← Predictions

Model v5 · Updated 23 Sep 2026

Our track record

Every forecast checked against the final 90-minute result, by league, competition and season. Nothing is hand-picked: every graded match counts.

The backtest runs model v5 over past seasons from Jun 2008 to Sep 2026. For every match it only uses information that existed before kickoff, exactly as if the forecast had been published that day. The model was tuned on matches from January 2025 to April 2026; every other month is an out-of-sample test.

Graded matches: 701
Match result hit rate: 46.8%
Brier score, 1X2: 0.624
Over/Under 2.5 hit rate: 65.0%
Both teams to score hit rate: 58.3%
Exact score hit rate: 18.8%

Every market

Results by market

Market	Matches	Hit rate	Brier score	Log loss
Match result (1X2)	701	46.8%	0.6243	1.0366
Over/Under 1.5 goals	701	58.1%	0.2491	0.6927
Over/Under 2.5 goals	701	65.0%	0.2288	0.6500
Over/Under 3.5 goals	701	84.7%	0.1307	0.4313
Both teams to score	701	58.3%	0.2437	0.6804
Exact score	701	18.8%	-	2.5684

Hit rate counts the outcome we marked as most likely. Brier score and log loss grade the full probabilities, so a confident miss costs more than a cautious one.

Season by season

Match result hit rate over time

Show as a table

Period	Matches	Hit rate	Brier score	Over/Under 2.5	Both teams to score
2025	173	48.6%	0.6092	64.7%	61.3%
2024	126	50.0%	0.6091	62.7%	56.3%
2023	126	49.2%	0.6383	67.5%	61.9%
2022	126	46.0%	0.6321	69.8%	60.3%
2021	126	40.5%	0.6313	61.9%	54.0%
2020	24	41.7%	0.6625	58.3%	41.7%

Paler bars have fewer than 100 matches and move a lot by chance. The earliest seasons have few matches and little history for the model to learn from, so they read lower.

Do the percentages mean what they say?

Calibration

Each dot groups home, draw and away probabilities of similar size. Dots close to the diagonal mean that when we say 60%, it happens about 60% of the time.

Where it works best

By competition group

Group	Matches	Hit rate	Brier score
Major leagues	150,938	49.5%	0.607
Other men's leagues	1,019,020	51.0%	0.597
Cups	109,572	55.1%	0.566
Women's football	35,577	62.6%	0.494
Youth & reserves	52,892	53.6%	0.583
National teams	17,077	60.2%	0.516
Friendlies	44,695	54.7%	0.577

Groups follow the period and model filters. Mismatched competitions such as cups and women's leagues are easier to call than balanced top divisions.

