Home win
53.84%
Osasuna
ESP1 · La Liga · Match Analysis
Osasuna vs
Racing Santander
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
24.84%
X
Away win
21.32%
Racing Santander
Score model
Most likely scorelines.
Exact-score probabilities from the Poisson + Dixon-Coles goal model.
Additional scorelines
3:15.80%
1:25.75%
3:05.58%
0:15.48%
2:25.10%
Team context
Strength and recent form.
Venue-adjusted team strength, Elo rating and recent xG performance.
Home
Osasuna
Elo1462.5
Overall attack1.21
Overall defense1.37
Home attack1.34
Home defense1.09
Form signal-0.874
Last 5 xGF / xGA1.15 / 2.21
Last 10 xGF / xGA1.25 / 1.78
Last 10 xGD-0.53
League statusSTABLE · from ESP1 2025/26
Away
Racing Santander
Elo1486.3
Overall attack1.42
Overall defense2.17
Away attack1.13
Away defense2.96
Form signal-1.374
Last 5 xGF / xGA1.30 / 2.83
Last 10 xGF / xGA1.37 / 2.45
Last 10 xGD-1.08
League statusNEW
Elo difference
-23.8
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.
06.75%
115.37%
224.49%
322.27%
415.67%
58.82%
64.14%
7+2.50%
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 matches1598
Target season2026/27
Previous seasons in history4
Season window2022/23 → 2023/24 → 2024/25 → 2025/26 → 2026/27
League tierTier 1
Historical league matches1598
xG matches1598 / 1598
Effective xG matches815.35
Observed home xG1.52
Observed away xG1.17
Model home xG baseline1.52
Model away xG baseline1.17
Total league xG baseline2.68
Home-advantage weight1.006
Model configdata/models/ESP1.json
Team attack half-life365.0 days
Team defense half-life365.0 days
Attack strength weight1.113
Defense strength weight1.038
Venue prior matches5.00
DistributionPoisson + Dixon-Coles
Dixon-Coles rho-0.068 (effective -0.068)
Dixon-Coles decay0.006
Goal overdispersion0.000
Implied goal variance H / A1.77 / 1.04
Relative strengths and form diagnostics
Osasuna
Home attack strength0.906
Home defense weakness0.956
Base lambda home1.89
Attack form xGF level1.185
Defense form xGA level (lower is better)2.059
Opponent-adjusted attack residual-0.105
Opponent-adjusted defense residual (higher is better)-0.658
Opponent-adjustment weight0.000
Model attack signal-0.157
Model defense signal-0.716
Opponent blend0.000
Form matches 5 / 105 / 10
Last 5 xGF / xGA / xGD1.15 / 2.21 / -1.06
Last 10 xGF / xGA / xGD1.25 / 1.78 / -0.53
Racing Santander
Away attack strength1.004
Away defense weakness1.372
Base lambda away1.12
Attack form xGF level1.325
Defense form xGA level (lower is better)2.700
Opponent-adjusted attack residual+0.010
Opponent-adjusted defense residual (higher is better)-1.326
Opponent-adjustment weight0.000
Model attack signal-0.017
Model defense signal-1.357
Opponent blend0.000
Form matches 5 / 105 / 8
Last 5 xGF / xGA / xGD1.30 / 2.83 / -1.53
Last 10 xGF / xGA / xGD1.37 / 2.45 / -1.08
Historical data coverage
Osasuna
Total matches160
Home / away matches80 / 80
xG matches160 / 160
xG home / away coverage80 / 80 · 80 / 80
Effective total49.68
Effective attack / defense49.68 / 49.68
Effective home matches24.83
Effective home attack / defense24.83 / 24.83
Racing Santander
Total matches8
Home / away matches4 / 4
xG matches8 / 8
xG home / away coverage4 / 4 · 4 / 4
Effective total7.42
Effective attack / defense7.42 / 7.42
Effective away matches3.71
Effective away attack / defense3.71 / 3.71
Historical data coverage
29.28%
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 1578 historical matches.
Positive improvement means the model performed better than the league baseline.
