insignificant. The gender and age interaction for the model in column (6) in Table ​ Table8 8 is presented in panel (b) of Fig.  3 . The figure shows the difference between men and women is greatest among 18–28-year-olds, with the responses of women in this age range on average 0.70 standard deviations lower. This compares with differences of less than 0.30 standard deviations for those over 60 years. The results are consistent with those in panel (a) of Fig.  3 , suggesting that gender age differences are robust whether we control for the betting category or the number of bets per year.

This study builds on existing literature at the intersection of sports betting and sports, providing a comprehensive analysis of the sports betting behaviour of sports fans, including many people who choose not to gamble at all. Survey respondents’ attitudes to sports betting were analysed using betting behaviour and a wide range of demographic characteristics. The approach differs from many previous studies as we targeted a broader demographic of sports fans, rather than focusing only on those engaged in gambling (sports or non-sports), which is a strength. We did not measure whether a person’s gambling behaviour is deemed ‘problematic’, but previous research has demonstrated a connection between frequency of sports betting and problematic gambling behaviour (Hing et al., 2016 ). Therefore, our analysis of the number of sports bets provides a useful proxy to identify those most at risk of experiencing gambling harm. The research was guided by two overarching questions addressed in turn through the following discussion.

Demographic Profile of Sports Bettors

The dominant theme emerging from our analysis is the importance of gender, age and their interaction. The gender difference in the probability of sports betting is wider among the youngest in the sample; 18-year-old men are about 25 percentage points more likely than their female counterparts to bet on sports, whereas this difference is less than 5 percentage points for those over 60 years. Similar patterns are evident for the number of sports bets placed, with younger men placing more bets than similar aged women and fewer bets being placed with each additional year of age. Consequently, young men are most at risk based on their sports betting engagement and number of bets placed. This aligns with previous studies (Hing et al., 2016 ; Williams et al., 2012 ), but widens our understanding of the sports betting behaviour of sports fans. Moreover, even though this has been described in smaller-scale qualitative research studies (Deans et al., 2017b ; Waitt et al., 2020 ) our results are based on empirical analytic techniques applied to a larger and more diverse sample. The results comprehensively demonstrate sports betting is predominantly pursued by young men, in sharp contrast to other forms of gambling. Given the recent growth of sports betting, its marketing, and increasing contribution to problem gambling (Hing et al., 2019 ), as well as the need for appropriately tailored prevention and early intervention public health initiatives, this finding is significant for highlighting the distinctive sports betting behaviour of young men aged 18–35. Recent research has started to examine young women aged 18–35 as an emerging gambling cohort (see McCarthy et al., 2020 ), but our results demonstrate no significant gender or age effects for women’s sports and non-sports betting behaviour.

Other important demographic factors included education level, relationship status, and employment status. People who are widowed or separated were less likely to bet on sports, but no other relationship types were significant at the 1% level. University educated individuals were less likely to bet on sports than those who did not complete high school. Employment status did not exhibit a strong relationship with sports betting, except students and unemployed were less likely than self-employed to bet on sports. Surprisingly, income did seemingly not influence whether people engaged in sports betting only. This was more important in the context of making both sports and non-sports bets—those reporting higher levels of income were more likely to engage in these gambling types.

Whilst other studies have reported various demographic risk factors for sports betting and gambling, our results contribute by clearly demonstrating the significant interaction between age and gender. The importance of our study for public health policy and harm reduction campaign strategies is twofold. First, our sampling frame is likely the target audience of sports betting marketers, providing strong evidence upon which to base public health policy and harm reduction campaigns. Second, such campaigns should be aimed specifically at young men to help counteract the increasing environmental and social normalisation of sports betting. The next section focuses on the key attitudinal differences that emerged from the results to answer our second research question.

Attitudes Associated with Sports Betting

Our results demonstrate there are significant differences between the attitudes of sports bettors (either sports betting or sports betting combined with non-sports betting), non-sport bettors and non-bettors. Not only do sports bettors feel more strongly that sports betting has a place in sport, they are also less concerned about the risks and harms of sports betting. These results help to demonstrate the effects of the normalisation processes outlined in previous studies. Whilst existing literature has documented how sports and sports betting have become synonymous (Milner et al., 2013 ; Nyemcosk et al., 2021 ; Pitt et al., 2016a ; Thomas, 2018 ), the attitudinal differences we identified highlight how the ‘gamblification’ (McGee, 2020 ) of sports has penetrated individual perceptions about sports betting as an activity and influenced behaviour. Moreover, the differences between sports bettors and non-sports bettors suggest something unique is happening for this group; it is not necessarily related to the act of gambling, but potentially broader environmental and socio-cultural influences. Gender and age effects are also apparent, with women less likely to agree that sports betting should be part of experiencing sports and more likely to agree that sports betting can place people at higher risk of other harms. A similar pattern is evident for age, with younger people generally being more permissive of sports betting.

The influence of the social aspects of sports betting, namely the characteristics of sports bettors’ social networks, is also a strong emerging theme, underpinned by several attitudinal measures. Sports bettors are more likely to have family and friendship groups where gambling is common and perceived as relatively harmless. Additionally, they are more likely to agree that discussions about odds and the placing of bets is the norm amongst peers. In this context, there are significant differences observed between both sports bettors and non-bettors, and sports bettors and non-sports bettors. This again highlights that there are potentially distinct socialisation processes specifically influencing the attitudes and behaviours of sports bettors. Whilst this has previously been described on a relatively smaller scale (Thomas, 2017a ), our research demonstrates that these interactional and socialisation factors are highly meaningful in an extensive cohort of sports fans.

