Polymarket has changed how it settles short-dated crypto prediction markets, replacing single-price snapshots with time-weighted average prices after months of trader complaints and a study pointing to likely manipulation around expiry.

The platform said the revised system will apply to its crypto up/down markets, with five-minute contracts using a 30-second average and 15-minute and four-hour contracts using a 60-second average. Prices will be delivered through Chainlink Data Streams.

Research and complaints drove the change

The rule change follows criticism that Polymarket’s prior settlement method allowed traders to profit by moving the underlying market at the exact moment a contract resolved. In a post announcing the update, the company said the move was intended to protect market integrity. It also said it would add $1 million in liquidity rewards across affected markets through August to support trading during the transition.

A recent study by researchers from Stanford University and Singapore Management University examined roughly two months of five-minute bitcoin contracts and identified patterns they said were consistent with settlement manipulation. Their analysis found that 821 accounts earned $8.2 million during settlement windows they classified as likely manipulated.

What the study found

According to the paper, unusually large orders appeared on Binance in the final seconds before Polymarket contracts settled, and bitcoin prices then quickly reversed. The researchers argued that this pointed to a structural weakness in contracts that settle on a financial price that can itself be moved through trading in the underlying market.

The authors did not claim to prove intent, and they did not directly establish that the traders placing spot orders were the same people holding positions on Polymarket. Still, they said the impact was concentrated on smaller participants: excluding market makers, 93% of losses in windows they deemed likely manipulated were borne by retail traders.

The study described a market in which outcomes priced as near-certain were sometimes overturned at settlement. The authors wrote that a bet treated by the market as almost guaranteed was reversed roughly one time in three.

Warnings had surfaced earlier

Concerns about Polymarket’s short-term crypto markets had been raised before the academic paper. On May 21, a pseudonymous onchain analyst known as Variance Lover published a detailed post arguing that traders could build positions on Polymarket and then push Binance prices during the settlement window to force favorable outcomes.

An Axis Robotics contributor using the name 郡主Christine on X also said manipulation in Polymarket’s five-minute bitcoin market appeared to be worsening, citing precise reversals in the final seconds before expiry. In response to that discussion, a Polymarket developer said the issue was being examined more closely.

How the new approach compares

Polymarket’s use of short TWAP windows mirrors safeguards used by rival prediction platform Kalshi, which says it resolves markets using a regulated CF Benchmarks price index and a 60-second moving average. Kalshi has argued that this makes brief price distortions harder and more expensive to carry out than systems based on instant snapshots.

Questions around resolution integrity are not unique to one venue. A Kalshi developer said the problem did not exist there, but another market participant replied that they had seen similar behavior. A Kalshi spokesperson told CoinDesk that offshore markets can still influence prices, though the company said arbitrageurs tend to correct artificial moves quickly.

Kalshi also said its identity-verified user base makes suspicious trading easier to investigate. The company reported conducting 150 to 250 material investigations per quarter and making about 40 to 50 referrals to the Commodity Futures Trading Commission so far this year, though those figures cover the business as a whole rather than only short-dated crypto markets.

Next step for Polymarket

For now, the confirmed change is the new averaging method for short-duration crypto contracts and the rollout of Chainlink Data Streams as the pricing source for those settlements. Whether the adjustment materially reduces the behavior identified by traders and researchers will likely become clearer only after more trading data accumulates under the new rules.

Source: www.coindesk.com