ITA1 · Serie A · Match Analysis

AC Milan vs
Atalanta

Model-based match probabilities derived from multi-season xG data, team strength, venue performance, recent form, Elo ratings and league context.

Analysis date18 Oct 2026

Target season2026/27

History window5 seasons

League tierTier 1

Model output

Match probabilities.

Probabilities are model estimates, not certainties or betting guarantees.

Home win

48.96%

AC Milan

Draw

25.93%

Away win

25.10%

Atalanta

Expected goals 1.74 - 1.20 2.93 total

BTTS 58.62% Both teams to score

Over 1.5 80.22% 2+ total goals

Over 2.5 56.17% 3+ total goals

Score model

Most likely scorelines.

Exact-score probabilities from the Poisson + Dixon-Coles goal model.

#11:112.21%

#22:19.61%

#31:08.10%

#42:08.04%

#51:26.61%

Additional scorelines

0:06.47%

2:25.74%

3:15.57%

0:15.21%

3:04.66%

Team context

Strength and recent form.

Venue-adjusted team strength, Elo rating and recent xG performance.

Home

AC Milan

Elo1590.0

Overall attack1.59

Overall defense1.16

Home attack1.76

Home defense1.08

Form signal+0.243

Last 5 xGF / xGA1.39 / 1.13

Last 10 xGF / xGA1.48 / 1.27

Last 10 xGD+0.20

League statusSTABLE · from ITA1 2025/26

Away

Atalanta

Elo1550.1

Overall attack1.49

Overall defense1.27

Away attack1.36

Away defense1.40

Form signal-0.613

Last 5 xGF / xGA1.23 / 1.88

Last 10 xGF / xGA1.22 / 1.78

Last 10 xGD-0.56

League statusSTABLE · from ITA1 2025/26

Elo difference

+40.0

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.

H2H matches8Effective: 2.31

Average goals3.10AC Milan 1.26 · Atalanta 1.84

H2H 3+ goals49.76%Historical context

H2H BTTS81.17%Model BTTS: 58.62%

H2H 2+ goals84.64%

H2H 3+ goals49.76%

H2H 4+ goals37.87%

H2H BTTS81.17%

Goal distribution

Total goals profile.

The model's probability distribution for the total number of goals.

06.47%

113.31%

224.05%

322.39%

416.42%

59.63%

64.71%

7+3.03%

0-1 goals19.78%Low total

2+ goals80.22%Over 1.5

3+ goals56.17%Over 2.5

4+ goals33.78%Over 3.5

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 matches1580

Target season2026/27

Previous seasons in history4

Season window2022/23 → 2023/24 → 2024/25 → 2025/26 → 2026/27

League tierTier 1

Historical league matches1580

xG matches1580 / 1580

Effective xG matches798.37

Observed home xG1.37

Observed away xG1.17

Model home xG baseline1.36

Model away xG baseline1.18

Total league xG baseline2.54

Home-advantage weight0.875

Model configdata/models/ITA1.json

Team attack half-life240.0 days

Team defense half-life240.0 days

Attack strength weight1.050

Defense strength weight1.081

Venue prior matches3.31

DistributionPoisson + Dixon-Coles

Dixon-Coles rho-0.106 (effective -0.104)

Dixon-Coles decay0.034

Goal overdispersion0.000

Implied goal variance H / A1.74 / 1.20

Relative strengths and form diagnostics

AC Milan

Home attack strength1.215

Home defense weakness0.940

Base lambda home1.69

Attack form xGF level1.424

Defense form xGA level (lower is better)1.181

Opponent-adjusted attack residual+0.147

Opponent-adjusted defense residual (higher is better)+0.087

Opponent-adjustment weight0.000

Model attack signal+0.152

Model defense signal+0.091

Opponent blend0.000

Form matches 5 / 105 / 10

Last 5 xGF / xGA / xGD1.39 / 1.13 / +0.26

Last 10 xGF / xGA / xGD1.48 / 1.27 / +0.20

Atalanta

Away attack strength1.127

Away defense weakness1.014

Base lambda away1.25

Attack form xGF level1.229

Defense form xGA level (lower is better)1.842

Opponent-adjusted attack residual-0.085

Opponent-adjusted defense residual (higher is better)-0.608

Opponent-adjustment weight0.000

Model attack signal-0.042

Model defense signal-0.571

Opponent blend0.000

Form matches 5 / 105 / 10

Last 5 xGF / xGA / xGD1.23 / 1.88 / -0.64

Last 10 xGF / xGA / xGD1.22 / 1.78 / -0.56

Historical data coverage

AC Milan

Total matches158

Home / away matches78 / 80

xG matches158 / 158

xG home / away coverage78 / 78 · 80 / 80

Effective total31.66

Effective attack / defense31.66 / 31.66

Effective home matches14.90

Effective home attack / defense14.90 / 14.90

Atalanta

Total matches158

Home / away matches80 / 78

xG matches158 / 158

xG home / away coverage80 / 80 · 78 / 78

Effective total31.59

Effective attack / defense31.59 / 31.59

Effective away matches14.96

Effective away attack / defense14.96 / 14.96

Historical data coverage

95.68%

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 1560 historical matches.