Age and gender are the key demographic factors related to responses in a similar way to that described in the previous section. Whilst women are more likely to agree that sports betting is common in broader society and amongst family members, men are more likely to indicate it is common within their peer groups. Men are also more likely to state that ‘odds talk’ is prevalent when socialising with their peers. In combination with the demographic risk factors outlined in the previous section, it is apparent that men also have different interactions with sports betting. They are more likely to agree it has a place in sports, less likely to think it is risky and can lead to other harms, more likely to have friends and peers who bet on sports, and more likely to have dialogue that supports and endorses the normalisation of sports betting. These attitudes combined suggest it is imperative public health prevention measures and harm reduction interventions target young men. The impact of peer socialisation processes and hegemonic masculine norms around sports betting have been described in previous studies (Ayandele et al., 2019 ; Bunn et al., 2019 ; Deans et al., 2017a ). Sports betting has also been related to the development of socially valorised identities for young men (Lamont and Hing, 2021 ). Our research supports and builds on these previous findings by demonstrating in a large sample that it has become a more prevalent part of sports fandom for younger adult men.

On a large and unique scale, we have demonstrated fundamental attitudinal and behavioural differences, and distinct and concerning trends, among those who engage in sports betting, thereby offering important insights about those most at risk. Most previous research has been qualitative or focused on those identified as having problematic gambling behaviours. By contrast, the scale and type of results generated from this study have afforded the ability to compare differences between non-bettors and bettors, providing compelling evidence of current issues amongst a general cohort of sports fans. Importantly, this study provides data about an emerging public health crisis in which younger men are most at risk because they are more exposed to sports betting normalisation processes, show greater engagement with sports betting and express more permissive attitudes. As such, the results of this study provide a foundation for public health interventions and programs.

Below is the link to the electronic supplementary material.

Acknowledgements

We would like to acknowledge the Victorian Responsible Gambling Foundation for supporting and funding this research. We would also like to acknowledge and thank Joe Vecci and Roger Wilkins for helpful comments and suggestions.

Author Contributions

All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Emma Seal, Buly Cardak and Matthew Nicholson. The first draft of the manuscript was written by Emma Seal and Buly Cardak with guidance from Matthew Nicholson and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Open Access funding enabled and organized by CAUL and its Member Institutions. This research received funding from the Victorian Responsible Gambling Foundation.

Declarations

The authors have no competing interests to declare that are relevant to the content of this article.

1 One participant reported 52,000 bets per year while some others reported 5,200 and 2,600 bets per year. Such observations were treated as outliers and excluded from the analysis. This amounted to 14 observations being excluded from the analysis.

2 Higher order age terms are not supported by a likelihood ratio (LR) test of a model with a fourth order age term with a test statistic of 4.63 ( p -value = 0.20).

3 Higher order age terms are not supported by a likelihood ratio (LR) test of a model with a fourth order age term with a test statistic of 3.33 ( p -value = 0.19) for the model of the number of non-sports bets and 2.74 ( p -value = 0.25) for the model of the number of sports bets.

