Home win
32.72%
PEC Zwolle
NED1 · Eredivisie · Match Analysis
PEC Zwolle vs
SC Cambuur
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
27.49%
X
Away win
39.79%
SC Cambuur
Score model
Most likely scorelines.
Exact-score probabilities from the Poisson + Dixon-Coles goal model.
Additional scorelines
0:26.36%
2:26.12%
1:05.90%
2:05.12%
1:34.55%
Team context
Strength and recent form.
Venue-adjusted team strength, Elo rating and recent xG performance.
Home
PEC Zwolle
Elo1448.4
Overall attack1.18
Overall defense1.93
Home attack1.28
Home defense1.76
Form signal-1.189
Last 5 xGF / xGA0.93 / 2.10
Last 10 xGF / xGA0.89 / 2.12
Last 10 xGD-1.23
League statusSTABLE · from NED1 2025/26
Away
SC Cambuur
Elo1428.8
Overall attack1.40
Overall defense2.03
Away attack1.33
Away defense2.05
Form signal-0.470
Last 5 xGF / xGA1.38 / 1.84
Last 10 xGF / xGA1.43 / 1.92
Last 10 xGD-0.49
League statusSTABLE · from NED1 2022/23
Elo difference
+19.6
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.
06.80%
112.64%
224.38%
322.39%
416.42%
59.63%
64.71%
7+3.03%
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
LeagueNED1
Loaded teams25
Loaded historical matches1287
Target season2026/27
Previous seasons in history4
Season window2022/23 → 2023/24 → 2024/25 → 2025/26 → 2026/27
League tierTier 1
Historical league matches1287
xG matches369 / 1287
Effective xG matches293.54
Observed home xG1.78
Observed away xG1.44
Model home xG baseline1.78
Model away xG baseline1.44
Total league xG baseline3.22
Home-advantage weight1.000
Model configdata/models/NED1.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 / A1.39 / 1.55
Relative strengths and form diagnostics
PEC Zwolle
Home attack strength0.809
Home defense weakness1.149
Base lambda home1.52
Attack form xGF level0.916
Defense form xGA level (lower is better)2.105
Opponent-adjusted attack residual-0.628
Opponent-adjusted defense residual (higher is better)-0.427
Opponent-adjustment weight0.000
Model attack signal-0.696
Model defense signal-0.494
Opponent blend0.000
Form matches 5 / 105 / 10
Last 5 xGF / xGA / xGD0.93 / 2.10 / -1.17
Last 10 xGF / xGA / xGD0.89 / 2.12 / -1.23
SC Cambuur
Away attack strength0.973
Away defense weakness1.054
Base lambda away1.62
Attack form xGF level1.399
Defense form xGA level (lower is better)1.868
Opponent-adjusted attack residual-0.170
Opponent-adjusted defense residual (higher is better)-0.217
Opponent-adjustment weight0.000
Model attack signal-0.213
Model defense signal-0.257
Opponent blend0.000
Form matches 5 / 105 / 7
Last 5 xGF / xGA / xGD1.38 / 1.84 / -0.46
Last 10 xGF / xGA / xGD1.43 / 1.92 / -0.49
Historical data coverage
PEC Zwolle
Total matches109
Home / away matches54 / 55
xG matches41 / 109
xG home / away coverage20 / 54 · 21 / 55
Effective total21.50
Effective attack / defense21.50 / 21.50
Effective home matches10.40
Effective home attack / defense10.40 / 10.40
SC Cambuur
Total matches41
Home / away matches21 / 20
xG matches7 / 41
xG home / away coverage4 / 21 · 3 / 20
Effective total6.19
Effective attack / defense6.19 / 6.19
Effective away matches2.69
Effective away attack / defense2.69 / 2.69
Historical data coverage
27.67%
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 1267 historical matches.
Positive improvement means the model performed better than the league baseline.
