AI football predictions: calibrated probabilities, published accuracy, no tips sold

Which fixtures is the model tracking right now?

4,003 fixtures are scheduled in the next seven days and 14,927 in the next thirty, across 646 active competitions. The table below lists the next six kick-offs among the major domestic leagues and continental cups.

Times given in Coordinated Universal Time (UTC). The 1X2 probabilities for each fixture are computed in the application; publishing them on a dedicated page per match is being rolled out.

How does Footlab AI predict a match?

With a Poisson model whose scoring intensities are fed by seven measurable inputs, all read from the database at computation time. When odds are published for the match, the displayed 1X2 probability combines this estimate with the market consensus, which carries most of the weight (85 to 95%, the model always keeping its share); no odds are displayed.

The computation is pure arithmetic: it calls no external service and consumes no AI credit.

How accurate are the predictions, measured?

54.7% correct 1X2 calls on the 1,522 cards actually served, on par with the bookmakers' favourite (55.8% on the 1,251 of those matches with known odds). The model alone, without odds, scored 53.9% on unseen matches from the top leagues and European cups (test of 16 July 2026), but 50.5% across all competitions, against 52.8% for the bookmakers: that is why the card combines the model with the odds (measured 5 October 2026).

Every served sheet is logged before kick-off then compared to the official result. Nothing is cherry-picked afterwards, nothing is deleted: that is what makes the figure contestable.

Why is 53% a good number and 87% a red flag?

Because the starting point is not zero. Over 29,415 matches in the top 20 domestic leagues, seasons 2022 to 2025, the hosting team wins 44.3% of the time: calling “home” on every fixture, with no model whatsoever, already returns 44.3% accuracy. On the matches for which Footlab AI served a card, “always home” scores 44.9% and Footlab AI's cards 54.7%: the real gain is 9.8 points, not 21.4 points above chance. That is also the bookmakers' level: their favourite scores 55.8% on those matches.

The arithmetic deserves to be spelled out, because it is checkable. Draws are 26.3% of the measured fixtures. A model that never designates the draw as its most likely outcome — which is the case for almost every published Poisson model — is therefore mechanically capped at 73.7%. Claiming 87% requires finding 13.3 extra points inside the draws, meaning calling more than half of them correctly. No published work comes close. An accuracy claim above 60% with no sample, no date and no protocol should be read as a marketing statement, not as a measurement.

Is that law stable over time?

Five consecutive seasons, same 20 leagues. The maximum spread on the home-win share is 1.9 points: the 44.3% benchmark is not a windowing artefact.

What does Footlab AI not do?

Footlab AI publishes no odds, computes no betting “value”, recommends no stake and displays no link to an operator. Those four absences are deliberate and permanent.

The reason is easy to verify: a site paid by bookmaker affiliation has a direct interest in the reader placing a bet, therefore in overstating its own reliability. Publishing a failure rate hurts it commercially. Footlab AI has no such conflict of interest, and that is exactly what makes its figure readable.

Which competitions are covered?

1,247 competitions are recorded in the Footlab AI database — 787 leagues and 460 cups, spread over 171 countries. The model produces a card for every scheduled fixture; when a team's history is thin (youth, amateur divisions, friendlies), the card flags it and its prediction should be read with caution.

Frequently asked questions

Does Footlab AI give betting tips?

No. Footlab AI publishes calibrated probabilities for each outcome of a fixture, with no odds, no recommended stake and no link to a betting operator. No bookmaker affiliation exists.

What does 54.7% accuracy mean?

It means that in 54.7% of the matches for which Footlab AI served a card, the outcome shown as most likely actually happened (1,522 matches evaluated as of 5 October 2026). The right comparison is the bookmakers' favourite, which scores 55.8% on the same matches: Footlab AI is at that level, not above it.

Why is a claimed accuracy of 87% suspicious?

Because draws are 26.3% of fixtures and almost never the most likely outcome of a match. A model that never designates the draw is therefore capped at 73.7%. Reaching 87% would require correctly predicting more than half of all drawn matches, which no published work achieves.

What is the best accuracy achievable with no model at all?

44.3%, by calling the home win on every fixture. Measured over 29,415 matches in the top 20 domestic leagues, seasons 2022 to 2025. That is the real starting line, not the 33.3% of a three-way coin toss.

Do the predictions use bookmaker odds?

The model does not: it is fed by pool-anchored Elo, country anchors, the probable starting eleven, player experience and ratings, absences, home advantage and cross-league damping. Then, when odds are published for the match, the displayed 1X2 probability combines that estimate with the market consensus. No odds are displayed, and no link leads to a betting operator.