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

• Armstrong A, Carroll M. Gambling activity in Australia—findings from wave 15 of the HILDA survey. Australian Gambling Research Centre, Australian Institute of Family Studies; 2017. [ Google Scholar ]
• Ayandele O, Popoola O, Obosi A. Influence of demographic and psychological factors on attitudes toward sport betting among young adults in Southwest Nigeria. Journal of Gambling Studies. 2019; 35 :343–354. doi: 10.1007/s10899-019-09882-9. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Browne M, Langham E, Rawat V, Greer N, Li E, Rose J, Rockloff M, Donaldson P, Thorne H, Goodwin B, Bryden G, Best T. Assessing gambling-related harm in Victoria: A public health perspective. Victorian Responsible Gambling Foundation; 2016. [ Google Scholar ]
• Bunn C, Ireland R, Minton J, Holman D, Philpott M, Chambers S. Shirt sponsorship by gambling companies in the English and Scottish Premier Leagues: Global reach and public health concerns. Soccer and Society. 2019; 20 (6):824–835. doi: 10.1080/14660970.2018.1425682. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Burke WJ. Fitting and interpreting Cragg's tobit alternative using Stata. The Stata Journal. 2009; 9 (4):584–592. doi: 10.1177/1536867X0900900405. [ CrossRef ] [ Google Scholar ]
• Cowlishaw S, Kessler D. Problem gambling in the UK. Implications for health, pyschosocial adjustment and healthcare utilisation. European Addiction Research. 2016; 22 :90–98. doi: 10.1159/000437260pmid:26343859. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Cragg JG. Some statistical models for limited dependent variables with application to the demand for durable goods. Econometrica. 1971; 39 (5):829–844. doi: 10.2307/1909582. [ CrossRef ] [ Google Scholar ]
• Deans EG, Thomas SL, Daube M, Derevensky J. “I can sit on the beach and punt through my mobile phone”: The influence of physical and online environments on the gambling risk behaviours of young men. Social Science and Medicine. 2016; 166 :110–119. doi: 10.1016/j.socscimed.2016.08.017. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Deans EG, Thomas SL, Daube M, Derevensky J. The role of peer influences on the normalisation of sports wagering: A qualitative study of Australian men. Addiction Research and Theory. 2017; 25 (2):1–11. doi: 10.1080/16066359.2016.1205042. [ CrossRef ] [ Google Scholar ]
• Deans EG, Thomas SL, Daube M, Derevensky J, Gordon R. Creating symbolic cultures of consumption: An analysis of the content of sports wagering in advertisements in Australia. BMC Public Health. 2016; 16 :208–215. doi: 10.1186/s12889-016-2849-8. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Deans EG, Thomas SL, Derevensky J, Daube M. The influence of marketing on the sports betting attitudes and consumption behaviours of young men: Implications for harm reduction and prevention strategies. Harm Reduction Journal. 2017; 14 (1):1–12. doi: 10.1186/s12954-017-0131-8. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Dowling, N. (2014).  The impact of gambling problems on families  (AGRC Discussion Paper No. 1). Australian Gambling Research Centre, Melbourne.
• Dowling N, Oldenhof E, Cockman S, Suomi A, Merkouris S, Jackson A. Problem gambling and family violence: Factors associated with family violence victimization and perpetration in treatment-seeking gamblers. Journal of Interpersonal Violence. 2019; 36 (15–16):7654–7669. [ PubMed ] [ Google Scholar ]
• Fulton, C. (2017). Developments in the gambling area: Emerging trends and issues supporting the development of policy and legislation in Ireland. Department of Justice and Equality Report . Accessed 10 August 2021. http://hdl.handle.net/10197/8612
• Gainsbury SM, Hing N, Delfabbro P, Dewar G, King D. An exploratory study of interrelationships between social casino gaming, gambling, and problem gambling. International Journal of Mental Health and Addiction. 2014; 13 (1):136–153. doi: 10.1007/s11469-014-9526-x. [ CrossRef ] [ Google Scholar ]
• Greene WH. Econometric analysis. 8. Pearson Education; 2018. [ Google Scholar ]
• Hare S. Study of gambling and health in Victoria: Findings from the Victorian prevalence study 2014. Victorian Responsible Gambling Foundation; 2015. [ Google Scholar ]
• Hing N, Lamont M, Vitartas P, Fink E. Sports-embedded gambling promotions: A study of exposure, sports betting intention and problem gambling amongst adults. International Journal of Mental Health and Addiction. 2015; 13 (1):115–135. doi: 10.1007/s11469-014-9519-9. [ CrossRef ] [ Google Scholar ]
• Hing N, Russell A, Lamont M, Vitartas P. Bet anywhere, anytime: An analysis of Internet sports bettors’ responses to gambling promotions during sports broadcasts by problem gambling severity. Journal of Gambling Studies. 2017; 33 :1051–1065. doi: 10.1007/s10899-017-9671-9. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Hing N, Russell A, Thomas A, Jenkinson R. Wagering advertisements and inducements: Exposure and perceived influence on betting behaviour. Journal of Gambling Studies. 2019; 35 :793–811. doi: 10.1007/s10899-018-09823-y. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Hing N, Russell A, Vitartas P, Lamont M. Demographic, behavioural and normative risk factors for gambling problems amongst sports bettors. Journal of Gambling Studies. 2016; 32 :625–641. doi: 10.1007/s10899-015-9571-9. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Jenkinson, R., Sakata, R., Khokhar, T., Tajin, R. & Jatkar, U. (2020). Gambling in Australia during COVID-19 . Australian Institute of Family Studies Report, Australia.
• Johnstone P, Regan M. Gambling harm is everybody’s business: A public health approach and call to action. Public Health. 2020; 184 :63–66. doi: 10.1016/j.puhe.2020.06.010. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Lamont M, Hing N. Sports betting motivations among young men: An adaptive theory analysis. Leisure Sciences. 2021; 42 (2):185–204. doi: 10.1080/01490400.2018.1483852. [ CrossRef ] [ Google Scholar ]
• Lopez-Gonzalez H, Russell AMT, Hing N, Estévez A, Giffiths MD. A cross-cultural study of weekly sports bettors in Australia and Spain. Journal of Gambling Studies. 2020; 36 :937–955. doi: 10.1007/s10899-019-09898-1. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• McCarthy S, Thomas SL, Pitt H, Daube M, Cassidy R. It’s tradition to go down the pokies on your eighteenth birthday—the normalisation of gambling for young women in Australia. Australia and New Zealand Journal of Public Health. 2020; 44 :476–381. doi: 10.1111/1753-6405.13024. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• McGee D. On the normalisation of online sports gambling among young adult men in the UK: A public health perspective. Public Health. 2020; 184 :89–94. doi: 10.1016/j.puhe.2020.04.018. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Milner L, Hing N, Vitartas P, Lamont N. An exploratory study of embedded gambling promotion in Australian football television broadcasts. Communication, Politics, and Culture. 2013; 46 :177–198. [ Google Scholar ]
• Nyemcsok C, Thomas SL, Pitt H, Pettigrew S, Cassidy R, Daube M. Young people’s reflections on the factors contributing to the normalisation of gambling in Australia. Australia and New Zealand Journal of Public Health. 2021; 45 :165–170. doi: 10.1111/1753-6405.13063. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Pitt H, Thomas SL, Bestman A. Initiation, influence, and impact: Adolescents and parents discuss the marketing of gambling products during Australian sporting matches. BMC Public Health. 2016; 16 :967–979. doi: 10.1186/s12889-016-3610-z. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Pitt H, Thomas SL, Bestman A, Daube M, Derevensky J. Factors that influence children’s gambling attitudes and consumption intentions: Lessons for gambling harm prevention research, policies and advocacy strategies. Harm Reduction Journal. 2017; 14 (11):1–12. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Productivity Commission (1999). Australia’s gambling industries . Report no. 10. Australian government.
• Pitt H, Thomas SL, Bestman A, Stoneham M, Daube M. “It's just everywhere!” Children and parents discuss the marketing of sports wagering in Australia. Australian and New Zealand Journal of Public Health. 2016; 40 (5):480–486. doi: 10.1111/1753-6405.12564. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Purves RJ, Critchlow N, Morgan A, Stead M, Dobbie F. Examining the frequency and nature of gambling marketing in televised boradcasts of professional sporting events in the United Kingdom. Public Health. 2020; 184 :71–78. doi: 10.1016/j.puhe.2020.02.012. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Raybould JN, Larkin M, Tunney RJ. Is there a health inequality in gambling related harms? A systematic review. BMC Public Health. 2021; 21 :305. doi: 10.1186/s12889-021-10337-3. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Raymen T, Smith O. Lifestyle gambling, indebtedness and anxiety: A deviant leisure perspective. Journal of Consumer Culture. 2017; 20 (4):381–399. doi: 10.1177/1469540517736559. [ CrossRef ] [ Google Scholar ]
• Rockloff M, Browne M, Hing N, Thorne H, Russell A, Greer N, Tran K, Brook K, Sproston K. Victorian population gambling and health study 2018–2019. Victorian Responsible Gambling Foundation; 2020. [ Google Scholar ]
• Russell A, Langham E, Hing N. Social influences normalize gambling-related harm among higher risk gamblers. Journal of Behavioral Addictions. 2018; 7 (4):1100–1111. doi: 10.1556/2006.7.2018.139. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Suomi A, Dowling NA, Jackson AC. Problem gambling subtypes based on psychological distress, alcohol abuse and impulsivity. Addiction Behaviour. 2014; 39 :1741–1745. doi: 10.1016/j.addbeh.2014.07.023. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Thomas, S. L. (2014). Parents and adolescents discuss gambling advertising: A qualitative study . Victorian Responsible Gambling Foundation, Melbourne. Available from: http://www.responsiblegambling.vic.gov.au/__data/assets/pdf_file/0006/14676/Parents-andadolescents-discuss-gambling-advertising-a-qualitative-study.pdf
• Thomas SL, Bestman A, Pitt H, Cassidy R, McCarthy S, Nyemcsok C, Daube M. Young people’s awareness of the timing and placement of gambling advertising on traditional and social media platforms: A study of 11–16-year-olds in Australia. Harm Reduction Journal. 2018; 15 (1):1–13. doi: 10.1186/s12954-018-0254-6. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Thomas S, Lewis S, Duong J, McLeod C. Sports betting marketing during sporting events: A stadium and broadcast census of Australian Football League matches. Australian and New Zealand Journal of Public Health. 2012; 36 (2):145–152. doi: 10.1111/j.1753-6405.2012.00856.x. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Thomas SL, Randle M, Bestman A, Pitt H, Bowe SJ, Cowlishaw S, Daube M. Public attitudes towards gambling product harm and harm reduction strategies: An online study of 16–88 year olds in Victoria, Australia. Harm Reduction Journal. 2017; 14 (1):1–11. doi: 10.1186/s12954-017-0173-y. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Waitt G, Hayden C, Gordon R. Young men’s sports betting assemblages: Masculinities, homosociality and risky places. Social and Cultural Geography, Online First. 2020 doi: 10.1080/14649365.2020.1757139. [ CrossRef ] [ Google Scholar ]
• Wardle H, Reith G, Best D, McDaid D, Platt S. Measuring gambling-related harms: A framework for action. Gambling Commission; 2018. [ Google Scholar ]
• Williams RJ, Volberg RA, Stevens RM. The population prevalence of problem gambling: Methodological influences, standardized rates, jurisdictional differences, and worldwide trends. Problem Gambling Research Centre; 2012. [ Google Scholar ]

