Predicted scorelines and win probabilities for football's biggest matches — powered by expected goals, current form and head-to-head record. Clear, honest, and built to compare.
★ Match of the day
The model makes Rayo Vallecano 37% to win, the draw 29% and Espanyol 34% — fair odds 2.70 · 3.45 · 2.94 — with 2-1 the most likely scoreline, at low confidence.
The day's most important matches across every competition we cover — ranked by our model. See all matches — recent results and the days ahead — on the fixtures page.
The 15 most important matches on 15 September 2026, across every competition we cover, ranked by the model.
Auto-generated previews and reviews, written from the model's own numbers — no punditry, just what the data says about the biggest matches of the next seven days.
The model makes it 36% Bournemouth, 28% draw, 36% Liverpool — most likely 2-1 (fair odds 2.78 on Bournemouth).
20 September 2026 · Premier LeagueThe model makes it 40% AS Roma, 26% draw, 34% Inter — most likely 2-1 (fair odds 2.50 on AS Roma).
19 September 2026 · Serie AThe model makes it 51% Brentford, 24% draw, 25% Chelsea — most likely 2-1 (fair odds 1.96 on Brentford).
18 September 2026 · Premier LeagueThe model makes it 33% Fulham, 28% draw, 39% Manchester United — most likely 1-2 (fair odds 2.56 on Manchester United).
20 September 2026 · Premier LeagueThe model makes it 38% Brighton, 31% draw, 31% Arsenal — most likely 1-0 (fair odds 2.63 on Brighton).
19 September 2026 · Premier LeagueThe model makes it 27% Sevilla, 27% draw, 46% Barcelona — most likely 1-2 (fair odds 2.17 on Barcelona).
19 September 2026 · La Liga6 of the most important matches played on 14 September 2026, with the final score and the call the model published before kick-off. Across all 42 it called 24 correct — 57%, against 47% over every prediction on record.
Every prediction is logged before kickoff and checked against the real result — a transparent, automatically-updated record of how the model actually does.
The hit rate is only half the question. The full track record scores every settled prediction against an uninformed baseline and shows, band by band, whether a 70% from this model really happens 70 times in 100 — including where it does not.