A recent scandal involving a Green Beret allegedly betting on a classified US raid may be more than an isolated incident, as new research suggests it could be an extreme example of the small group of informed traders who significantly influence prices on platforms like Polymarket. According to a working paper released by researchers from the London Business School and Yale, the accuracy of prediction markets can be attributed to a tiny group of informed traders, rather than the collective knowledge of the crowd.
By analyzing 1.72 million accounts and $13.76 billion in trading volume on Polymarket from 2023 to 2025, the researchers found that just 3% of traders are responsible for the majority of price discovery, consistently predicting outcomes and moving prices in the right direction. The remaining 97% of traders, on the other hand, tend to lose money, providing liquidity and generating volume, but ultimately being on the losing side of trades against the informed minority. To distinguish between skill and luck, the researchers simulated each trader's bets 10,000 times, using a coin flip to determine the direction of the trade.
The results showed that among the biggest winners, only 12% consistently outperformed the benchmark, while many apparent winners were simply lucky. The study also found that when skilled traders account for a larger share of trading, prices move closer to the correct outcome, especially in the final stretch before resolution. However, the same edge that makes skilled traders valuable to price discovery raises concerns when that information is not public or is not supposed to be.
The researchers cite the example of the US removal of Nicolás Maduro from power in Venezuela, where three newly created Polymarket accounts placed unusually large bets on the contract before the price moved, collectively making over $630,000. While there is no evidence of wrongdoing, the incident highlights the risk of insider trades, which can move prices aggressively per dollar. Overall, the study challenges the idea that prediction markets work due to the collective knowledge of the crowd, instead suggesting that they work because of the presence of informed traders.