Sports Betting Strategy

In this guide, we’ll highlight the best sports betting strategies for beginner and advanced bettors. We’ll also include some expert tips to help you develop smart sports betting habits.

A good sports betting strategy can be the difference between success and failure. Sports are unpredictable by nature and no betting strategy is guaranteed to make money in the long run. But adopting good habits can help maximize your winnings and minimize your losses.

The Basics of Sports Betting Strategy

Betting strategies are available for beginner, intermediate, and advanced sports bettors. Regardless of how much experience you have, remember that a sound strategy can improve your chances of winning on a consistent basis. These four basic principles will give you a good sports betting foundation:

Manage Your Bankroll

When betting on sports, always have a set budget in mind to bet with. This money, referred to as your ‘bankroll’, is disposable income you can comfortably afford to lose. It may be a small fraction of your monthly income, for example.

With a set bankroll in mind, you can then think about stake size. We recommend only making bets with 1-5% of your bankroll as this will help you withstand any spells of bad luck.

A common mistake for a sports bettor on a losing streak is increasing bet size to try and win back their money. This is called chasing your losses and it’s an incorrect mindset. This often results in making irrational decisions and losing even more money.

The best way to avoid this is by sticking within your bankroll limits and applying the management strategy. This will help minimize your losses when you are on a cold streak.

Bankroll Example:

Let’s say your bankroll for the whole NFL season was $1,000 and you want to bet on 5 NFL games every week:

• Total bankroll: $1,000
• Bankroll per week: ($1,000/18) = $55.55
• Total game units: ($55.55/5) = $11.11

In this scenario, your average stake should be $11.11. Therefore, you could stake between $8 – $12 per game with a $1,000 NFL bankroll.

Bet Sober/Don’t Tilt

There is a reason why Las Vegas casinos offer free alcoholic drinks to gambling patrons. Betting while under the influence can impair your judgment and lead to irrational decision-making. Avoid being intoxicated if you want to become a winning sports bettor.

The second part of this tip refers to ’tilt.’ Tilt is a commonly used term in gambling to refer to players who let their emotions take over. Playing while sad, frustrated, scared, or with extremely positive emotions, are all classified as playing with tilt. These emotions often lead to bad decisions.

