A recent study has shed light on the inner workings of prediction markets, suggesting that a mere 3% of traders are responsible for the accuracy of market predictions. This finding challenges the long-held assumption that the collective knowledge of participants is the driving force behind market accuracy. The study, which analyzed over 1.72 million accounts and $13.76 billion in trading volume on Polymarket, reveals that a small group of informed traders consistently predict outcomes and drive prices towards the correct outcome, while the remaining 97% of traders tend to lose money. The researchers employed a novel approach to distinguish between skilled traders and those who are simply lucky, by simulating each trader's bets 10,000 times and comparing the results to a benchmark.
The findings indicate that only 12% of the biggest winners by raw profit can be attributed to skill, rather than luck. Furthermore, the study shows that the activity of skilled traders improves market accuracy, particularly in the final stages before resolution, and that they are the first to react to new information.
However, the study also raises concerns about the potential for insider trading, highlighting a case where three newly created accounts placed large bets on a contract related to a classified US raid, making over $630,000 in profits. While insider trades are rare, they can have a significant impact on market prices. Overall, the study's findings challenge the conventional wisdom that prediction markets work due to the collective efforts of the crowd, and instead suggest that they are driven by the actions of a small group of informed traders.