Total goals log-loss vs baseline+1.53%
1X2 log-loss vs baseline+8.46%
BTTS Brier vs baseline+0.35%
Over 2.5 Brier vs baseline+3.43%
Full backtest metrics
Model
MAE total goals1.2686
Exact score log-loss2.8030
Total goals Brier0.8153
Total goals log-loss1.8368
1X2 Brier0.5779
1X2 log-loss0.9733
BTTS Brier0.2478
Over 1.5 Brier0.1908
Over 2.5 Brier0.2410
Over 3.5 Brier0.1840
League baseline
MAE total goals1.3152
Exact score log-loss2.9255
Total goals Brier0.8204
Total goals log-loss1.8654
1X2 Brier0.6420
1X2 log-loss1.0633
BTTS Brier0.2487
Over 1.5 Brier0.1953
Over 2.5 Brier0.2495
Over 3.5 Brier0.1904
Calibration diagnostics
Home win · ECE 2.73%
0- 10% n=9 avg= 8.4% actual= 11.1% 10- 20% n=97 avg= 15.9% actual= 20.6% 20- 30% n=220 avg= 25.5% actual= 23.6% 30- 40% n=339 avg= 35.3% actual= 36.3% 40- 50% n=393 avg= 44.7% actual= 46.1% 50- 60% n=231 avg= 54.5% actual= 55.8% 60- 70% n=144 avg= 64.4% actual= 69.4% 70- 80% n=99 avg= 74.5% actual= 87.9% 80- 90% n=46 avg= 83.0% actual= 87.0%
Draw · ECE 1.95%
0- 10% n=6 avg= 9.0% actual= 0.0% 10- 20% n=198 avg= 16.0% actual= 9.6% 20- 30% n=1141 avg= 26.3% actual= 27.5% 30- 40% n=233 avg= 31.4% actual= 29.6%
Away win · ECE 2.89%
0- 10% n=122 avg= 7.2% actual= 6.6% 10- 20% n=286 avg= 15.3% actual= 13.6% 20- 30% n=459 avg= 24.9% actual= 20.7% 30- 40% n=341 avg= 34.3% actual= 32.8% 40- 50% n=202 avg= 44.4% actual= 46.0% 50- 60% n=105 avg= 54.6% actual= 57.1% 60- 70% n=42 avg= 64.4% actual= 45.2% 70- 80% n=18 avg= 73.2% actual= 77.8% 80- 90% n=3 avg= 81.7% actual=100.0%
BTTS · ECE 3.15%
30- 40% n=67 avg= 37.8% actual= 47.8% 40- 50% n=729 avg= 45.9% actual= 50.6% 50- 60% n=662 avg= 54.0% actual= 53.6% 60- 70% n=115 avg= 63.1% actual= 67.8% 70- 80% n=5 avg= 73.1% actual= 60.0%
Over 2.5 · ECE 1.86%
20- 30% n=21 avg= 27.7% actual= 38.1% 30- 40% n=277 avg= 36.4% actual= 33.6% 40- 50% n=624 avg= 45.0% actual= 46.0% 50- 60% n=396 avg= 54.5% actual= 52.8% 60- 70% n=191 avg= 64.4% actual= 64.4% 70- 80% n=66 avg= 73.7% actual= 65.2% 80- 90% n=3 avg= 81.2% actual=100.0%
Walk-forward - latest periods
| Period | N | Model 1X2 LL | Baseline 1X2 LL | Improvement |
|---|---|---|---|---|
| 2025-09 | 39 | 0.9408 | 1.0522 | +10.58% |
| 2025-10 | 31 | 1.0046 | 1.0833 | +7.27% |
| 2025-11 | 38 | 0.8748 | 1.0447 | +16.26% |
| 2025-12 | 32 | 0.9668 | 1.0451 | +7.49% |
| 2026-01 | 43 | 0.9594 | 1.0625 | +9.71% |
| 2026-02 | 40 | 0.9680 | 1.0579 | +8.49% |
| 2026-03 | 36 | 0.9530 | 1.0198 | +6.56% |
| 2026-04 | 40 | 1.0503 | 1.0215 | -2.82% |
| 2026-05 | 50 | 1.0095 | 1.0529 | +4.12% |
| 2026-08 | 30 | 0.9317 | 1.0290 | +9.46% |
| 2026-09 | 39 | 0.9791 | 1.0986 | +10.88% |
| 2026-10 | 9 | 0.9149 | 1.0794 | +15.24% |
Football is data.
NILYTIX turns match data into transparent model probabilities and measurable analytical signals.