Every modern bookmaker price is built on data, which means a punter working from vibes is bringing an opinion to a maths fight. The good news is that an unusual amount of serious football and sports data is free. The honest news is that free data narrows the gap with the bookmaker without closing it — and misread data is worse than none. Here is what is worth using and how not to fool yourself with it.
Official sources first
League and governing-body sites publish the ground truth: fixtures, results, team news and official statistics. The Premier League, UEFA and the GAA all publish extensive stats sections, the League of Ireland publishes results and tables, and Horse Racing Ireland and the Irish Racing Regulatory Board cover declarations, going and form in racing. Official sources settle arguments other sites start — when a stats aggregator and a league site disagree, the league site wins.
Free advanced stats: xG and friends
Expected goals (xG) measures chance quality: each shot is assigned the probability a similar shot historically became a goal, and the sum tells you what a team “should” have scored. Public sources such as FBref and Understat publish xG for the big European leagues free of charge, along with shot maps, possession-adjusted numbers and season-level tables. xG’s honest use is as a corrective: a team winning matches while being consistently out-created is usually flattered by its results, and league tables lie over small samples in ways xG tables lie less.
How to read stats without overfitting
The classic mistake is treating a pattern in a small sample as a law of nature. “Six of their last seven matches went over 2.5 goals” is seven coin flips dressed up as insight — with so few matches, streaks like that arise constantly by chance. Prefer season-long rates to last-five form, prefer underlying numbers (shots, xG) to outcomes, and distrust any stat with a suspiciously specific filter (“on Tuesdays, away, after a red card”). The more conditions a trend needs, the more likely it was found by searching rather than discovered by understanding.
The benchmark is the price, not the pundit
A stat only matters if the market has missed it — and the market has already read the same free sources. Convert the odds into an implied probability, strip the margin, and ask whether your data genuinely justifies a different number. Most of the time it will not, because xG tables and form figures are priced in within minutes of publication. The realistic win from free data is avoiding clearly bad bets, not discovering golden ones.
Honest conclusions
Use official sources for facts, public xG for context, season-sized samples for judgement, and the market price as your benchmark. Free data makes you a better-informed punter; it does not make betting profitable, because the margin still has to be beaten on top of the market’s own accuracy. Bet what the entertainment is worth, not what the spreadsheet seems to promise. 18+ — Gambling Awareness Trust, 1800 753 753, gamblingcare.ie.
Frequently Asked Questions
Official league and governing-body sites for ground truth (Premier League, UEFA, the GAA, Horse Racing Ireland for racing), and public xG sources such as FBref and Understat for advanced metrics on the big European leagues. When an aggregator and a league site disagree, the league site wins.
Expected goals measures chance quality: each shot is assigned the probability that a similar shot historically became a goal. Its best use is as a corrective — spotting teams flattered or wronged by recent results — over season-sized samples, not five-match streaks.
Realistically no. The market reads the same free sources and prices them in within minutes, and the margin must be beaten on top of the market’s own accuracy. Free data helps you avoid clearly bad bets rather than discover golden ones.