Model Calibration

How accurate are the ELO-based win probability estimates? Each game's pre-match ELO ratings are used to predict the outcome, then compared to what actually happened across 66222 historical games.

For the full explanation of goal difference, forfeits, card points, and the season simulator, see Model Details.

Brier Score
0.179
Lower is better · 0 = perfect · 0.25 = coin flip
Favorite Win Rate
57.0%
Games where the higher-ELO side won outright
Games Analyzed
66,222
Non-forfeit games across all historical seasons
Season-Start Regression
65%
Win probability pulled toward 50% at game 0, fading to 0% by game 6

Calibration Curve

Each dot is a bucket of games grouped by predicted win probability. The diagonal line is perfect calibration — dots above mean the model underestimates; dots below mean it overestimates. Dot size reflects number of games in that bucket.

Game Count by Predicted Probability

How many games fall into each 5% probability bucket — a well-spread distribution means the model uses its full range.

Season Outlook Calibration

At each game-week checkpoint, the simulation predicts each team's chance of promotion or relegation. How well do those predictions match what actually happened across historical seasons? Use the slider to see how calibration improves as more games are played.

Promotion Calibration
Relegation Calibration