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 380 matches, the model lands within 97.4% 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 gave Liverpool only 49% versus the market's 73%, likely because its rating of Burnley kept the home and draw outcomes more plausible than the market priced in. The result favoured the market, though the 0.14 to 2.64 xG gap and the 95th-minute winner show Liverpool dominated a game that stayed goalless until the very end.
The model saw only a modest xG edge for Liverpool (2.22 vs 1.67) and priced the home side at 52%, while the market was far more confident at 75%, likely reflecting information a public-data model cannot capture. Liverpool's 4-2 win, aided by two late goals at 88 and 94 minutes, vindicated the market's stronger lean.
The model favored Crystal Palace (50%) based on their superior xG (2.94 vs 2.08), while the market priced Liverpool as favorites (51%), likely factoring in information a public-data model cannot capture. Crystal Palace's 2-1 win, sealed by a 97th-minute goal, matched their xG edge and vindicated the model.
The model was more cautious on Liverpool (42% vs the market's 66%), likely reflecting a tighter recent xG picture, while the market priced in stronger home-win expectations. Despite dominating xG 2.12 to 0.68, Liverpool failed to convert and the goalless draw fell closer to the model's more balanced read.
The model favoured Forest at 52% while the market leaned toward Newcastle at 45%, a gap likely reflecting information a public-data model cannot see. The result was a draw shaped by late goals at the 74th and 88th minutes, with Newcastle's edge in xG (1.58 vs 1.15) aligning more with the market's read.
"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.