Polymarket's five-minute binary contracts on Bitcoin price movement have proven to be not just a betting tool, but an effective mechanism for capital redistribution. A new study by analysts from Stanford and Singapore Management University has uncovered a troubling pattern: these contracts systematically funneled funds from retail traders to a narrow group of manipulators, while simultaneously distorting the spot price of the leading cryptocurrency.
The product, which did not exist before February 12, 2026, traded over $4 billion in just a few months, tripling the platform's daily trading volume. The mechanics are simple: the contract pays $1 if Bitcoin closes a five-minute window above the opening level, and $0 otherwise. The key flaw, according to the authors of the study "Settlement Manipulation in Prediction Markets," lay in the settlement mechanism via the Chainlink oracle, which averaged the price across major spot exchanges.
How the Scheme Worked
A trader holding a contract could buy or sell real Bitcoin in the final seconds of the window, shifting the reference price past the required threshold. The protection of mixing exchanges in the oracle proved illusory. The largest platform, Binance, deviated from the oracle's average by only 2.5 basis points and moved almost in sync with it. Pressure was applied precisely through Binance: in 85% of cases, the exchange ended up on the same side of the threshold as the settlement result. To win, it was enough to push the price on Binance by a few basis points past the threshold — and the outcome was decided.
A characteristic pullback confirms the manipulation: within ten seconds after settlement, the price retraced roughly a quarter of the movement in near-equilibrium cycles. Interventions mainly occurred during quiet hours — 56% at night and 44% on weekends.
Numbers and Consequences
In near-equilibrium cycles, a push against the favorite changed the winner in 65% of cases, compared to 41% in normal trading. Even when one side had a 90–100% chance before closing, a push reversed the outcome in 34% of cases. Only 821 traders fit the manipulator profile — roughly one in three hundred among the 243,000 participants. In cycles with pushes, they earned $8.2 million, while in others they broke even. Meanwhile, 93% of all losses fell on retail traders.
The authors dismissed the notion of harmless hedging: when hedging, a binary contract carries almost no risk when one side is close to winning — yet such cycles were precisely the ones where a push reversed the outcome. Moreover, trades were executed in a single burst in the last fifty seconds, not accumulated gradually.
According to the researchers, the solution lies in the contract duration. No manipulation was found on fifteen-minute contracts: a longer window allowed more regular trades to pass through, diluting a single push. Pulling off the manipulation became much more difficult.
My expert opinion: Polymarket, positioning itself as a decentralized and transparent prediction market, has in reality become an ideal environment for "skimming" retail players. The problem is not the blockchain itself — every transaction is visible — but the contract architecture, where a short window and weak oracle protection make manipulation trivial. If Nasdaq and Cboe, which have filed applications with the SEC for binary contracts on stock indices, do not learn this lesson, the same problem will migrate to much larger markets. Polymarket should either lengthen the window or revise the settlement mechanism; otherwise, trust in the platform will be completely undermined.