Only 3% of Traders Drive Prediction Market Accuracy
Polymarket study shows prediction markets work because of a tiny group of informed traders, not crowd wisdom. The Green Beret insider case may be an extreme example.
Study Findings
^ Scope: 1.72M accounts, $13.76B volume from 2023-2025Key
^ Result: 3% of traders drive price discovery and accuracy
^ Other 97%: Add liquidity but lose to the informed 3%
Skill vs Luck
Researchers ran 10K simulations per trader, flipping trade direction:
* Skill Test: Consistently beating random outcomes = skill
* Results: Only 12% of top profit makers passed
* Mean Reversion: ∼60% of "lucky winners" lost in follow-up tests
How Skilled Traders Move Markets
^ More skilled traders = higher accuracy, especially near resolution
^ React first to news like FOMC or earnings. Others don’t
^ Usually repeat players with consistent records
Insider Risk
Skill advantage raises issues when info isn’t public:
* Case: US overthrow of Nicolás Maduro in Venezuela
* Activity: 3 new Polymarket accounts bet big on "Maduro Overthrown" at 10% odds pre-operation
* Outcome: Made $630K+. Accounts went dormant. No charges filed
* Impact: Insider trades move markets 7-12x more per dollar than skilled trades, but are rare
Polymarket and Kalshi ban inside trading. Researchers say markets work due to informed traders, not crowds.
#PredictionMarkets #MarketEfficiency