Football statistics

FootIQ's predictions are only as good as the data behind them. This page lists the statistics we use and where they come from.

Data source

Fixtures, results, match statistics, standings, injuries, lineups, live events and bookmaker odds come from API-Football, a commercial football data provider. We store the finished matches of the current and previous season for every covered competition and refresh them regularly.

Statistics used by the models

Goals scored and conceded

The basis of every model, separately for home and away matches.

Half-time scores

Used by the time-split model to estimate how a team's goals are distributed across the two halves.

Expected goals (xG)

Used by the empirical-Bayes model where the provider has it; xG is less noisy than goals over short periods.

Match age

Older matches receive less weight; the weight halves after about one year.

Context statistics

Standings, head-to-head results, recent form, rest days, injuries and suspensions and lineups are shown in the analysis and are part of what the AI layer reviews. They are not added to the models as extra factors, to keep the models simple and testable.

Live statistics

During matches FootIQ shows live scores, events and statistics (for example shots and possession). Live scores and statistics are free for every account.

Known limitations

• Not every competition has xG data; where it is missing the models rely on goals.
• Newly promoted teams or teams with few matches have less reliable estimates; the data-quality gate handles this.
• Data providers occasionally correct results after the fact; a correction of a settled pick is recorded publicly on the track record.