Home win
54.00%
Real Madrid
ESP1 · La Liga · Match Analysis
Real Madrid vs
Villarreal
Model-based match probabilities derived from multi-season xG data, team strength, venue performance, recent form, Elo ratings and league context.
Model output
Match probabilities.
Probabilities are model estimates, not certainties or betting guarantees.
Draw
23.04%
X
Away win
22.96%
Villarreal
Score model
Most likely scorelines.
Exact-score probabilities from the Poisson + Dixon-Coles goal model.
Additional scorelines
1:26.00%
1:05.34%
3:04.91%
3:24.55%
0:04.19%
Team context
Strength and recent form.
Venue-adjusted team strength, Elo rating and recent xG performance.
Home
Real Madrid
Elo1650.5
Overall attack2.10
Overall defense1.10
Home attack2.37
Home defense0.97
Form signal+1.031
Last 5 xGF / xGA2.32 / 1.33
Last 10 xGF / xGA2.30 / 1.20
Last 10 xGD+1.10
League statusSTABLE · from ESP1 2025/26
Away
Villarreal
Elo1561.7
Overall attack1.73
Overall defense1.24
Away attack1.53
Away defense1.22
Form signal+1.473
Last 5 xGF / xGA2.78 / 1.22
Last 10 xGF / xGA2.52 / 1.21
Last 10 xGD+1.31
League statusSTABLE · from ESP1 2025/26
Elo difference
+88.8
Elo is one input among several and does not determine the final 1X2 probability on its own.
Head-to-head
Historical matchup context.
H2H values provide context and are not used as a standalone prediction.
Goal distribution
Total goals profile.
The model's probability distribution for the total number of goals.
04.19%
18.33%
219.70%
321.60%
418.86%
513.18%
67.67%
7+6.48%
Technical detail
How the model reached the estimate.
Detailed diagnostics are kept below the headline probabilities so the public report remains readable.
League baseline and model setup
LeagueESP1
Loaded teams29
Loaded historical matches1589
Target season2026/27
Previous seasons in history4
Season window2022/23 → 2023/24 → 2024/25 → 2025/26 → 2026/27
League tierTier 1
Historical league matches1589
xG matches1589 / 1589
Effective xG matches812.65
Observed home xG1.52
Observed away xG1.16
Model home xG baseline1.52
Model away xG baseline1.16
Total league xG baseline2.68
Home-advantage weight1.000
Model configdata/models/ESP1.json
Team attack half-life240.0 days
Team defense half-life240.0 days
Attack strength weight1.000
Defense strength weight1.000
Venue prior matches0.00
DistributionPoisson + Dixon-Coles
Dixon-Coles rho-0.130 (effective -0.130)
Dixon-Coles decay0.000
Goal overdispersion0.000
Implied goal variance H / A2.13 / 1.36
Relative strengths and form diagnostics
Real Madrid
Home attack strength1.428
Home defense weakness0.872
Base lambda home1.84
Attack form xGF level2.312
Defense form xGA level (lower is better)1.281
Opponent-adjusted attack residual+0.971
Opponent-adjusted defense residual (higher is better)+0.059
Opponent-adjustment weight0.000
Model attack signal+0.971
Model defense signal+0.060
Opponent blend0.000
Form matches 5 / 105 / 10
Last 5 xGF / xGA / xGD2.32 / 1.33 / +0.99
Last 10 xGF / xGA / xGD2.30 / 1.20 / +1.10
Villarreal
Away attack strength1.243
Away defense weakness0.848
Base lambda away1.26
Attack form xGF level2.690
Defense form xGA level (lower is better)1.217
Opponent-adjusted attack residual+1.290
Opponent-adjusted defense residual (higher is better)+0.066
Opponent-adjustment weight0.000
Model attack signal+1.348
Model defense signal+0.124
Opponent blend0.000
Form matches 5 / 105 / 10
Last 5 xGF / xGA / xGD2.78 / 1.22 / +1.56
Last 10 xGF / xGA / xGD2.52 / 1.21 / +1.31
Historical data coverage
Real Madrid
Total matches159
Home / away matches79 / 80
xG matches159 / 159
xG home / away coverage79 / 79 · 80 / 80
Effective total33.10
Effective attack / defense33.10 / 33.10
Effective home matches16.19
Effective home attack / defense16.19 / 16.19
Villarreal
Total matches159
Home / away matches79 / 80
xG matches159 / 159
xG home / away coverage79 / 79 · 80 / 80
Effective total33.26
Effective attack / defense33.26 / 33.26
Effective away matches17.07
Effective away attack / defense17.07 / 17.07
Historical data coverage
95.95%
This measures usable historical evidence. It is not the probability that the prediction is correct.
