Skip to content

EPL model accuracy

Premier League

Every final EPL game, forecast with only what was known the day before and then scored. Lower Brier and log loss are better; a coin flip scores 0.250 Brier in two-outcome leagues, and a blind three-way guess 0.333. All leagues

Scorecard

Accuracy of each model on final games
ModelGamesBrierLog lossAccuracySkillCalib. error
Game simulator (backtest)90small sample0.3251.07241.1%+2.9%6.2%
Elo90small sample0.3311.09338.9%+1.3%6.8%
Base rate90small sample0.3351.10540.0%—4.0%

Calibration

When a model gives the home side a 60% chance, does it win about 60% of the time? Dots on the dashed line mean yes; whiskers show the 90% range the observed rate could land in by chance.

Brier score by month

Early-season forecasts lean on last season's regressed ratings, so expect them to improve as games accumulate.

Head to head

Game simulator (backtest) vs Elo

90 games both forecast

Brier
0.325 vs 0.331
Log loss
1.072 vs 1.093
Accuracy
41.1% vs 38.9%

Game simulator (backtest) has the lower Brier score here, but on too few games to call.

Market consensus isn't connected: it needs a licensed odds source, which an admin can import. Until then the base rate and Elo are the benchmarks.

How this is measured

The backtest replays the game simulator for every final game from the efficiency ratings and Elo as they stood before that date, the league's scoring and home advantage so far, rest and travel, the venue and the game's weather. Injuries and lineups of past games aren't on record, so the replay is team level; the published forecasts, which include them, are scored separately as they accumulate.

Brier = ½ Σ (probability − outcome)² over home, draw, away. Log loss = −ln(probability given to the result), with probabilities clipped at 0.5%. Full details are in the model methodology document.

Data as of Source: SportsAPI Pro

Cookies

Cookies keep you signed in. Optional ones remember your settings. No ads or tracking. Cookie policy