Can you use insider trading in an investigation into your own insider trading? Polymarket just turned that question from philosophical to practical.
Blockchain sleuth ZachXBT published findings Thursday morning naming Axiom, a cryptocurrency trading platform, as the company whose employees he believed had used non-public information to make profitable trades.
The research had been teased for days, and Polymarket had created a contract that allowed users to bet on which company would be named, raising roughly $40 million in volume as of Monday.
The problem is that someone clearly knew the answer before it appeared.
Lookonchain identified 12 wallets that bet heavily on Axiom before the reveal, making a combined profit of more than $1 million.
A separate analysis by Polysights, a data terminal that tracks suspicious activity on Polymarket’s public ledger, detected five wallets that collectively staked around $50,000 and walked away with $266,000.
More on-chain data analyzed by CoinDesk tells the full story. The largest holder of Yes on the Axiom market, an account called predictorxyz, accumulated 477,415 shares at an average price of $0.14 and is now sitting on a profit of $411,000.
That’s about a 7x return on a bet placed before the answer was public. The second largest holder, an anonymous wallet, bought 109,450 shares at $0.33. The concentration is remarkable. This was not a broad market full of educated guesses. A handful of portfolios dominated the Axiom portion of the book.
For most of the week, another platform called Meteora had been the market favorite with more than 50% odds, as CoinDesk reported.
The odds changed in favor of Axiom on Wednesday night, which peaked at 46.2%. Anyone who bought Axiom stock in the period between that denial and ZachXBT’s release on Thursday morning was either reading the room very well or already knew what was coming.
ZachXBT acknowledged on social media that it had contacted Axiom for comment and conducted several interviews before publishing, making the leak “probably inevitable.”
That means several people at the company knew the report would be released before it was released. Any of them could have placed bets directly or tipped someone who did.
Polymarket’s offshore platform does not perform identity checks, making attribution difficult without the cooperation of the exchange itself.
Axiom said it was “shocked and disappointed” by the findings and would continue to investigate. He did not respond to questions about whether he was aware of any employees dealing with the Polymarket bet.
The structural irony here is that the mechanism worked exactly as designed. The people who were the subject of the research were simply rewarded rather than those who carried it out.
