Columbia Study Reveals 25% Wash Trading on Polymarket

Columbia University researchers have uncovered evidence that about 25 percent of trading volume on Polymarket, a leading blockchain-based prediction market platform, consists of wash trading, an artificial practice that inflates apparent activity levels. The findings, detailed in an 80-page academic paper titled "Network-Based Detection of Wash-Trading" and published this week, analyzed the platform's transaction history from July 2024 onward and identified suspicious patterns in 14 of its 1.26 million active wallets.

Wash trading involves traders buying and selling the same assets among themselves to simulate higher demand and liquidity, a tactic that is illegal in regulated U.S. markets because it distorts price signals and misleads participants about genuine interest. The study, which has not yet undergone peer review, used algorithmic clustering to detect clusters of wallets engaging in repetitive back-and-forth trades with minimal profit or loss, sometimes thousands of times. Researchers estimated that these activities accounted for roughly $4.5 billion in trades overall.

The analysis showed wash trading peaking at nearly 60 percent of weekly volume in July 2024, persisting...