Football ratings

Monday 10 August 2026 · times UK

The model's strongest calls

58%Santa Clara v Nacional — under 2.5 goals20:15 · market 56% · this strong, 57% land (7,790 cases)
£10 on all 1 as one multiple returns about £17 if every one lands. The model puts that at 58%, the market at 56%, and the price needs 60% just to break even.Bookmaker margin compounds with every leg, which is why the break-even sits above the market's own view. A same-match bet builder pays less than this — the bookmaker prices in that the legs move together. Our chance is a straight multiplication and is if anything a little low: where two legs of one match both cleared, they landed together 55.5% of the time against a multiplied 53.4%.

Every fixture

primeira liga
Santa Clara v Nacional20:15
HomeDrawAway
Model463024
Market512821
Over 2.5Under 2.5
Model4258
Market4456
Under 4.5 goals: 91%  ·  happens 86% of the time anyway

How the calls did

✓FC Porto 2–0 AlvercaFC Porto home win 83%
✕Heerenveen 1–0 Twente over 2.5 goals 71%
✕FC Porto 2–0 Alverca over 2.5 goals 65%
✓PEC Zwolle 0–2 AjaxAjax away win 64%
✓Heart Of Midlothian 4–0 Dundee Utd over 2.5 goals 64%
✓Kilmarnock 1–5 CelticCeltic away win 63%
✕PEC Zwolle 0–2 Ajax over 2.5 goals 61%

4 of 7 settled since yesterday · 12 of 20 since the model went live — far too few to read anything into yet.

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 6/11, we claimed 60% · 12/28 all-time
what actually happenedwe gave it 45% on average
·Over/under 2.5top pick right 6/11, we claimed 58% · 15/28 all-time
what actually happenedwe gave it 49% 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

🕒Heart Of Midlothian v Dundee Utdmoved to Sun 9 Aug · 15:00
🕒Motherwell v Falkirkmoved to Sun 9 Aug · 15:00

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 come off a single scoreline distribution and cannot contradict each other. 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%.

1 rated · 0 not rated · model goals_v3_cal · built 2026-08-10 07:01 UK