You want to be calm and rational when sports betting. To avoid playing with tilt, take regular breaks and check in with yourself. Take a walk and clear your mind before placing any more bets.

Do Your Homework

Yes, sports betting comes with homework! You will want to conduct thorough, high-quality research before placing your bets. Going with a hunch may work every once in a while, but it will not lead to consistent returns over time.

To profit from sports betting, look at stats, find trends, create sports betting systems , and analyze past games. Do everything possible to ensure you have value in your pick before placing a wager.

Reading opinions from experts can be a great way to catch up on the latest news including key injuries, travel schedules and trades. There are also other tools available to bettors that can help increase your chances of finding good bets.

Utilize Free Picks and Prediction Sites

Check out these resources next time you are researching which bets to make:

• Daily free picks from sports betting experts
• Free sports picks daily from state-of-the-art AI software
• Betting odds calculator

Don’t Overlook Line Shopping

Lastly, get the most out of every bet by shopping for the best possible odds. Check the lines at the  best sports betting sites  to find the highest value. This can be time-consuming, but it’s well worth the effort to find the best potential payout available.

Over the course of a sports betting season, you can pocket a lot of extra cash by shopping the lines. We only recommend sites with great odds, but the fluctuating, dynamic nature of these means that no one sportsbook always has the best value. For that reason, line shopping is a must.

What’s the Best Sports Betting Strategy?

There is no single secret to betting and winning on sports. There are simply too many variables, from different sports and bet types, to different betting styles and bankroll sizes. That said, a combination of multiple betting strategies is usually the most effective.

Some sports bettors swear by hedge , parlay or teaser betting . However, you should be wary of anyone who claims to have a foolproof sports betting strategy. The perfect strategy doesn’t exist!  However, combining sound bankroll management with thorough research and sport-specific strategy offers higher chances of success.

Sports Betting Strategy Guides

Our team of experts want to help you make smarter sports bets. We have dedicated guides to help you develop your strategy, whether you’re a betting rookie or looking to build on your existing skills:

Advanced Betting Strategy Guides

For more advanced bettors, we have dedicated guides to help you take your game to the next level:

Free Expert Picks and Predictions

The Sports Geek uses an AI computer to compile stats and crunch the numbers to give you free betting picks. In addition, our team of experts provide analysis and predictions on the biggest games of the week. These resources provide you valuable information to help inform your betting strategy.

Best Betting Strategies By Sport

Given the differences in gameplay, leagues and outcomes, you’ll want to hone your sports betting strategy based on your chosen sport. We’ve created sport-specific guides to help refine your staking:

Tips For Effective Sports Betting Strategy

Whether you’re adopting a basic or advanced sports betting strategy, there are some key principles you should stick to:

Understanding each bet type will help you find the best value in every match. Moneylines are the most popular, but eagle-eyed players will be able to spot when the  over/under ,  spread  or  prop bet  is a better option. Keep an open mind when reading the lines and don’t be afraid to make different types of bets.

Betting lines are dynamic. The Patriots might be the -110 favorites at  NFL betting sites  today, but big money bets on their opponents could see the odds shift to +120 by tomorrow. Bookmakers move odds to ensure that they will make profit no matter the outcome. Understanding this line movement can help you cash in when you see great odds.

Fading the public means to bet against the strong favorites. Betting against popular belief can be risky. But line movement, sporting unpredictability and good odds can make it profitable, too. If you’re looking to fade the public, we’d recommend betting futures early in the season when bookmakers tend to have better odds. Either that or betting on the home underdog.

As previously mentioned, a great sports betting strategy requires a sport-specific approach coupled with thorough research. Become an expert in one sport, or even one team, first. Once you’ve developed a method of  online sports betting  and have (hopefully) made a little money, you can start to look at betting on other sports too.

Backing your team is one of the greatest joys, but betting with your heart is not a sound sports betting strategy. Let’s say you’re a loyal Houston Rockets fan. While you may see the potential and talent within the team, statistically speaking, the squad has been the worst performing team in the NBA throughout 2023. Therefore, placing an  NBA bet  on them might not be the best choice. Betting with your personal feelings involved could lead to irrational wagers.

If you research thoroughly, become an expert, and bet based on the facts, you will be a highly selective bettor. This is to your benefit. Those who bet on a large range of outcomes at the same time are rarely successful in the long term. To increase your chances of success, be picky.

Online Betting Strategy Mistakes to Avoid

It’s hard to make money sports betting in the long term, but it does happen. If you have any hopes of being successful, you’ll need to avoid these common betting pitfalls:

Your bankroll budget should be a strict cap of only money that you can comfortably afford to lose. From there, you can divide your bankroll up and bet small fractions on each wager. Betting over and above a small amount is high-risk bankroll management .

Before you confirm your bet, double and triple check your betting slip. Make sure you’re betting on the right event, outcome, and with odds you’re happy with. Check your stake as well, as it’s easy to add an extra 0 when you’re working quickly.

Parlay bets come with higher odds and a larger house edge, so most professional bettors tend to avoid them. Nevertheless, if you’re keen to cash in on the extra value, it’s particularly important to shop around for top parlay odds. Each bookmaker uses a slightly different odds calculator , so you could miss out on significant value by not finding the best price.

Research using independent, respected outlets. Avoid searching for statistics and opinions that will reinforce your beliefs. Instead try to approach each bet objectively. Inaccurate statistics and misinformation (found on fan forums) can contradict official league statistics.

No sports betting system is fool-proof. For that reason, never follow one blindly. Keep track of your profit and loss, and check in regularly to see how the system is working out for you. Don’t be afraid to change up your habits if you need to.