1064 leagues with at least 30 matches

By league

League	Matches	Hit rate	Brier score	Over/Under 2.5	Both teams to score
Friendlies ClubsWorld	39,552	55.0%	0.576	62.6%	56.5%
FA CupEngland	10,696	48.2%	0.619	62.4%	57.6%
ChampionshipEngland	8,450	45.7%	0.635	53.5%	52.5%
League TwoEngland	8,318	44.3%	0.647	53.7%	51.9%
League OneEngland	8,285	46.9%	0.631	52.5%	52.5%
National LeagueEngland	8,200	47.1%	0.625	54.0%	54.7%
2. LigTurkey	7,387	52.0%	0.586	55.7%	51.0%
Primera NacionalArgentina	7,368	43.6%	0.642	66.6%	58.8%
Torneo Federal AArgentina	7,300	48.7%	0.614	62.0%	56.9%
J2 LeagueJapan	6,408	44.7%	0.641	55.3%	52.2%
Serie ABrazil	6,304	49.6%	0.614	56.2%	52.1%
Major League SoccerUnited States	6,202	50.0%	0.617	56.8%	56.9%
La LigaSpain	6,149	52.7%	0.582	57.2%	52.3%
Serie AItaly	6,131	53.4%	0.584	54.6%	53.8%
Premier LeagueEngland	6,120	53.1%	0.583	55.1%	54.6%
Ligue 2France	5,900	43.5%	0.645	56.8%	52.7%
Ligue 1France	5,813	49.7%	0.606	57.1%	52.8%
3. LigaGermany	5,761	45.9%	0.639	53.7%	54.2%
U19 BundesligaGermany	5,713	56.6%	0.558	69.7%	61.5%
Primera CArgentina	5,687	41.7%	0.652	63.7%	57.2%
Serie BBrazil	5,610	47.0%	0.625	58.9%	53.0%
Serie DBrazil	5,511	48.2%	0.614	58.0%	53.8%
Primera B MetropolitanaArgentina	5,470	40.1%	0.655	65.6%	57.6%
USL ChampionshipUnited States	5,298	48.3%	0.618	55.8%	55.0%
Serie C - Girone AItaly	5,278	43.0%	0.644	59.1%	53.4%
USL League TwoUnited States	5,159	59.2%	0.537	68.1%	59.9%
Primera DivisionGuatemala	5,110	62.8%	0.520	56.2%	54.1%
EredivisieNetherlands	5,023	53.9%	0.572	60.0%	57.6%
Süper LigTurkey	5,001	50.8%	0.602	53.9%	53.0%
FA TrophyEngland	4,985	47.3%	0.630	62.8%	59.9%
1. LigTurkey	4,972	47.5%	0.614	53.8%	51.9%
BundesligaGermany	4,948	51.4%	0.597	58.7%	56.7%
Ligue 2Algeria	4,915	56.1%	0.572	63.3%	58.0%
Jupiler Pro LeagueBelgium	4,818	51.2%	0.598	55.1%	54.2%
J1 LeagueJapan	4,793	47.0%	0.630	55.5%	55.5%
Liga Profesional ArgentinaArgentina	4,784	44.0%	0.639	61.7%	56.2%
Segunda DivisiónSpain	4,743	44.7%	0.638	60.4%	53.3%
Primeira LigaPortugal	4,705	54.5%	0.560	56.0%	52.8%
2. BundesligaGermany	4,654	44.7%	0.645	55.6%	55.4%
Ligue 3France	4,606	41.8%	0.652	56.8%	51.3%
Primera DivisiónVenezuela	4,601	47.4%	0.624	55.0%	51.3%
NB IIHungary	4,582	47.4%	0.623	55.4%	54.8%
Liga AlefIsrael	4,558	47.4%	0.625	54.7%	52.3%
UEFA Europa LeagueWorld	4,454	50.2%	0.621	54.0%	52.8%
Primera AColombia	4,446	47.2%	0.621	59.2%	52.9%
3. Lig - Group 3Turkey	4,344	48.3%	0.615	59.0%	53.8%
3. Lig - Group 1Turkey	4,340	48.4%	0.607	57.2%	52.2%
2. DivisionBelarus	4,334	64.0%	0.479	72.5%	55.5%
3. Lig - Group 2Turkey	4,322	46.6%	0.625	55.3%	52.7%
Premier LeagueEgypt	4,236	46.3%	0.614	60.3%	52.8%

Show all 1064 leagues

With a few hundred matches, a league's hit rate can move by several points from luck alone. Compare leagues with large samples.

Reading the numbers

How we measure accuracy

Hit rate: How often the outcome we marked as most likely happened. Simple, but it ignores how confident the forecast was.
Brier score: The squared gap between our probabilities and what happened. 0 is perfect; for match results, spreading a third on each outcome scores 0.667.
Log loss: Punishes confident mistakes harder than the Brier score. Lower is better.
Backtest: The current model run over past matches with only the information available before each kickoff. It shows how the model behaves across many seasons, but it is a simulation, not a record of published forecasts.
Published forecasts: Forecasts we actually showed, stored with a timestamp before kickoff and never edited afterwards.
Result used: The score after 90 minutes plus stoppage time. Extra time and penalties do not count.

Forecast archive → Methodology →

Forecasts are probabilities, not certainties. Past accuracy does not guarantee future results.