Model accuracy

Every prediction this model publishes is written to a log before kick-off and never touched again. This page scores all of them against what actually happened. So far 938 predictions have been made before kick-off and have a final result.

The model called the correct result — home win, draw or away win — in 456 of them, 48.6%, and named the exact scoreline 88 times, 9.4%.

Its Brier score over the three outcomes is 0.638 (lower is better). For scale, a model that learned nothing and gave every match an even 33.3% across the three results would score 0.667. The model is ahead of that baseline.

Predictions logged between 19 July 2026 and 18 August 2026.

Two honest limits. This is a young sample, so every figure here will move; and the hit rate on its own flatters any model that simply backs the favourite, which is why the calibration table below matters more than the headline.

Is the model well calibrated?

Accuracy on its own is a weak test: back the favourite every time and any football model lands near the same hit rate. The harder question is whether the numbers mean what they say — when the model gives an outcome a 60% chance, does it happen about 60 times in 100? Each match contributes three forecasts (home, draw, away), so the 938 settled matches give 2814 probability-and-outcome pairs.

Read it row by row: the closer the last two columns are, the more literally the percentages on this site can be taken.

By competition

Only competitions with at least 10 settled predictions are listed — below that a hit rate is noise, and printing it would invite a conclusion the data does not support.

The last 15 settled predictions

How the predictions are made

A Poisson / Dixon-Coles model built on each team's rolling attack and defence strength, estimated from recent expected-goals data and adjusted for the strength of the opposition and of the league. There are no bookmaker odds anywhere in it. The full method is on the About & Methodology page.