Very few people make a significant amount of money through sports betting in the long run. We encourage all bettors to see it as a form of entertainment, rather than a moneymaker. Keep realistic expectations with regards to your betting strategy, and remember that Rome wasn’t built in a day.

Latest Betting Strategy Blogs and News

Is It Better to Have High or Low Odds in Sports Betting?

Build the Ultimate Sports Betting Bankroll

7 Facts About Wong Strategy in NFL Teaser Betting That Can Help You Profit

Get Exclusive Access to Winning Sports Betting Picks for Free

Click on the button below to get free picks delivered to your email daily...

Enter Your Details Below To Get FREE Picks...

Research Your Sports Bets

There is no right method for being successful at sports gambling. Like with most things in life, there are many different paths that can lead us to the same result.

When it comes to successful sports teams, some will focus on offense, some will specialize in defense, some teams prefer to target players with great speed, and others focus on size and strength.

Winning comes in all different shapes and sizes on the field and the same can be said for those who achieve success with sports betting.

Whether you prefer to bet favorites on the road, underdogs only at home, or something in between, you can be successful with a strategy that works for you.

The one underlying similarity between all successful sports gamblers is that they spend the time and effort to properly research a game before they bet it.

Knowledge Is Power

There are no shortcuts in gambling when researching your wager before a particular game. Information is king and is the ultimate tool to help you along the journey of making your final decision on which bet to make.

Traditional Stats

Traditional stats are the common statistics we associate with each major sport. In baseball, we are talking about batting average, home runs, RBI’s, and several others. In football, it’s passing yards, rushing yards, points for and against, just to name a few.

These stats are considered staples in each of their respective sports. They often provide a very good overview of how a team is playing and more importantly for bettors, they are very easy to find and understand.

There is a bit of a tendency to undervalue traditional stats, because as sports continue to evolve, so do sports analytics. However, these stats have been used for decades and are always a great place to begin your research.

Yes, there is new analytics and information available to consume, but just because there are many new stats doesn’t mean we should forget about the many old ones.

Advanced Analytics

Now that we have effectively explained to you the importance of traditional stats and reminded you not to neglect them, we also want to suggest that you embrace some of the new advanced analytics that is available for bettors of all skill sets.

For many of us who have been following sports passionately over a long period of time, some of the new analytics being talked about today are hard to understand and can be confusing as to what their actual value is.

All of us who watch baseball understand the batting average and how it is calculated. But how many of us are familiar with O-swing% or Z-contact%?

A quick google search for advanced analytics in baseball will give you a list of over 50 different new metrics being used to track success and failure in that sport. All of them are very cool, but very few of them are going to help you win a bet.

Spend the time to educate yourself on a couple of the more popular “new” stats being used in some of your favorite sports.

Baseball Advanced Analytics

In baseball, I will always look at a hitter’s batting average and on-base percentage, which gives a pretty good indication of how they are doing overall. However, I might also check out a hitters Exit Velocity or Barrells percentage of late, just to see how they are performing in the last 10 days.

If someone is 2 for their last 20 but has a bunch of really hard-hit balls (high Exit Velocity) then maybe you can determine that hitter was a little unlucky and hit a few line drives right at some defenders.

Maybe instead of being in a slump, that hitter is actually hitting balls very hard and feeling very confident at the plate.

Maybe that is the type of player I want to target in a home run prop bet because the odds are a reflection of their 2 for 20 average and not their Exit Velocity of late.

Setting Your Own Odds

One of the most effective tools a gambler can use to their advantage is the ability to create their own betting odds for games of interest and then compare those to the odds set out by sportsbooks.

Many advanced bettors won’t even look at the odds set by bookmakers until after they have made their own. By setting your own odds, you can learn a couple of key things.

The first thing a gambler can learn is : How good you are at handicapping sports. Although we are all trying to beat the sportsbooks, we must also acknowledge that they are very good at what they do.

Take the time to research five NFL games and then after your research, set a point spread and compare your spreads against the ones posted.

If your odds are significantly different than the odds being set by the ‘books, that means you still have some work to do on how to properly handicap games.

Game Previews Intel

Read up on opinions and game previews to see what information you didn’t include in your research or maybe you were over-valuing or under-valuing specific information about a game.

A common occurrence for a lot of us is to overestimate the impact of an injury to one player on the final outcome of a game. Of course, key players will be missed, and odds will reflect that.

But in team sports where every athlete on both rosters is uber-talented, we more likely over-estimate the value of an injury as opposed to under-estimate it.

With regular effort and practice, your spreads and odds will start to reflect those being posted by the sportsbooks. Once you develop an ability to accurately handicap games, you can start to look at some specific games where you might have a slight edge.

By creating your own lines and comparing those regularly to the public betting lines, you will start to learn your strengths and weaknesses vs the professional oddsmakers .

When we say knowledge is power, this kind of knowledge can be very powerful.

The Eyes Have It

Stats, stats, and more stats are the only way to go. Today’s game is all about spreadsheets, data scientists, and extremely complicated analytics. Stats never lie.

None of that is actually true of course. Analytics were in fact created to help verify what our eyes are telling us. We can look at a box score and quickly form a conclusion of how a game played out.

• Las Vegas Raiders 15
• Kansas City Chiefs 31

At a quick glance, we can see that the Chiefs won by 16, and knowing what we know about Kansas City, we can quickly convince ourselves that Patrick Mahomes must have played well and that Travis Kelce and Tyreek Hill also had good or great games.

Effectively we just recapped this game in our mind, based on a score and it took us less than 10 seconds. This is a very convenient way to handicap games, but unfortunately not very effective.

