Results
Most recent 60 matches, newest first
What xG diff tells you
xG is the quality of chances each side created. Compare it with the score:
a team that won 2-0 while losing the xG 0.7 to 2.1 was outplayed and got the result anyway.
Over a season, results drift back toward xG — which is why the model is fitted on expected
goals rather than goals.
| Date | Match | Score | xG | xG diff | Model pick | Called | |
|---|---|---|---|---|---|---|---|
| 19 Oct 26 |
 Tottenham Hotspur v
 Coventry |
3–1 | 0.97–1.14 | -0.17 | called it | analysis → | |
| 18 Oct 26 |
 Real Madrid v
 Sevilla |
4–1 | 2.13–1.32 | +0.81 | called it | analysis → | |
| 18 Oct 26 |
 Osasuna v
 Racing Santander |
2–0 | 3.08–0.22 | +2.86 | called it | analysis → | |
| 18 Oct 26 |
 Malaga v
 Real Sociedad |
2–5 | 1.92–2.57 | -0.65 | missed | analysis → | |
| 18 Oct 26 |
 Celta Vigo v
 Alaves |
2–1 | 0.76–1.18 | -0.42 | called it | analysis → | |
| 18 Oct 26 |
Nottingham Forest v
 Arsenal |
0–1 | 0.75–0.92 | -0.17 | called it | analysis → | |
| 18 Oct 26 |
 Leeds United v
 Manchester United |
1–1 | 1.41–1.08 | +0.33 | missed | analysis → | |
| 18 Oct 26 |
 Brighton and Hove Albion v
 Crystal Palace |
5–0 | 2.25–0.63 | +1.62 | called it | analysis → | |
| 18 Oct 26 |
 Bournemouth v
 Sunderland |
2–0 | 4.39–0.09 | +4.30 | called it | analysis → | |
| 17 Oct 26 |
 Venezia v
 Napoli |
1–0 | 0.68–2.56 | -1.88 | missed | analysis → | |
| 17 Oct 26 |
 Roma v
 Genoa |
2–0 | 1.41–0.51 | +0.90 | called it | analysis → | |
| 17 Oct 26 |
 Bologna v
 Internazionale |
1–3 | 1.77–3.60 | -1.83 | called it | analysis → | |
| 17 Oct 26 |
 Werder Bremen v
 Paderborn |
2–2 | 2.67–1.37 | +1.30 | missed | analysis → | |
| 17 Oct 26 |
 Union Berlin v
 Borussia Dortmund |
0–0 | 0.94–0.99 | -0.05 | missed | analysis → | |
| 17 Oct 26 |
 Schalke 04 v
 Mainz 05 |
0–1 | 0.41–1.53 | -1.12 | called it | analysis → | |
| 17 Oct 26 |
 Hamburger SV v
 VfB Stuttgart |
1–0 | 1.67–2.64 | -0.97 | missed | analysis → | |
| 17 Oct 26 |
 Elversberg v
 Augsburg |
4–3 | 4.00–3.74 | +0.26 | called it | analysis → | |
| 17 Oct 26 |
 Bayern Munich v
 RasenBallsport Leipzig |
6–1 | 4.90–0.70 | +4.20 | called it | analysis → | |
| 17 Oct 26 |
 Troyes v
 Lens |
1–3 | 1.47–1.85 | -0.38 | called it | analysis → | |
| 17 Oct 26 |
 Strasbourg v
 Paris Saint Germain |
1–3 | 1.31–2.36 | -1.05 | called it | analysis → | |
| 17 Oct 26 |
 Lille v
 Brest |
1–1 | 2.05–0.79 | +1.26 | missed | analysis → | |
| 17 Oct 26 |
 Newcastle United v
 Aston Villa |
0–1 | 0.92–2.08 | -1.16 | missed | analysis → | |
| 17 Oct 26 |
 Manchester City v
 Ipswich |
3–2 | 4.01–1.37 | +2.64 | called it | analysis → | |
| 17 Oct 26 |
 Fulham v
 Hull |
1–0 | 2.27–0.13 | +2.14 | called it | analysis → | |
| 17 Oct 26 |
 Everton v
 Chelsea |
0–2 | 0.57–1.60 | -1.03 | called it | analysis → | |
| 17 Oct 26 |
 Brentford v
 Liverpool |
1–2 | 0.86–1.55 | -0.69 | called it | analysis → | |
| 12 Oct 26 |
Coventry v
Newcastle United |
4–1 | 1.66–2.22 | -0.56 | missed | analysis → | |
| 11 Oct 26 |
 Sassuolo v
 AC Milan |
1–1 | 1.34–1.87 | -0.53 | missed | analysis → | |
| 11 Oct 26 |
 Lecce v
Bologna |
1–2 | 1.21–2.00 | -0.79 | called it | analysis → | |
| 11 Oct 26 |
 Lazio v
 Monza |
3–1 | 2.22–1.30 | +0.92 | called it | analysis → | |
| 11 Oct 26 |
 Como v
Roma |
