Study: Polymarket Accuracy from Small Informed Traders

A new academic study analyzing every transaction on the Polymarket prediction market platform has concluded that its accuracy stems from a small group of informed traders, rather than the collective wisdom of broad participation. Researchers from London Business School and Yale University examined data from 2023 to 2025, covering 1.72 million accounts, 210,322 markets, and 13.76 billion dollars in trading volume.

The paper, authored by Roberto Gómez-Cram, Yunhan Guo, Theis Ingerslev Jensen, and Howard Kung, challenges the common industry claim that prediction markets aggregate diverse knowledge to produce reliable forecasts. Instead, the authors found that just 3.14 percent of accounts qualified as skilled winners, whose trading patterns consistently predicted short-term price movements and final event outcomes. Combined with market makers, this informed minority, comprising less than 3.5 percent of users, captured more than 30 percent of all gains on the platform.

To distinguish genuine skill from luck, the researchers developed a sign-randomization test. They simulated each trader's history 10,000 times by randomly flipping buy and sell directions while keeping trade sizes,...