If you were watching the game and not just surfing boxscores, you might have noticed that the Raiders defense played very well and that Patrick Mahomes and company did enough to win the game, but didn’t look overly sharp.

Maybe turnovers were a big difference and the Chiefs were the benefit of some short fields. Maybe it was special teams or penalties or a third-string right tackle who was in over his head.

It’s very possible Mahomes and Kelce and Hill were good but not actually great and when the Chiefs are listed as -10 point favorites the following week, an -8 point spread is really more accurate.

The most effective way to accurately handicap a football team and their potential each week is to watch them play.

How To Bet On Your Favorite Teams

Many of our peers will tell you that successful gamblers should not bet on teams they prefer to cheer for. Others will suggest that if you aren’t prepared to bet against a team, then you probably shouldn’t bet for them.

At BestOdds.com we are here to encourage you to throw away all of that advice and if betting on your favorite team is something you want to do – then our advice is to bet away.

There are a lot of things to consider when you are betting on your favorite teams. If your main goal as a gambler is to maximize profits and show the highest possible ROI from all your bets, then you should be aware that sometimes we all have a bias towards the things we love and care for, including our favorite teams.

To Be Truly Profitable

To be truly profitable, you will have to do some self-analysis and try to determine if you are indeed a little biased towards your favorite club. You can research the opinions of media experts and even your own friend base to get their thoughts on your team and compare that to your own expectations.

Do other people share your sense of optimism about your squad, do others see the same potential in players who may not yet have produced, but who you believe is about to put it all together?

At the same time, if you love a specific club and follow them regularly, then maybe they are exactly the team you should be betting on.

If you live in Indiana and regularly watch the Pacers play, more than you watch the Lakers, Bulls, or Heat, it would seem logical that you are the most educated and most informed about the Pacers.

Maybe you have some insight the majority of the betting world (who live outside of Indiana) doesn’t have.

If being profitable is your main goal, when you are tracking and analyzing your betting history (you do track and analyze your betting results, right?) then you would expect to see an appropriate amount of bets made for and against your favorite team.

If you notice that you have an overwhelming amount of bets on your favorite team to win, then re-evaluate your strategy and whether betting on your most cherished squad is actually the right play.

Our biggest issue with those who suggest that one should never bet on their most favorite teams is because they are totally neglecting the fact that sports gambling is a very entertaining activity that can add additional levels of excitement and enjoyment to any event.

This is a fact regardless of how much time you choose to spend on researching a bet.

Sports gambling to the overwhelming majority of those who partake is a form of sports entertainment and is not a career. Sports fans come in all shapes, sizes and with varying levels of enthusiasm and knowledge of their team.

But they are all fans.

Some sports bettors follow every team and every game. Others passionately cheer on the home squad and often little else. Does this mean that those who only have the time and passion to cheer for a few local teams should be excluded from placing a wager?

Of course not.

When it comes to placing a wager on teams that you root for the most, it is important to understand your motives behind that bet.

If you treat sports gambling as a business and are trying to maximize profits, then you must take the time to analyze your betting history and make sure any biases, known or unknown aren’t sneaking in.

But if you are an enthusiastic fan of a specific team and you just want to add a little bit of excitement to a particular game, then we say you do you, and make sure you have a little fun along the way.

Thirsty For More Betting Knowledge? Check Our  Sports Betting Guides Hub .

About the author

For almost two decades, Sadonna has remained at the forefront of the gambling industry in the US and abroad, covering the latest news and legal update...

Our Newsletter

Thursday's most entertaining betting analysis and cherry-picked insights in your inbox .

Join +10,000 Sports Bettors & Fans

Book a demo

Submit an inquiry.

2024 US sports betting trends – How live streaming is shaping user engagement

In our new 2024 US bettor's research report, we explore the importance of live streaming in sportsbook apps and how it varies across different age groups. Among the youngest group of sports bettors, aged 21-24, 28% are very interested in live streaming, and two-fifth are somewhat interested, showing a strong engagement with this feature.

The 25-34 age group also displays high interest, with a third very interested and two in five bettors somewhat interested. This group shows a significant inclination towards using live streaming features in sportsbook apps. For those aged 35-44, interest remains strong, with 32% very interested and 41% somewhat interested, indicating that live streaming is a valuable addition for maintaining user engagement.

Bettors aged 45-54 and 55+ show slightly lower levels of interest, but the numbers are still substantial. In the 45-54 age group, a quarter are very interested (26%) and three in 10 bettors are somewhat interested, while among those aged 55+, 29% are very interested and 32% somewhat interested. This data highlights that live streaming features in sportsbook apps are appealing across all age groups, with younger users leading the trend.

The 2024 US Bettor’s Report seeks to answer three questions:

• Which sportsbooks are trending right now, and with which age groups? 
• Which in-app features matter most to current and prospective bettors? 
• What's the next frontier in sports betting innovation?

Related Content

What are Britons gifting their dads this Father’s Day?

10 key takeaways on American attitudes towards sustainability in 2024

US PR & Social Surveys Calendar – Summer 2024 Edition

UK public wary of AI yet open to its potential in healthcare and manufacturing

How often do Americans clean their appliances? The answer may surprise you

Deceptive deals or real savings? 51% of consumers say brands regularly float fake discounts

Bill rounding for charity - Are consumers happy to help or forced to donate?

2024 US sports betting analysis – How different age groups bet differently

The 2024 Grand National was safer and just as exciting, say Brits

Defusing consumer boycotts: How can brands regain customers and bounce back?

Unraveling brand boycotts: What are the top triggers for consumer boycotts across global markets?

A fifth of Brits wagered on the Grand National

AI app appeal among Britons – Who’s interested?

