How the signals have actually performed — including where the sample is still too small to say.
Last updated Sep 1, 2026
When a qualified-wallet cohort takes a directional position, that statement is recorded together with the market price at that moment. Once the market resolves, the pair is scored: was the direction right, and — the number that actually matters — was the signal closer to the outcome than the price it disagreed with?
The second question is the point. A hit rate of 61 percent against a price that already said 65 percent is a failure, and the bare hit rate would report it as a success. Both numbers are therefore shown together, the signal's Brier score next to the market's on the same cases.
A figure is published per signal strength only once at least 30 scored cases stand behind it. Below that, an accuracy number says more about the sample than about the signal, and the honest answer is that it is not measurable yet. The threshold is stated here so that a published figure can be read against it — not as a target to be reached, but as the point below which a number would mislead.
The corpus is also younger than the product. The wallet-qualification rules were corrected in August 2026, and cases recorded under the earlier cohort are excluded rather than mixed in — a track record assembled from two different definitions of the same signal would be worse than none. The corpus has been refilling since, one scheduled run at a time.
Snapshots are only taken on the markets the coverage job reaches, which are the highest-volume open ones. Whatever the figures above say, they say it about those markets — not about Polymarket as a whole. This is the same honest boundary that applies to the cohort itself, and it does not disappear once the sample is large enough.
The calibration of the AI probability estimates is a separate measurement. Every analysis has been logged with its probability since July 2026, so the data path exists; the evaluation against resolved outcomes is not built yet, and this page will say so until it is.
No figure is published yet. Every signal strength needs at least 30 scored cases behind it, and the corpus has been refilling since the cohort rules were corrected. The table shows how far along that is.
Statements are on record and waiting for their markets to settle. The earliest of them reaches its resolution date on Sep 5, 2026; scoring starts from there.
| Signal strength | Scored cases | Hit rate | Brier (signal / market) |
|---|---|---|---|
| weak | 0 / 30 | not measurable yet | not measurable yet |
| moderate | 0 / 30 | not measurable yet | not measurable yet |
| strong | 0 / 30 | not measurable yet | not measurable yet |