0–0 | 1.85–1.14 | +0.71 | missed | analysis → | |
| 11 Oct 26 |
 Cagliari v
 Juventus |
1–1 | 0.81–2.38 | -1.57 | missed | analysis → | |
| 11 Oct 26 |
Real Sociedad v
 Deportivo La Coruna |
3–2 | 1.33–2.34 | -1.01 | called it | analysis → | |
| 11 Oct 26 |
 Real Betis v
Osasuna |
5–2 | 2.66–1.15 | +1.51 | called it | analysis → | |
| 11 Oct 26 |
Racing Santander v
 Valencia |
2–2 | 1.40–1.75 | -0.35 | missed | analysis → | |
| 11 Oct 26 |
 Elche v
Celta Vigo |
0–0 | 0.73–0.43 | +0.30 | missed | analysis → | |
| 11 Oct 26 |
Liverpool v
Manchester City |
0–3 | 0.76–1.59 | -0.83 | missed | analysis → | |
| 11 Oct 26 |
Hull v
Everton |
2–2 | 1.53–2.80 | -1.27 | missed | analysis → | |
| 11 Oct 26 |
Crystal Palace v
Nottingham Forest |
3–1 | 2.25–3.76 | -1.51 | called it | analysis → | |
| 10 Oct 26 |
Union Berlin v
Elversberg |
1–0 | 1.24–0.28 | +0.96 | called it | analysis → | |
| 10 Oct 26 |
RasenBallsport Leipzig v
 Eintracht Frankfurt |
6–0 | 4.74–1.96 | +2.78 | called it | analysis → | |
| 10 Oct 26 |
Paderborn v
VfB Stuttgart |
0–2 | 2.09–2.59 | -0.50 | called it | analysis → | |
| 10 Oct 26 |
Mainz 05 v
 Bayer Leverkusen |
3–4 | 2.17–2.29 | -0.12 | called it | analysis → | |
| 10 Oct 26 |
 Hoffenheim v
Hamburger SV |
2–3 | 0.93–1.51 | -0.58 | missed | analysis → | |
| 10 Oct 26 |
Augsburg v
Bayern Munich |
2–2 | 1.96–4.58 | -2.62 | missed | analysis → | |
| 10 Oct 26 |
Paris Saint Germain v
 Le Mans |
3–1 | 4.23–2.23 | +2.00 | called it | analysis → | |
| 10 Oct 26 |
 Monaco v
 Toulouse |
2–2 | 1.23–1.41 | -0.18 | missed | analysis → | |
| 10 Oct 26 |
 Lorient v
 Paris FC |
1–0 | 0.49–0.90 | -0.41 | missed | analysis → | |
| 10 Oct 26 |
Lille v
 Le Havre |
1–0 | 1.32–0.84 | +0.48 | called it | analysis → | |
| 10 Oct 26 |
Brest v
 Angers |
5–1 | 4.04–1.19 | +2.85 | called it | analysis → | |
| 10 Oct 26 |
Sunderland v
Brighton and Hove Albion |
0–2 | 0.87–1.64 | -0.77 | called it | analysis → | |
| 10 Oct 26 |
Manchester United v
Tottenham Hotspur |
1–1 | 0.87–2.03 | -1.16 | missed | analysis → | |
| 10 Oct 26 |
Ipswich v
Fulham |
2–1 | 1.26–1.78 | -0.52 | missed | analysis → | |
| 10 Oct 26 |
Chelsea v
Bournemouth |
5–1 | 1.37–1.68 | -0.31 | called it | analysis → | |
| 10 Oct 26 |
Aston Villa v
Brentford |
2–2 | 1.84–1.28 | +0.56 | missed | analysis → | |
| 10 Oct 26 |
Arsenal v
Leeds United |
2–1 | 2.32–1.20 | +1.12 | called it | analysis → | |
| 20 Sep 26 |
 Parma Calcio 1913 v
Genoa |
2–1 | 0.86–0.86 | +0.00 | missed | analysis → | |
| 20 Sep 26 |
Juventus v
 Atalanta |
2–0 | 1.92–0.49 | +1.43 | called it | analysis → | |
| 20 Sep 26 |
 Frosinone v
Como |
2–0 | 2.26–2.29 | -0.03 | missed | analysis → | |
| 20 Sep 26 |
 Fiorentina v
Napoli |
1–1 | 1.06–2.13 | -1.07 | missed | analysis → |
34 of 60 most-likely calls landed — 57%. Always backing the home side over the same matches would have called 50%, so the model is +7 points ahead of that.
Picking the most likely of three outcomes is a coarse test: it ignores how confident the model was, and a model can be right more often while forecasting worse. The measured versions are on Model, Calibration and Receipts.
The model pick is the out-of-sample forecast: made by a model fitted only on matches played before that one, so it is what the model actually said in advance. A dash means no forecast was on record before kick-off.