How long is too long to wait for a new season of TV?

Standout vs. Standard: Measuring ROI of product placements

APAC Spotlight Calendar - April, May and June

Exploring American’s interest in AI apps – who's engaged and who’s not?

US bettor research report 2024

Betting on the Rise: The Soaring Popularity of Football as a Gamblers' Haven

In the world of sports gambling, football has emerged as a clear champion, captivating the attention of bettors worldwide. With its thrilling matches, passionate fan base, and potential for lucrative wins, football has become a haven for those willing to take their chances.

As the sport's popularity continues to soar, so does its appeal as a gambling opportunity. Whether it's the English Premier League, the UEFA Champions League, or the World Cup, football events draw in millions of viewers eagerly placing their bets in the hopes of hitting the jackpot.

The rise of online betting platforms has further contributed to the exponential growth of football gambling. Now, enthusiasts can conveniently place their bets from the comfort of their homes, with many options and markets at their fingertips. If you are wondering how, click here to know more.

However, it's important to approach football gambling responsibly. While it can be an exhilarating pastime, it's crucial to set limits and gamble within one's means. With the right strategy and knowledge, football betting can offer an electrifying experience for both avid fans and casual bettors alike.

The Growth of the Football Betting Industry

The football betting industry has experienced remarkable growth in recent years, fueled by several factors. Firstly, the global popularity of football as a sport provides a vast potential market for betting. With matches taking place throughout the year, there is always an opportunity for bettors to get involved.

Furthermore, advancements in technology have made football betting more accessible than ever before. The proliferation of smartphones and the internet has allowed bookmakers to offer their services online, enabling users to place bets from anywhere at any time. This convenience has played a significant role in the industry's growth.

According to recent statistics, the football betting market is expected to reach a value of $155.49 billion by 2024, with a compound annual growth rate of 9.8%. This staggering growth is a testament to the increasing popularity of the sport as a gambling avenue.

Football Betting Statistics and Trends

To understand the phenomenon of football betting, it's essential to delve into some key statistics and trends that shape the industry. One notable trend is the rise of in-play betting, where bettors can place wagers during the course of a match. This real-time betting experience adds an extra layer of excitement and engagement for gamblers.

Another noteworthy statistic is the increasing prevalence of accumulator bets, also known as parlays. These bets involve combining multiple selections into a single wager, with the potential for significantly higher payouts. Accumulator bets have gained popularity due to their ability to turn small stakes into substantial winnings.

Additionally, the popularity of football betting has also given rise to a variety of specialized markets. Alongside traditional bets on match outcomes, bettors can now wager on specific events within a game, such as the number of corners, yellow cards, or goalscorers. These niche markets provide an extensive range of options for gamblers to explore.

Tips for Successful Football Betting

While football betting offers immense potential for excitement and profit, it's crucial to approach it with a strategic mindset. Here are some tips to enhance your chances of success:

Research and Knowledge: 

Stay informed about the latest news, team line-ups, injuries, and other relevant factors that may impact match outcomes. The more knowledge you have, the better equipped you'll be to make informed betting decisions.

Bankroll Management: 

Set a budget for your betting activities and stick to it. Avoid chasing losses or betting more than you can afford to lose. Responsible bankroll management is key to long-term success.

Focus on Value Bets: 

Look for bets where the odds offered by the bookmaker are higher than what you believe the actual probability of the outcome is. Identifying value bets can lead to consistent profits over time.

Specialize in Specific Leagues or Competitions: 

By focusing on a particular league or competition, you can become an expert in analyzing teams, players, and trends. This specialization can give you an edge over more casual bettors.

Track Your Bets and Analyze Results: 

Keep a record of your betting activities to identify patterns and areas for improvement. Analyzing your results can help refine your approach and identify profitable strategies.

Remember, successful football betting requires patience, discipline, and continuous learning. It's a journey of trial and error, but with the right approach, it can be a rewarding one.

Responsible Gambling in Football Betting

While the excitement of football betting is undeniable, it's crucial to engage in responsible gambling practices. Here are some guidelines to ensure a safe and enjoyable experience:

Set Limits: 

Establish a budget for your betting activities and stick to it. Avoid chasing losses or betting more than you can afford to lose.

Take Breaks: 

Take regular breaks from betting to maintain a healthy balance and perspective. Don't let gambling consume your life or negatively impact your relationships or financial well-being.

Recognize the Signs of Problem Gambling: 

Be aware of the signs of problem gambling, such as betting beyond your means, betting to escape problems, or neglecting other responsibilities due to gambling. Seek help if you suspect you or someone you know may have a gambling problem.

Use Self-Exclusion Tools: 

Many online betting platforms offer self-exclusion tools that allow users to limit their access to their accounts for a specified period. Utilize these tools if you feel the need to take a break or control your gambling habits.

Seek Support: 

If you need assistance or guidance regarding responsible gambling, reach out to helplines, support groups, or professional counselors specializing in gambling addiction.

Remember, gambling should be seen as a form of entertainment, and any potential winnings should be considered a bonus rather than a guaranteed income.

Conclusion: The Future of Football Betting

As the popularity of football continues to rise, the allure of football betting shows no signs of slowing down. With advancements in technology, the industry is poised for further growth and innovation.

In conclusion, football betting has become a global phenomenon, providing an exciting avenue for fans to engage with the sport on a deeper level. With its captivating matches, vast betting markets, and potential for lucrative wins, football betting continues to soar in popularity. By approaching it responsibly and arming ourselves with knowledge and strategy, we can fully enjoy the electrifying experience it offers. So, let's embrace the rise of football as a gamblers' haven and place our bets wisely!