Polymarket’s Insider Trading Problem Survives Bloomberg Fact Check

Published by James Harris on June 19, 2026June 19, 2026

Polymarket’s reputation for clean markets got a stress test this week, not from a regulator, but from a pseudonymous analyst picking apart a Bloomberg investigation line by line.

The analyst, Car, published a thread arguing that Bloomberg’s insider trading report on Trump-related prediction markets was built on a flawed foundation: blockchain analytics flags from Polysights that labeled $45 million in volume and 34,225 wallets as suspicious without distinguishing between algorithmic pattern-matching and actual evidence of wrongdoing. Car traced the specific wallet Bloomberg highlighted and found winnings of “a couple hundred thousand,” not the $1.5 million the report claimed, a discrepancy he says Bloomberg could have caught by checking Polymarket’s own public leaderboard, where the top winner on the relevant contract made $1.1 million and the flagged wallet never appeared. The wallet’s history, according to Car, showed a pattern of large bets on elections and sports, the kind of behavior that looks suspicious in a spreadsheet and unremarkable to anyone who has spent time watching how high-conviction traders actually operate on these platforms.

The problem is that the insider trading concern on Polymarket is not invented. It is documented.

In January 2026, a newly created account placed $32,000 on Nicolas Maduro’s removal hours before U.S. forces seized him, eventually collecting over $430,000. Israeli authorities charged two individuals in February for using classified military intelligence to bet on strikes against Iran, and Bubblemaps identified six accounts that collectively won $1 million on the exact timing of those strikes, all funded within 24 hours of the attack. Polymarket rewrote its integrity rules in March 2026 in direct response, banning trading on confidential information and tips from insiders. Congressman Ritchie Torres introduced legislation in January that would bar anyone with access to material non-public government information from trading on prediction markets at all, a bill that now carries more than 40 Democratic co-sponsors.

Car’s argument, then, is not that insider trading on Polymarket is impossible. He explicitly acknowledges one or two genuine cases may exist. His point is narrower: that journalists treating every well-timed large bet as evidence of insider knowledge misread how these markets function, where concentrated, high-conviction wagers are routine among experienced participants. That distinction matters for the platform’s regulatory trajectory because legislation written to address genuine misconduct can easily be drafted broadly enough to sweep in behavior that is simply skilled trading. Research published in April 2026 found that 84.1% of Polymarket traders have lost money and only 35 out of 2.5 million wallets have sustained earnings equivalent to a U.S. monthly salary for 12 consecutive months, which suggests the edge that looks like insider information usually belongs to a very small group of people who are simply better at this than everyone else.

The Torres bill’s 40-plus co-sponsors means this is no longer a theoretical regulatory threat. How Polymarket and its legal team engage with that process over the coming months will determine whether prediction markets get workable rules or ones written by people who cannot tell a whale from a whistleblower.