What is a good win rate on Polymarket?

The distribution, by evidence depth

Method in one line: each wallet’s resolved positions are graded against the price actually paid, with bets at 2.5 cents or less set aside; each set of mutually exclusive outcomes, such as the candidates in one race, counts as one event, and only when the wallet held a position on just one of those markets, on one side, through resolution; a record lands in the ahead or behind column when its excess wins clear the 95% interval chance produces at those prices, in that direction. Full method on the methodology page.

The median wallet lands close to its own prices

In the 20 to 99 event tier, the median wallet wins 72.0% of its events while its entry prices implied 71.3%. Measured wallet by wallet, the median record runs 0.7 wins per hundred ahead of what its own prices predicted, and 54% of these wallets beat their prices. The typical record in this tier is mostly bets on favorites. The significance test runs in both directions. At 95% confidence, 1.8% of these records are ahead of their prices and 3.6% are behind them. The test is set so luck alone flags about 2.5% of records in each direction, so at this depth the ahead share is about what luck produces. The middle tier sits almost exactly at its prices. At 100 to 999 events the median record runs 0.1 wins per hundred behind them, 49% beat them, and at 95% confidence 8.1% are ahead against 9.1% behind. The 2026 SSRN study “Who Wins and Who Loses in Prediction Markets” found from account-level data that a small minority of participants drives the accuracy.

A perfect win rate can mean almost nothing

The 90th percentile win rate among wallets with 1,000 or more graded events is 99.9%. About one in ten of the deepest graded records has lost about one event in a thousand held to resolution. A record that close to perfect over that many events in practice only happens at extreme favorite prices, where winning was already close to certain, and the margin left over those prices is small. Across the whole tier the median excess is 0.9 wins per hundred. Meanwhile the folk thresholds you will find elsewhere, 60% or 65% win rate as a mark of skill, sit comfortably inside what chance plus favorable prices produces. If someone quotes a win rate at you without the prices behind it, they have told you nothing yet.

Evidence depth separates provable skill from noise

The ahead shares climb from 1.8% at 20 to 99 events, to 8.1% at 100 to 999, to 42.9% at 1,000 or more. Deep records make real edges provable and lucky streaks impossible to sustain. This is also why the leaderboard gates on resolved events. A short hot run, however profitable, cannot carry the evidence a ranking claim needs.

We retracted the July 2026 versions of these shares. The earlier figures, 6%, 27%, and 45%, were computed as the ahead tail alone, although the wording around them did not say so. They also came from a test that counted events where a wallet held both sides, or entered and exited before resolution; scoring those against a single entry price credited hedging and profit-taking as forecasting accuracy. On the corrected test, run on 2026-07-26, the same ahead-tail statistic read 1.3%, 4.5%, and 44.5%. The thin tiers fell to about a fifth and a sixth of their published values while the deepest tier held. Counting both directions, 13.3%, 34.3%, and 66.9% of records were distinguishable from their own prices one way or the other, and below 1,000 events the behind tail dominated. The table above carries the current figures, which drift a little with each data refresh. Mixed events were the raw material of thin records that looked provably ahead; deep directional records carry their evidence either way. The full revision history is in the correction log.

The figures moved further on 2026-09-24, when methodology v1.1 stopped scoring bets at 2.5 cents or less. An event that held a favorite beside a penny bet on the other side now grades as a one-sided favorite bet, so more wallets reach the 20 graded events this page counts, and thin records lean further toward favorites. Between the 2026-09-21 and 2026-09-24 runs the graded population rose from 402,094 wallets to 484,465. The share of records behind their prices at 95% confidence fell from 12.2% to 3.6% at 20 to 99 events and from 30.7% to 9.4% at 100 to 999. This page’s earlier heading, that the median wallet loses to its own prices, no longer holds and was withdrawn that day.

From the 2026-09-27 update the test counts only positions on a single market held on one side. Before, an event where a wallet bet on several of its markets, such as No on two candidates in a three-way race, counted only when every one of those bets won or every one lost, which depended on the result (see the correction log). Between the 2026-09-24 and 2026-09-27 runs, which span that change and three days of newly resolved markets, the graded population fell from 484,465 wallets to 459,913. At 20 to 99 events the median win rate moved from 75.0% to 72.0% and the median price-implied rate from 74.5% to 71.3%. Measured wallet by wallet, the median excess moved only from +0.8 to +0.7 wins per hundred. The ahead share at 1,000 or more events moved from 44.3% to 42.9%.

What is a good win rate on Polymarket?

There is no win rate that means skill on its own, because a win rate carries no information without the entry prices behind it. Among the 385,301 graded wallets with 20 to 99 resolved events (of 459,913 graded in all), the median win rate is 72.0% against a median price-implied win rate of 71.3%, so a typical win rate mostly restates the prices paid. The useful question is whether a record beat the prices it paid by more than chance allows, which is what the free checker tests.

How often does the Polymarket favorite win?

It depends on the price. In the 2026 Q3 edition of OVERROUND's calibration benchmark (markets resolved between July 2025 and June 2026), a market where the outcome Polymarket lists first (Yes in a yes/no market) was priced 50 to 60 cents a day before resolution resolved to that outcome 53.2% of the time (23,381 markets); at 60 to 70 cents, 63.6% (12,019); at 70 to 80 cents, 73.4% (9,402); at 80 to 90 cents, 80.5% (7,127); at 90 to 100 cents, 89.7% (13,257). Those bands count the first-listed outcome only, so they are a subset of favorites, and the benchmark's liquidity cuts, taken six hours and 30 days out, put most of the favorite shortfall in thinly traded markets (six hours out, favorites above 85 cents resolved 7.4 points under their price in the thinnest third of markets and 0.3 under in the most liquid third). Counting every market's above-50-cent side, including the No side of long shots, the side favored a day out went on to win 81.2% of the time across 259,361 markets, and 95.7% on the 21,992 longer-lived markets that already had a price a month before they resolved. Each market counts once and is scored on its last traded price a day before resolution; this measures Polymarket prices, not any trader, and a market with no trade a day out is absent from these counts.

Is a 60% or 65% win rate good on Polymarket?

No. Both thresholds sit inside what chance produces for a wallet that mostly buys favorites. A wallet buying 65 cent favorites should win about 65% of the time, and among graded wallets with 20 to 99 resolved events the median win rate is 72.0% against a median price-implied win rate of 71.3%. A win rate becomes informative only next to the prices paid.