Home win
50.86%
Napoli
ITA1 · Serie A · Match Analysis
Napoli vs
Frosinone
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
27.43%
X
Away win
21.72%
Frosinone
Score model
Most likely scorelines.
Exact-score probabilities from the Poisson + Dixon-Coles goal model.
Additional scorelines
1:25.92%
0:15.48%
3:15.33%
3:05.23%
2:24.91%
Team context
Strength and recent form.
Venue-adjusted team strength, Elo rating and recent xG performance.
Home
Napoli
Elo1598.3
Overall attack1.38
Overall defense1.01
Home attack1.42
Home defense0.96
Form signal+0.123
Last 5 xGF / xGA1.47 / 1.39
Last 10 xGF / xGA1.46 / 1.25
Last 10 xGD+0.21
League statusSTABLE · from ITA1 2025/26
Away
Frosinone
Elo1488.9
Overall attack1.43
Overall defense1.92
Away attack1.40
Away defense1.86
Form signal-0.340
Last 5 xGF / xGA1.67 / 2.08
Last 10 xGF / xGA1.59 / 1.80
Last 10 xGD-0.21
League statusSTABLE · from ITA1 2023/24
Elo difference
+109.3
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.
08.37%
115.37%
226.14%
321.99%
414.73%
57.89%
63.52%
7+1.98%
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
LeagueITA1
Loaded teams27
Loaded historical matches1570
Target season2026/27
Previous seasons in history4
Season window2022/23 → 2023/24 → 2024/25 → 2025/26 → 2026/27
League tierTier 1
Historical league matches1570
xG matches1570 / 1570
Effective xG matches794.53
Observed home xG1.37
Observed away xG1.17
Model home xG baseline1.37
Model away xG baseline1.17
Total league xG baseline2.53
Home-advantage weight1.000
Model configdata/models/ITA1.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 / A1.66 / 1.02
Relative strengths and form diagnostics
Napoli
Home attack strength1.027
Home defense weakness0.863
Base lambda home1.59
Attack form xGF level1.462
Defense form xGA level (lower is better)1.339
Opponent-adjusted attack residual+0.186
Opponent-adjusted defense residual (higher is better)-0.079
Opponent-adjustment weight0.000
Model attack signal+0.195
Model defense signal-0.071
Opponent blend0.000
Form matches 5 / 105 / 10
Last 5 xGF / xGA / xGD1.47 / 1.39 / +0.08
Last 10 xGF / xGA / xGD1.46 / 1.25 / +0.21
Frosinone
Away attack strength1.073
Away defense weakness1.133
Base lambda away1.08
Attack form xGF level1.644
Defense form xGA level (lower is better)1.984
Opponent-adjusted attack residual+0.375
Opponent-adjusted defense residual (higher is better)-0.720
Opponent-adjustment weight0.000
Model attack signal+0.377
Model defense signal-0.717
Opponent blend0.000
Form matches 5 / 105 / 10
Last 5 xGF / xGA / xGD1.67 / 2.08 / -0.41
Last 10 xGF / xGA / xGD1.59 / 1.80 / -0.21
Historical data coverage
Napoli
Total matches157
Home / away matches78 / 79
xG matches157 / 157
xG home / away coverage78 / 78 · 79 / 79
Effective total31.33
Effective attack / defense31.33 / 31.33
Effective home matches15.45
Effective home attack / defense15.45 / 15.45
Frosinone
Total matches43
Home / away matches22 / 21
xG matches43 / 43
xG home / away coverage22 / 22 · 21 / 21
Effective total6.67
Effective attack / defense6.67 / 6.67
Effective away matches2.89
Effective away attack / defense2.89 / 2.89
Historical data coverage
28.11%
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 1550 historical matches.
Positive improvement means the model performed better than the league baseline.
