A new study commissioned by the American Gaming Association found that gamblers scored higher on financial literacy than non-gamblers, while prediction market users tended to rate their mathematical ability more highly than their test results supported.
Researchers surveyed 3,201 U.S. adults online from July 15 to 21, 2026. After data-quality screening, the final sample included 3,149 participants: 537 sports bettors, 574 prediction market users, 557 casino players, 495 iGaming users and 986 non-gamblers.
Across all gambling groups, participants averaged 3.93 out of five on the financial literacy index, compared with 3.72 among non-gamblers. The study also classified 41% of gamblers as highly financially literate, compared with 27% of non-gamblers.
“This study shows that, contrary to popular belief, gamblers are financially savvy entertainment consumers – especially sports bettors who demonstrate a particularly high level of financial literacy and mathematical capability,” said David Forman, American Gaming Association’s Vice President of Research. “Prediction market” users tend to overestimate their mathematical skillset, often thinking they are better at mathematical reasoning than they actually are, unlike traditional sports bettors who have a more accurate self-assessment.
Sports Bettors Post the Highest Scores
The study measured financial literacy through five questions covering inflation, interest rates, diversification, bonds and mortgages.
Sports bettors achieved the strongest financial literacy score among gambling groups at 4.18. Prediction market users followed at 4.05, with casino players at 3.84 and iGaming users at 3.75.
The researchers also measured objective numeracy. Sports bettors again ranked highest with a score of 9.56. Prediction market users scored 9.30, almost matching the 9.28 recorded by non-gamblers despite expressing substantially greater confidence in their mathematical ability.
“Prediction market users are more likely to misjudge their mathematical ability,” according to the AGA. “Sports bettors show the greatest mathematical confidence and objective capability, meanwhile ‘prediction market’ users expressed nearly as much confidence in their mathematical skills but performed more in-line with non-gamblers when those skills were objectively tested.”
The association said that gap could create consumer-protection concerns when prediction platforms present sports-related activity using concepts associated with investing or trading.
“The research highlights the risks of “prediction markets” marketing sports betting as investing against “peers” and the dangers of misleading consumers into thinking they just have to be smart to win,” Forman stated.
Among the 574 prediction market users in the study, 258 reported using Kalshi and 226 used Polymarket. Robinhood had 173 users, DraftKings Predictions 130, Crypto.com 106, FanDuel Predicts 103 and Fanatics Markets 42.
The study found that 31% of prediction market participants used two platforms, while another 31% used three or more. Sports attracted 374 users in this group, with 270 participating only in sports markets.
“A substantial share of prediction market users are functioning as sports bettors, using prediction platforms as a functional substitute for regulated sportsbooks,” the study said.
Only 61 participants reported trading mention markets.
Casino Players Lead on Positive Play
Researchers also examined the relationship between financial literacy and responsible gambling behavior using the Positive Play Scale. The measure covers personal responsibility, gambling literacy, honesty and control, and setting limits in advance.
Participants with moderate or high financial literacy achieved significantly stronger Positive Play scores than those with low financial literacy.
Casino players recorded the highest Positive Play result at 4.31. Sports bettors followed at 4.21, prediction market users at 4.18 and iGaming players at 4.15.
The casino group’s result stood out because its financial literacy score trailed both sports bettors and prediction market users.
“Casinos have operated within regulated frameworks for decades and have developed extensive responsible gaming infrastructure like mandatory signage, pre-commitment tools, staff training, and operator-led intervention programs,” the AGA study said. “Consumers who gamble in these environments are routinely exposed to RG messaging and have had more time to internalize its norms.”
The study also linked differences across gambling segments to the maturity of their respective responsible gaming environments.
“Sports betting and iGaming, while regulated, are newer to the landscape and still maturing in their RG programming,” the study said. “The Positive Play data reflect these environmental differences.”
Findings Point to Further Research
The results suggest a positive relationship between financial literacy and responsible play, although the survey measured participants at a single point in time and relied on self-reported behavior.
The AGA said financial literacy education could strengthen responsible gaming resources by improving consumers’ understanding of risk. It also highlighted the difference between prediction market users’ confidence and their measured mathematical performance as an area requiring further attention.
“As we close out another Responsible Gaming Education month, this research also identifies a potential link between responsible gaming and financial literacy, laying the groundwork for future research into this important topic,” said Forman. “It reinforces the principle that gambling must be promoted – and approached – as what it is: a form of entertainment, not a financial strategy.”
Colin López, Ph.D., founder and principal of IN Research & Analytics LLC, and Jackson Sears, Ph.D., founder and principal of Matrix Consulting Group of North Carolina, conducted the research and analysis. Both are co-founders of the Betting, Experience, and Trading in Sports Research Center and assistant professors at the University of North Carolina at Chapel Hill.
The survey carried a margin of error of plus or minus two percentage points at the 95% confidence level.