Data collection and processing
Data Collection — Fixtures, scores, odds, weather, injuries, lineups, and team statistics from multiple sources.
We turn football data into probabilities using machine learning, advanced statistics and a transparent methodology.
A transparent approach combining statistics, advanced models and continuous evaluation.
Data Collection — Fixtures, scores, odds, weather, injuries, lineups, and team statistics from multiple sources.
Elo Ratings — Dynamic team strength ratings updated after every match, with home advantage correction.
Feature Engineering — Features grouped by statistical topic including form, goals, head-to-head, weather, squad quality, and more.
Dixon-Coles Model — A classical statistical model that estimates team attack and defence parameters.
XGBoost Models — Specialist gradient-boosted models for 14 markets, trained with ablation-tested feature selection.
Predictions — Probabilistic forecasts generated daily for upcoming matches.
A 60% prediction means the model estimates the outcome will occur in roughly 6 out of 10 similar situations. It does not guarantee the result of an individual match.
Compare predictions against actual results and review calibration by probability range on our performance page.
View performance →Each percentage is an estimated probability. Explore the data and models behind our analysis.
Learn more →Compare match results with the predictions published before kickoff.
Learn more →Explore accuracy statistics, calibration and the history of our models.
Learn more →Models analyse historical data, recent form, Elo ratings and other statistics available before each match.
No. A probability is an estimate, not a guarantee. Football is unpredictable and any result can occur.
No. We are a football statistical analysis platform and do not accept bets. Use the information responsibly.