We use privacy-friendly analytics to understand traffic. With your consent we also load Google Analytics. No analytics cookies are set until you accept. See our privacy policy.
What the rolling-strength model forecast before each kickoff — win/draw/loss probability and expected score — graded against what actually happened. Out-of-sample (the model never saw the result). What do these mean?
The model's forecast for the upcoming fixtures — make your own picks before kickoff and see how you stack up.
Model probabilities are this site's own forecast. Picks lock at kickoff and are graded from the final result — no stakes, just bragging rights.
Every call the model made this season, graded against the result — and, more importantly, whether its probabilities held up.
Accuracy is the wrong lens for a three-way outcome: draws are rarely any side's single most-likely result, so ~25% of games are "wrong" by design. What matters is whether a stated 60% really happens ~60% of the time — that's calibration, and it's the number that actually matters.
Over 306 matches, the model lands within 97.7% of the consensus forecast's accuracy. The consensus wins narrowly — as expected, since it aggregates all available information. Source: football-data.co.uk (closing: Pinnacle/Bet365). Analysis only.
The model strongly favored Alavés at 61%, while the market was far more cautious with the home side at just 35% and rated Celta's away chances more than double the model's (33% vs 16%). Celta's superior xG (1.38 vs 0.42) and their 55th-minute winner showed the market's more balanced pricing was closer to the actual outcome.
The model heavily favored Espanyol (58%) despite near-identical xG (1.01 vs 1.08), while the market saw a much tighter contest and even leaned toward Girona (38% vs 33%). The goalless draw, with both sides posting almost equal xG, vindicated the market's more balanced pricing over the model's away-side conviction.
The model rated Athletic Club heavily (80%) while the market was more cautious (56%), likely pricing in factors a public-data model cannot capture. Athletic's dominant xG (3.21 vs 0.85) and 2-1 win, sealed by an 89th-minute goal, validated the model's stronger read.
The model favored Villarreal at 45%, while the market backed Barcelona at 55%, likely pricing in information a public-data model cannot capture. Barcelona's away win, supported by a 2.78-to-1.76 xG edge and goals in the 12th and 63rd minutes, proved the market closer.
The model rated Espanyol clear favourites at 66%, likely reflecting their edge in recent xG, while the market was more cautious at 43% and priced in a more open contest. The narrow 2.39–2.25 xG split and five goals across the match pointed to a tight game, but Espanyol's 3-2 win validated the model's stronger lean toward the home side.
The model favored Real Sociedad far more heavily (81% vs the market's 58%), likely weighing their strong underlying performance that the final 3-1 scoreline and 2.61-1.19 xG edge confirmed, while the market priced in more uncertainty toward a draw or away win. The result, decided by three first-half goals, validated the model's confidence.
The model saw a more balanced contest than the market, giving Rayo a 32% chance versus the market's 68% lean toward Barcelona, likely because it weighed recent xG while the market may have priced in information a public-data model cannot capture. The near-even xG (1.89 vs 1.88) and the 1-1 draw supported the model's more competitive read.
The market priced Barcelona as heavy favourites at 81% versus the model's more cautious 59%, likely reflecting information a public-data model cannot fully capture. Barcelona's early goals (minutes 9 and 12) and superior xG (1.93 vs 0.72) delivered the comfortable 3-1 win the market anticipated.
"closer" = the side that gave the actual result the higher probability. Pick badges: H/D/A = home / draw / away. One-line reads are AI-written strictly from each match's data (xG, red cards, goals, score) — no outside news.