Every selection on this page carries two numbers: the chance we give it, and the chance the bookmaker’s price implies. Where ours is higher, the odds are longer than the record says they should be.
Odds of 10/11 mean the bookmaker thinks it lands 52% of the time. Evens is 50%. 11/10 is 48%. That is all a price is — a percentage with a margin added.
We work out the same percentage from the records instead. How often a player has been fouled three times in his last six matches. How often two sides have both scored across their last 20. Then we put the two numbers side by side, and where ours is higher by enough to matter, it goes on this page.
Where Our Model Disagrees with the Market
Every selection we priced, not only the ones we are backing.
How the Model Works
Every figure starts from a count rather than an opinion.
For a player line we take his last 6 matches and ask how often he cleared it, then check the rate per 90 minutes behind that so a single big return does not flatter an average. A player under 450 minutes does not appear at all — 4 goals in 266 minutes is a hot streak, not a rate.
For a team market we use both sides’ last 20 matches. Corners, cards, goals at both ends and total goals all come from counting what has actually happened rather than from a rating.
Where a side has changed division, we weight by level. A promoted team’s last 20 include matches against weaker opposition, and taking those figures at face value makes them look better than they are. Four sides in this division came up from LaLiga 2, and their figures are discounted accordingly — which is why some fixtures here have fewer selections than the raw numbers would suggest.
Where a rate comes off a small sample, we pull it towards a baseline before quoting it. A forward with three goals in four appearances is regressed towards what a forward normally does, because four appearances is not enough to know otherwise.
What We Are Not Pricing in La Liga, and Why
Card markets. Not one card selection appears on this page, and the reason is worth stating. Our card figures come from each side’s last 20 matches, and we have never applied the level adjustment to them that we apply to goals. A side promoted from LaLiga 2 brings its second-division card rate with it, which inflates the fixture by more than a card a game.
We found this by pricing a fixture at 72% where the market said 33% — a 39-point gap with no honest explanation except that our number was wrong. Until the adjustment is built, card markets involving a promoted side stay out.
Match results in lopsided fixtures. A 20-match model counts what has happened and cannot see squad quality. It reads corners and fouls well and results badly, so where the market makes a side a heavy favourite we publish the price rather than argue with it.
Referees, for now. Only one fixture in this round has an appointed official. La Liga referees range from 2.95 to 5.53 yellows a game — a spread more than twice as wide as the English officials we have measured — so an unappointed referee is a much bigger unknown here than at home.
Our Record
We track every selection on this page and settle it against the official match data, losses included.
The tracker started on 1 September 2026 and La Liga is our newest competition, so the sample here is a single round. We would rather tell you that than quote a strike rate off a handful of bets.
The method behind it has been running longer, and it has already changed once because of what the results showed. The 20-point ceiling on the edge table came from settled bets rather than from theory, and it arrived after four selections had gone out above it.
La Liga Results
No selections settled in the last 7 days. The figures above cover the full record.
More Predictions
- Value bets across every competition we cover
- Premier League best bets for every round
- Championship best bets fixture by fixture
- Both teams to score tips this weekend
18+ only. Please gamble responsibly. GambleAware.org