Polymarket has changed how its five-minute Bitcoin prediction markets determine opening and closing prices, replacing single-point snapshots with 60-second time-weighted averages. Early research suggests the reform was followed by the disappearance of a trading pattern previously observed around settlement on Binance, although the findings do not establish current trader profits or prove that every possible manipulation strategy has ended.
According to CryptoSlate, the change matters because the platform’s earlier contract design could make a brief movement in Bitcoin’s reference market especially important if it occurred at the precise moment used for settlement. Under the revised structure, a price move generally affects the benchmark according to how long it remains part of the measured average.
How Polymarket’s Bitcoin contracts now settle
A five-minute Bitcoin market gives traders a simple “Up” or “Down” outcome based on the relationship between two reference values. In the Oct. 27 contract covering 12:20 p.m. to 12:25 p.m. Eastern Time, the opening benchmark was compared with the closing benchmark. “Up” wins when the ending value is at least as high as the opening value; otherwise, “Down” wins.
The market used Chainlink’s BTC/USD 60-second time-weighted average price, or TWAP, as its resolution source. A TWAP is an average that gives prices weight according to the amount of time they cover, rather than treating every individual observation as equally important.
Polymarket’s changelog says crypto up/down markets moved from single snapshots to Chainlink-computed averages on Aug. 7. The five-minute markets were extended to use 60-second benchmarks from Aug. 14 at 00:00 UTC. The contract itself still lasts five minutes, but each boundary is now assessed through a one-minute average. That distinction is important: the benchmark window is not the same as the duration of the prediction market.
The averaging process also does not necessarily correspond to 60 equally weighted price ticks or begin and end at a round clock minute. Chainlink’s timestamp documentation describes report validity periods and observation times that can vary, including cases where a later report covers a gap. As a result, reviewing only the timestamp displayed on an exchange chart may not be enough to reconstruct the exact values used for settlement.
Early evidence after the reform
The newer research reviewed Binance one-second data from July 1, 2025, through Sept. 4, 2026. It examined 82,752 five-minute Bitcoin boundaries where only the five-minute contract settled and measured the proportion of Binance trading volume occurring during the final 10 seconds of each window.
In the first three weeks after the 60-second benchmark regime began, the study found no systematic shift of trading into earlier time periods. It also found no return of the near-strike reversal pattern identified in earlier research. In practical terms, the Binance trading behavior observed around settlement before the reform appeared to subside in the initial post-change sample.
That result provides an early behavioral test of the new settlement method. However, it has a narrower meaning than a complete assessment of trader activity. The research measured whether a particular trading pattern changed; it did not calculate the current gains or losses of individual wallets and did not show that every possible way of influencing a settlement price had disappeared.
Why the old $8.2 million estimate is not a current profit figure
Earlier research by Dai, Jia and Yu reported spikes in spot-market flows near settlement and subsequent reversals after the launch of the five-minute contract. The work estimated that 821 wallets earned about $8.2 million during earlier cycles in which prices could be influenced around a settlement snapshot.
That figure belongs to the historical analysis and cannot automatically be applied to the revised contracts. The change from a single observation to a 60-second average alters the economic calculation. A trader seeking to influence the prediction-market outcome may need to keep the underlying Bitcoin price at a level for longer, potentially increasing the cost of the strategy. The relevant calculation would also need to include execution costs and the result of any Bitcoin trades used to move or hedge the reference price.
A profit on the prediction-market position is therefore not the same as the net return from a combined strategy involving Bitcoin. Without current wallet-level information and a full accounting of underlying-market trades, the earlier estimate cannot show how much traders earn or lose under the new rule.
Moneyiar market picture
Moneyiar market data recorded at 22:39 on Oct. 25, 2026, showed Bitcoin at $80,430.51, down 0.08% over 24 hours and 1.32% over seven days. BNB was priced at $740.86, up 0.81% over 24 hours and 3.32% over seven days. Polygon (POL), formerly associated with the MATIC ticker, stood at $0.08981, up 0.39% over 24 hours but down 0.74% over seven days.
These market readings provide a snapshot of the underlying crypto market at that stated time; they do not measure the prices or benchmarks used to settle the Oct. 27 Polymarket contract.