Total goals log-loss vs baseline+0.41%
1X2 log-loss vs baseline+3.63%
BTTS Brier vs baseline+0.48%
Over 2.5 Brier vs baseline+0.86%
Full backtest metrics
Model
MAE total goals1.4011
Exact score log-loss3.1841
Total goals Brier0.8384
Total goals log-loss1.9586
1X2 Brier0.6224
1X2 log-loss1.0352
BTTS Brier0.2414
Over 1.5 Brier0.1405
Over 2.5 Brier0.2434
Over 3.5 Brier0.2450
League baseline
MAE total goals1.4076
Exact score log-loss3.2466
Total goals Brier0.8397
Total goals log-loss1.9666
1X2 Brier0.6498
1X2 log-loss1.0742
BTTS Brier0.2426
Over 1.5 Brier0.1411
Over 2.5 Brier0.2455
Over 3.5 Brier0.2473
Calibration diagnostics
Home win · ECE 10.96%
10- 20% n=19 avg= 16.7% actual= 21.1% 20- 30% n=50 avg= 25.1% actual= 16.0% 30- 40% n=328 avg= 37.6% actual= 22.9% 40- 50% n=732 avg= 42.7% actual= 53.4% 50- 60% n=66 avg= 54.8% actual= 50.0% 60- 70% n=49 avg= 65.2% actual= 59.2% 70- 80% n=21 avg= 74.5% actual= 71.4% 80- 90% n=2 avg= 80.3% actual=100.0%
Draw · ECE 3.07%
10- 20% n=49 avg= 17.8% actual= 24.5% 20- 30% n=1216 avg= 28.0% actual= 25.1% 30- 40% n=2 avg= 30.2% actual= 0.0%
Away win · ECE 11.16%
0- 10% n=19 avg= 8.3% actual= 0.0% 10- 20% n=73 avg= 15.1% actual= 16.4% 20- 30% n=622 avg= 27.7% actual= 18.0% 30- 40% n=456 avg= 32.2% actual= 47.4% 40- 50% n=53 avg= 44.1% actual= 39.6% 50- 60% n=32 avg= 54.6% actual= 75.0% 60- 70% n=11 avg= 63.9% actual= 63.6% 70- 80% n=1 avg= 70.3% actual=100.0%
BTTS · ECE 0.92%
40- 50% n=8 avg= 48.4% actual= 62.5% 50- 60% n=1029 avg= 55.3% actual= 55.0% 60- 70% n=213 avg= 64.0% actual= 66.2% 70- 80% n=17 avg= 71.9% actual= 88.2%
Over 2.5 · ECE 7.63%
40- 50% n=784 avg= 49.4% actual= 57.8% 50- 60% n=267 avg= 53.5% actual= 62.9% 60- 70% n=170 avg= 64.6% actual= 66.5% 70- 80% n=45 avg= 72.9% actual= 77.8% 80- 90% n=1 avg= 80.7% actual=100.0%
Walk-forward - latest periods
| Period | N | Model 1X2 LL | Baseline 1X2 LL | Improvement |
|---|---|---|---|---|
| 2025-08 | 34 | 0.9435 | 1.0019 | +5.83% |
| 2025-09 | 29 | 0.9687 | 1.0497 | +7.71% |
| 2025-10 | 28 | 0.9236 | 1.0610 | +12.95% |
| 2025-11 | 34 | 1.0679 | 1.0447 | -2.22% |
| 2025-12 | 27 | 1.0159 | 1.1117 | +8.62% |
| 2026-01 | 30 | 1.1277 | 1.1318 | +0.36% |
| 2026-02 | 38 | 1.0330 | 1.0726 | +3.69% |
| 2026-03 | 32 | 1.0069 | 1.0449 | +3.64% |
| 2026-04 | 27 | 0.9071 | 1.0747 | +15.59% |
| 2026-05 | 27 | 0.9953 | 1.1101 | +10.34% |
| 2026-08 | 33 | 1.0148 | 1.1182 | +9.25% |
| 2026-09 | 30 | 1.0027 | 1.1404 | +12.08% |
Football is data.
NILYTIX turns match data into transparent model probabilities and measurable analytical signals.