Accuracy
We do not need to beat the market. We need to not be lying. This page publishes our model's own calibration results, unedited, whether they pass or fail.
No model has been fit and evaluated yet. This page will show the real numbers as soon as one has — there is nothing to show in the meantime, and nothing here is estimated or placeholder.
What these gates mean
- RPS within 5% of the market's own RPS on the same matches. This reads “not evaluated” until we hold market prices for the matches being scored — our historical corpus has none, and a model compared against nothing is not a model that beat the market.
- Expected Calibration Error under 3% for each outcome (home, draw, away).
- No probability decile disagreeing with observed results by more than chance allows at the sample size we actually have. The original fixed 3-point bound is still reported above, and still fails — see the note on why it no longer decides.
- Our published interval contains the true outcome close to its stated rate.