Citizens Study Finds Retail Traders Lose More on Prediction Markets Than Sportsbooks

Retail users on prediction markets posted a median return of -8% from July 2025 through mid-March 2026, compared to -5% for sportsbook users over the same period, as the absence of profitable-user restrictions leaves casual traders exposed to professional market makers, according to a report from Citizens JMP Securities.

Only the highest-volume prediction market traders—those moving more than $500,000—generated positive returns (+2.6%), consistent with sharp-bettor benchmarks, while every cohort below that level was negative, sliding to -26.8% for users trading less than $100. No sportsbook cohort was profitable either, but the decay was less severe, with the $500,000-plus group at -0.6% and the smallest accounts at -29.3%.

The research, which cited transaction data from analytics company Juice Reel, found that prediction markets are drawing a younger demographic than traditional sportsbooks, with 24% of Kalshi users under 25 compared to 7% for DraftKings and FanDuel.

Why Retail Traders Fare Worse on Prediction Markets

No Limits on Winning Players

Unlike regulated sportsbooks, which limit or ban profitable users, prediction markets allow professional traders and market makers to operate freely. Isaac Rose-Berman, a professional bettor interviewed for the report, stated: “Those platforms [prediction markets], whatever their flaws, they don’t kick bettors out. So if you are a bettor with a lot of money, that is very appealing. They also allow you to capture that retail flow. … If you’re on a sportsbook, you can’t really capture the losses of other people, but on prediction markets, you can.”

Market-Making Opportunity

Rose-Berman noted that while he can log onto Kalshi and “put in $50,000 or $100,000 and start market making,” he is often limited to $50 or $100 per wager on DraftKings or FanDuel. Fellow professional bettor Canzhi Ye added: “It’s a new game. And that is what is attractive to me, being able to be on the other side of bets, that’s what I want do. We all want to be FanDuel or DraftKings, that’s the dream.”

The report noted that this market-making opportunity may be temporary, as platforms could eventually shift to institutional market makers.

Sportsbook Downloads Decline as Prediction Markets Attract Younger Users

User Demographics

According to Sensor Tower data cited in the report, the median age of a traditional sportsbook user is 35, compared to 31 for prediction markets. The under-25 demographic represents 7% of traditional sportsbook customers versus 24% on Kalshi. Roughly 90% of DraftKings revenue comes from users over 30.

Download Trends

From September 2025 through February 2026, FanDuel and DraftKings downloads fell 18% and 13% year-over-year, respectively. Over the same period, Kalshi logged 6.3 million downloads. Citizens JMP analyst Jordan Bender wrote that prediction markets may be “diverting potential new players, particularly in the regulated states, who might have otherwise downloaded a sports betting app.”

Geographic Distribution

The bulk of prediction market players are located in non-legal gaming states including California, Georgia, and Texas, as well as legal gaming states such as Florida, Illinois, New York, Pennsylvania, Washington, New Jersey, and North Carolina.

Industry Response and Future Outlook

Cannibalization Estimates

Gaming executives surveyed for the report downplayed the threat of prediction markets:

• DraftKings’ Jason Robins: Prediction markets are not materially incremental to existing customers

• Flutter’s Peter Jackson: No evidence of material cannibalization

• BetMGM’s Adam Greenblat: Estimated a low-to-mid-single-digit percentage impact on betting revenue

• Citizens JMP estimate: Approximately 5% cannibalization

Market Concentration Risk

Citizens JMP noted that of the 61 online sports betting platforms launched outside Nevada in the last eight years, only 31 remain active, with the top eight accounting for 98% of revenue. The report warned that prediction markets may follow a similar consolidation pattern.

March Madness Pricing Advantage

In a separate note, Citizens JMP declared Kalshi a winner through the first four days of March Madness, with the platform offering superior pricing (4.04% implied hold) compared to DraftKings (4.28%), FanDuel (4.40%), and Fanatics (4.47%) on moneyline and over/under bets. Polymarket’s pricing was “superior” to all due to its nearly zero-fee structure, though the platform is not widely available in the U.S.

Frequently Asked Questions

Why do retail traders lose more on prediction markets than sportsbooks?

Prediction markets do not limit or ban profitable users, leaving retail traders directly exposed to professional market makers and high-volume participants who consistently take the other side of less informed flow. Sportsbooks, by contrast, manage risk by limiting winning players and capturing revenue through parlay products.

Are prediction markets cannibalizing sportsbook revenue?

Gaming executives estimate a low-to-mid-single-digit percentage impact (approximately 5%), with little evidence of direct cannibalization. However, prediction markets are attracting a significantly younger demographic—24% of Kalshi users are under 25, compared to just 7% for DraftKings and FanDuel—suggesting they may be capturing the next generation of bettors before they ever download a sportsbook app.

Which prediction market users are profitable?

Only the highest-volume traders—those moving more than $500,000 on prediction markets—generated positive returns (+2.6%). Every lower-volume cohort posted negative returns, with users trading less than $100 experiencing a -26.8% median return.