Validation
Historical model performance.
Leakage-safe backtesting compares the fitted model with a league baseline over 1569 historical matches.
Positive improvement means the model performed better than the league baseline.
Total goals log-loss vs baseline+1.50%
1X2 log-loss vs baseline+8.05%
BTTS Brier vs baseline+0.37%
Over 2.5 Brier vs baseline+3.17%
Full backtest metrics
Model
MAE total goals1.2675
Exact score log-loss2.8109
Total goals Brier0.8166
Total goals log-loss1.8408
1X2 Brier0.5824
1X2 log-loss0.9795
BTTS Brier0.2480
Over 1.5 Brier0.1911
Over 2.5 Brier0.2416
Over 3.5 Brier0.1839
League baseline
MAE total goals1.3119
Exact score log-loss2.9284
Total goals Brier0.8217
Total goals log-loss1.8688
1X2 Brier0.6432
1X2 log-loss1.0652
BTTS Brier0.2489
Over 1.5 Brier0.1956
Over 2.5 Brier0.2495
Over 3.5 Brier0.1899
Calibration diagnostics
Home win · ECE 3.61%
0- 10% n=3 avg= 8.4% actual= 0.0% 10- 20% n=73 avg= 16.1% actual= 19.2% 20- 30% n=221 avg= 25.3% actual= 23.5% 30- 40% n=389 avg= 35.2% actual= 35.5% 40- 50% n=420 avg= 44.7% actual= 47.9% 50- 60% n=224 avg= 54.6% actual= 59.4% 60- 70% n=139 avg= 64.7% actual= 74.8% 70- 80% n=86 avg= 74.4% actual= 87.2% 80- 90% n=14 avg= 81.9% actual= 85.7%
Draw · ECE 2.16%
10- 20% n=128 avg= 16.9% actual= 7.8% 20- 30% n=917 avg= 26.7% actual= 26.3% 30- 40% n=524 avg= 31.6% actual= 28.1%
Away win · ECE 2.30%
0- 10% n=89 avg= 7.7% actual= 5.6% 10- 20% n=318 avg= 15.4% actual= 12.6% 20- 30% n=508 avg= 25.2% actual= 22.6% 30- 40% n=344 avg= 34.4% actual= 36.0% 40- 50% n=189 avg= 44.6% actual= 47.1% 50- 60% n=82 avg= 54.0% actual= 53.7% 60- 70% n=33 avg= 64.8% actual= 60.6% 70- 80% n=6 avg= 74.3% actual= 83.3%
BTTS · ECE 2.57%
30- 40% n=36 avg= 38.0% actual= 50.0% 40- 50% n=605 avg= 46.1% actual= 50.4% 50- 60% n=775 avg= 54.1% actual= 53.2% 60- 70% n=145 avg= 63.0% actual= 63.4% 70- 80% n=8 avg= 72.2% actual= 50.0%
Over 2.5 · ECE 1.75%
20- 30% n=18 avg= 28.2% actual= 38.9% 30- 40% n=265 avg= 36.7% actual= 34.0% 40- 50% n=639 avg= 45.1% actual= 46.2% 50- 60% n=415 avg= 54.5% actual= 53.3% 60- 70% n=189 avg= 64.2% actual= 62.4% 70- 80% n=42 avg= 73.1% actual= 66.7% 80- 90% n=1 avg= 81.7% actual=100.0%
Walk-forward - latest periods
| Period | N | Model 1X2 LL | Baseline 1X2 LL | Improvement |
|---|---|---|---|---|
| 2025-08 | 31 | 0.9981 | 1.0681 | +6.56% |
| 2025-09 | 39 | 0.9340 | 1.0517 | +11.19% |
| 2025-10 | 31 | 1.0050 | 1.0881 | +7.64% |
| 2025-11 | 38 | 0.8962 | 1.0476 | +14.46% |
| 2025-12 | 32 | 0.9782 | 1.0496 | +6.80% |
| 2026-01 | 43 | 0.9692 | 1.0643 | +8.94% |
| 2026-02 | 40 | 0.9765 | 1.0599 | +7.87% |
| 2026-03 | 36 | 0.9551 | 1.0211 | +6.46% |
| 2026-04 | 40 | 1.0381 | 1.0228 | -1.50% |
| 2026-05 | 50 | 1.0149 | 1.0590 | +4.16% |
| 2026-08 | 30 | 0.9403 | 1.0318 | +8.87% |
| 2026-09 | 39 | 0.9923 | 1.1010 | +9.87% |
Football is data.
NILYTIX turns match data into transparent model probabilities and measurable analytical signals.