Football ratings

Monday 19 October 2026 · times UK

The model's strongest calls

Nothing clears the bar today. The thresholds are set where the model has historically been as good as it claims — home win 70%, away win 60% — and a quiet day is a real answer, not a gap.

Goals — supporting, not headline

These are calibrated — a 65% call lands about 65% of the time. But the model separates high- from low-scoring matches far less sharply than it separates winners, so there is less information behind a goals number than a result one.

59%Getafe v Rayo Vallecano — under 2.5 goals20:00 · market 64% · this strong, 58% land (10,295 cases)
55%Parma v Torino — under 2.5 goals19:45 · market 60% · this strong, 58% land (10,295 cases)

Every fixture

Premier League
Tottenham v Coventry20:00
HomeDrawAway
Model522622
Market642116
Over 2.5Under 2.5
Model5644
Market6040
Under 4.5 goals: 84%  ·  happens 86% of the time anyway
Model is 12 points below the market on home win (pinnacle). Worth a look at why.
La Liga
Getafe v Rayo Vallecano20:00
HomeDrawAway
Model333136
Market323336
Over 2.5Under 2.5
Model4159
Market3664
Under 4.5 goals: 92%  ·  happens 86% of the time anyway
Serie A
Monza v Cagliari17:30
HomeDrawAway
Model442927
Market393131
Over 2.5Under 2.5
Model4753
Market4555
Under 4.5 goals: 88%  ·  happens 86% of the time anyway
Parma v Torino19:45
HomeDrawAway
Model302941
Market293239
Over 2.5Under 2.5
Model4555
Market4060
Under 4.5 goals: 90%  ·  happens 86% of the time anyway
La Liga 2
Celta de Vigo II v Tenerife19:30
HomeDrawAway
Model382932
Market362935
Over 2.5Under 2.5
Model4555
Market4654
Under 4.5 goals: 89%  ·  happens 86% of the time anyway
Primeira Liga
SC Braga v GIL Vicente20:15
HomeDrawAway
Model552619
Market612415
Over 2.5Under 2.5
Model5446
Market4753
Under 4.5 goals: 85%  ·  happens 86% of the time anyway
Super Lig
Trabzonspor v Beşiktaş18:00
HomeDrawAway
Model432829
Market382537
Over 2.5Under 2.5
Model5248
Market6238
Under 4.5 goals: 86%  ·  happens 86% of the time anyway

How the calls did

✓Real Madrid 4–1 SevillaReal Madrid home win 75%
✕Cambuur 2–2 AZ AlkmaarAZ Alkmaar away win 71%
✓Cambuur 2–2 AZ Alkmaar over 2.5 goals 67%
✓Bayer Leverkusen 2–1 SC Freiburg over 2.5 goals 67%
✓Borussia Mönchengladbach 1–2 1899 Hoffenheim over 2.5 goals 66%
✓Groningen 2–5 Ajax over 2.5 goals 64%
✓PEC Zwolle 4–1 GO Ahead Eagles over 2.5 goals 63%
✓Real Madrid 4–1 Sevilla over 2.5 goals 62%
✓Celtic 7–1 Heart Of Midlothian over 2.5 goals 61%
✓Burgos 1–0 Sabadell under 2.5 goals 60%
✓Lorient 1–0 Monaco under 2.5 goals 58%
✓Osasuna 2–0 Racing Santander under 2.5 goals 56%

11 of 12 settled since yesterday · 313 of 469 since the model went live.

Every match, for the record

The model rates the whole card, not just the calls, and learns from all of it. Judged as probabilities rather than picks — a 38% favourite losing is the model working, not failing.

·Resulttop pick right 21/33, we claimed 51% · 478/957 all-time
what actually happenedwe gave it 42% on average
·Over/under 2.5top pick right 21/33, we claimed 58% · 552/957 all-time
what actually happenedwe gave it 55% on average

A well-calibrated model's average claim matches how often it happens. That is the number Rule 9 gates on, and it is tracked for results and goals alike whether or not a fixture ever produced a call.

Changed after we rated them

🕒SC Braga v GIL Vicentemoved to Mon 19 Oct · 20:15

What these numbers are. The model's own estimate of how likely each result is, from form, Elo, rest, shots and xG as they stood at 06:00 UK on match day. One bivariate Poisson produces all of them, so the result and the goals figures share one underlying view of the match. Each published figure is then calibrated against what actually happened in past matches, one market at a time — so they agree closely but are not forced to agree exactly. Market figures are the bookmaker's prices with their margin removed, so the two are comparable.

What they are not. Across 57,633 out-of-sample fixtures the market separated matches better than the model in every league, every season, every outcome and every confidence band — 24 slices, no exceptions, re-tested in August 2026 on the current model. This is a second opinion, not an edge. Where it earns its keep is honesty about its own confidence: when it says 74%, that happens about 74% of the time.

Why a call appears. Thresholds — home 70% · away 60% · over 60% · under 55% — sit where the model has historically been as good as its claim. Raising them would surface fewer calls and be less accurate about them, because its confidence above those levels is inflated everywhere except home wins. Draws are never flagged: the highest one ever rated was 62%.

7 rated · 0 not rated · model goals_v7_cal · built 2026-10-19 07:02 UK