Total goals log-loss vs baseline+0.67%
1X2 log-loss vs baseline+8.58%
BTTS Brier vs baseline+0.50%
Over 2.5 Brier vs baseline+1.36%
Full backtest metrics
Model
MAE total goals1.2186
Exact score log-loss2.7775
Total goals Brier0.8123
Total goals log-loss1.8045
1X2 Brier0.5943
1X2 log-loss0.9942
BTTS Brier0.2495
Over 1.5 Brier0.1911
Over 2.5 Brier0.2462
Over 3.5 Brier0.1842
League baseline
MAE total goals1.2465
Exact score log-loss2.9028
Total goals Brier0.8142
Total goals log-loss1.8167
1X2 Brier0.6590
1X2 log-loss1.0874
BTTS Brier0.2507
Over 1.5 Brier0.1929
Over 2.5 Brier0.2496
Over 3.5 Brier0.1867
Calibration diagnostics
Home win · ECE 3.20%
0- 10% n=6 avg= 8.4% actual= 33.3% 10- 20% n=131 avg= 16.3% actual= 11.5% 20- 30% n=287 avg= 25.2% actual= 24.4% 30- 40% n=370 avg= 35.1% actual= 32.2% 40- 50% n=332 avg= 44.6% actual= 47.0% 50- 60% n=244 avg= 54.7% actual= 57.0% 60- 70% n=141 avg= 64.2% actual= 70.9% 70- 80% n=36 avg= 73.2% actual= 83.3% 80- 90% n=3 avg= 82.8% actual= 33.3%
Draw · ECE 1.11%
10- 20% n=45 avg= 17.9% actual= 17.8% 20- 30% n=951 avg= 26.9% actual= 25.1% 30- 40% n=554 avg= 31.7% actual= 31.6%
Away win · ECE 4.02%
0- 10% n=46 avg= 8.2% actual= 2.2% 10- 20% n=319 avg= 15.4% actual= 11.9% 20- 30% n=417 avg= 25.1% actual= 22.8% 30- 40% n=370 avg= 34.6% actual= 34.1% 40- 50% n=227 avg= 45.0% actual= 56.4% 50- 60% n=123 avg= 54.6% actual= 57.7% 60- 70% n=43 avg= 64.2% actual= 79.1% 70- 80% n=5 avg= 73.8% actual= 60.0%
BTTS · ECE 1.86%
30- 40% n=43 avg= 38.3% actual= 44.2% 40- 50% n=766 avg= 46.2% actual= 48.2% 50- 60% n=697 avg= 53.8% actual= 52.4% 60- 70% n=44 avg= 62.1% actual= 59.1%
Over 2.5 · ECE 2.60%
20- 30% n=12 avg= 28.2% actual= 25.0% 30- 40% n=319 avg= 37.1% actual= 40.4% 40- 50% n=722 avg= 45.0% actual= 47.2% 50- 60% n=437 avg= 53.8% actual= 51.0% 60- 70% n=59 avg= 63.1% actual= 64.4% 70- 80% n=1 avg= 70.2% actual=100.0%
Walk-forward - latest periods
| Period | N | Model 1X2 LL | Baseline 1X2 LL | Improvement |
|---|---|---|---|---|
| 2025-08 | 20 | 0.9795 | 1.0890 | +10.05% |
| 2025-09 | 30 | 0.9876 | 1.0892 | +9.33% |
| 2025-10 | 40 | 1.0472 | 1.1022 | +4.99% |
| 2025-11 | 39 | 1.0238 | 1.0935 | +6.37% |
| 2025-12 | 37 | 1.0663 | 1.0804 | +1.31% |
| 2026-01 | 58 | 0.9498 | 1.0988 | +13.56% |
| 2026-02 | 40 | 0.9627 | 1.0955 | +12.12% |
| 2026-03 | 36 | 0.9898 | 1.0612 | +6.73% |
| 2026-04 | 40 | 1.0062 | 1.0889 | +7.60% |
| 2026-05 | 40 | 1.1031 | 1.1105 | +0.67% |
| 2026-08 | 20 | 0.8152 | 1.0882 | +25.09% |
| 2026-09 | 30 | 1.0258 | 1.0708 | +4.20% |
Football is data.
NILYTIX turns match data into transparent model probabilities and measurable analytical signals.