Algorithmic football predictions
Under the lights, the numbers decide
Football predictions from a transparent statistical model — 1X2, Over/Under 2.5 and both teams to score — with a public, append-only track record. No hunches, no hindsight edits.
- 1X2 hit rate · 189 settled
- 51%
- Predictions recorded
- 410
- Leagues covered
- 11
Nice79%Over 1.5 goalsFair odds 1.27
Le Mans
Track record — in the open
Full history →
How it works
Full methodology →- 01
Data in
Results, fixtures and market odds for our covered leagues, refreshed three times a day.
- 02
The model prices every match
A Dixon-Coles statistical model turns team strength and form into 1X2, Over/Under 2.5 and BTTS probabilities.
- 03
Predictions locked before kickoff
Every call is recorded with its model version and timestamp before the match starts — after that, nothing can be edited.
- 04
Every call graded in public
Predictions are settled after full time and stay on the record forever — hits and misses alike.
From the blog
All articles →First team to score betting: the formula that prices the market
First team to score betting explained: the competing-Poisson formula that prices it, a worked example, and why the margin hides in the No Goal price.
TransparencyHow to compare football prediction sites: a fair test in five steps
Most rankings compare numbers that were never comparable. How to compare football prediction sites properly: one shared fixture set, one market, one score.
ModelMonte Carlo simulation in football: predicting a whole season
Monte Carlo simulation in football, explained: how thousands of simulated seasons turn match probabilities into title and relegation odds — and their limits.