Published on: April 27, 2026, 11:55 AM.
Updated on: April 27, 2026, 11:55 AM.
- Only 3% of Polymarket users are responsible for the majority of accuracy on the platform, as revealed by a new analysis.
- The remaining users do not contribute to accuracy but serve as financial backers, the research suggests.
- A substantial portion of Polymarket users are classified as “unfortunate losers.”
Prediction markets like Kalshi and Polymarket rely on the principle of “wisdom of the crowd,” wherein a large number of participants expressing similar views can help validate those perspectives.
Contrary to common perceptions, a recent study by researchers from the London Business School and Yale University indicates that only 3.14% of Polymarket participants represent nearly all of the platform’s accuracy and credible price setting. They analyzed 1.72 million Polymarket accounts and $13.76 billion in trades between 2023 and 2025, concluding that “skilled winners” overwhelmingly dominate accuracy.
“Employing this classification, we segment ‘takers’—accounts primarily submitting market orders—into four categories based on profit significance and direction,” the study reveals. “Skilled winners, who consistently generate substantial profits, comprise 3.14% of Polymarket accounts. Lucky winners, yielding negligible profits, make up 29%. Unfortunate losers, facing minor losses, constitute 61.4%. Unskilled losers, experiencing significant losses, represent 6.4%. The leftover 0.1% are market makers, offering liquidity by placing limit orders.”
As the largest prediction market by trading volume until lately, Polymarket serves as a pertinent testing ground for this type of analysis, given that it settles transactions on a public blockchain, facilitating easy access to data for verifying traders’ successes or failures.
Challenges to Success on Polymarket
Although focusing on a separate aspect, the findings from the London Business School/Yale analysis align with prior studies confirming that a significant majority of prediction market traders struggle to achieve lasting success.
A 2025 analysis by Felix Reichenbach from Technische Universität Berlin and Martin Walther from the German International University revealed that 70% of Polymarket users report negative profit/loss figures. An even more recent study by researcher Andrey Sergeenkov shows that a staggering 84.1% of Polymarket users experience losses, with just 2% earning over $1,000 on the platform.
The findings from the London Business School/Yale research clearly indicate that skilled traders on Polymarket hold a degree of predictive ability, while the rest largely serve as liquidity providers.
“The order imbalance created by skilled traders strongly predicts subsequent price alterations and final market outcomes,” the researchers state. “Conversely, lucky winners, despite their positive returns, lack significant predictive capability. Unskilled losers consistently bet against the eventual results, whereas market makers support short-term order flow but ultimately end up on the losing side of informed trades at resolution, generating profits from providing liquidity rather than from directional skill.”
Distinguishing Skill from Chance
Some retail traders utilize copy-trading technologies to gain profitability in prediction markets; however, this approach carries risks. As highlighted by the London Business School/Yale team, a few fortuitous trades from a specific wallet can inflate perceived profits without demonstrating actual trader skill.
Essentially, it’s critical to differentiate skill from sheer luck; this insight is vital in prediction markets, where market makers and skilled takers make up only 3.5% of all accounts yet account for 30% of the total profits.
“Thus, the accuracy of prediction markets reflects the insight of an informed minority, rather than the collective wisdom of the crowd,” concludes the study.