Positive improvement means the model performed better than the league baseline.

Total goals log-loss vs baseline+0.63%

1X2 log-loss vs baseline+9.07%

BTTS Brier vs baseline+0.37%

Over 2.5 Brier vs baseline+1.34%

Full backtest metrics

Model

MAE total goals1.2171

Exact score log-loss2.7695

Total goals Brier0.8119

Total goals log-loss1.8027

1X2 Brier0.5899

1X2 log-loss0.9879

BTTS Brier0.2496

Over 1.5 Brier0.1909

Over 2.5 Brier0.2463

Over 3.5 Brier0.1841

League baseline

MAE total goals1.2442

Exact score log-loss2.8998

Total goals Brier0.8136

Total goals log-loss1.8143

1X2 Brier0.6583

1X2 log-loss1.0864

BTTS Brier0.2505

Over 1.5 Brier0.1925

Over 2.5 Brier0.2497

Over 3.5 Brier0.1866

Calibration diagnostics

Home win · ECE 2.51%

 0- 10%  n=22   avg=  8.3% actual=  9.1%
10- 20%  n=177  avg= 15.9% actual= 15.8%
20- 30%  n=282  avg= 25.2% actual= 25.5%
30- 40%  n=333  avg= 35.2% actual= 33.6%
40- 50%  n=297  avg= 44.7% actual= 48.1%
50- 60%  n=218  avg= 54.8% actual= 51.8%
60- 70%  n=161  avg= 64.5% actual= 70.8%
70- 80%  n=59   avg= 73.8% actual= 79.7%
80- 90%  n=11   avg= 83.1% actual= 63.6%

Draw · ECE 1.65%

 0- 10%  n=1    avg=  8.3% actual=  0.0%
10- 20%  n=94   avg= 17.1% actual= 21.3%
20- 30%  n=1059 avg= 26.5% actual= 25.3%
30- 40%  n=406  avg= 31.5% actual= 33.7%

Away win · ECE 3.79%

 0- 10%  n=73   avg=  7.7% actual=  6.8%
10- 20%  n=308  avg= 15.0% actual= 12.3%
20- 30%  n=368  avg= 25.0% actual= 22.0%
30- 40%  n=343  avg= 34.6% actual= 30.6%
40- 50%  n=208  avg= 44.6% actual= 53.4%
50- 60%  n=160  avg= 54.3% actual= 53.8%
60- 70%  n=74   avg= 63.6% actual= 70.3%
70- 80%  n=25   avg= 73.6% actual= 76.0%
80- 90%  n=1    avg= 83.8% actual=  0.0%

BTTS · ECE 1.92%

30- 40%  n=64   avg= 37.9% actual= 45.3%
40- 50%  n=823  avg= 45.9% actual= 48.4%
50- 60%  n=637  avg= 53.6% actual= 53.1%
60- 70%  n=36   avg= 61.8% actual= 58.3%

Over 2.5 · ECE 2.49%

20- 30%  n=17   avg= 28.1% actual= 23.5%
30- 40%  n=300  avg= 36.9% actual= 40.0%
40- 50%  n=705  avg= 44.9% actual= 47.1%
50- 60%  n=455  avg= 53.9% actual= 51.6%
60- 70%  n=78   avg= 63.4% actual= 60.3%
70- 80%  n=5    avg= 71.5% actual= 80.0%

Walk-forward - latest periods

Period	N	Model 1X2 LL	Baseline 1X2 LL	Improvement
2025-09	30	0.9825	1.0882	+9.71%
2025-10	40	1.0491	1.1060	+5.14%
2025-11	39	1.0141	1.0922	+7.15%
2025-12	37	1.0639	1.0763	+1.15%
2026-01	58	0.9365	1.0986	+14.75%
2026-02	40	0.9507	1.0931	+13.03%
2026-03	36	0.9894	1.0574	+6.43%
2026-04	40	1.0026	1.0882	+7.87%
2026-05	40	1.1007	1.1080	+0.66%
2026-08	20	0.7777	1.0805	+28.03%
2026-09	30	1.0237	1.0700	+4.33%
2026-10	10	0.8909	1.0354	+13.96%

Football is data.

NILYTIX turns match data into transparent model probabilities and measurable analytical signals.

Back to Matches Follow on X

AC Milan vsAtalanta

Match probabilities.

Most likely scorelines.

Strength and recent form.

AC Milan

Atalanta

Historical matchup context.

Total goals profile.

How the model reached the estimate.

AC Milan

Atalanta

AC Milan

Atalanta

Historical model performance.

Home win · ECE 2.51%

Draw · ECE 1.65%

Away win · ECE 3.79%

BTTS · ECE 1.92%

Over 2.5 · ECE 2.49%

Football is data.

AC Milan vs
Atalanta