SCO1 · Scottish Premiership · Match Analysis

Celtic vs
Rangers

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

Analysis date20 Sep 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

55.03%

Celtic

Draw

23.61%

Away win

21.36%

Rangers

Expected goals 2.06 – 1.24 3.30 total

BTTS 63.18% Both teams to score

Over 1.5 85.32% 2+ total goals

Over 2.5 63.99% 3+ total goals

Score model

Most likely scorelines.

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

#11:110.66%

#22:19.71%

#32:07.83%

#43:16.66%

#51:06.39%

Additional scorelines

2:26.01%

1:25.84%

3:05.37%

0:04.93%

3:24.12%

Team context

Strength and recent form.

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

Home

Celtic

Elo1676.6

Overall attack2.35

Overall defense0.95

Home attack2.39

Home defense0.68

Form signal+1.954

Last 5 xGF / xGA3.02 / 1.00

Last 10 xGF / xGA2.78 / 0.94

Last 10 xGD+1.83

League statusSTABLE · from SCO1 2025/26

Away

Rangers

Elo1614.9

Overall attack2.06

Overall defense1.13

Away attack1.81

Away defense1.20

Form signal+1.479

Last 5 xGF / xGA2.63 / 0.99

Last 10 xGF / xGA2.35 / 1.16

Last 10 xGD+1.19

League statusSTABLE · from SCO1 2025/26

Elo difference

+61.7

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 matches10Effective: 4.21

Average goals3.35Celtic 1.68 · Rangers 1.68

H2H 3+ goals79.42%Historical context

H2H BTTS75.58%Model BTTS: 63.18%

H2H 2+ goals88.55%

H2H 3+ goals79.42%

H2H 4+ goals61.82%

H2H BTTS75.58%

Goal distribution

Total goals profile.

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

04.93%

19.75%

221.34%

322.10%

418.21%

512.01%

66.60%

7+5.07%

0–1 goals14.68%Low total

2+ goals85.32%Over 1.5

3+ goals63.99%Over 2.5

4+ goals41.89%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

LeagueSCO1

Loaded teams14

Loaded historical matches948

Target season2026/27

Previous seasons in history4

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

League tierTier 1

Historical league matches948

xG matches264 / 948

Effective xG matches212.83

Observed home xG1.60

Observed away xG1.28

Model home xG baseline1.60

Model away xG baseline1.28

Total league xG baseline2.87

Home-advantage weight1.000

Model configdata/models/SCO1.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 / A2.06 / 1.24

Relative strengths and form diagnostics

Celtic

Home attack strength1.351

Home defense weakness0.672

Base lambda home1.77

Attack form xGF level2.936

Defense form xGA level (lower is better)0.982

Opponent-adjusted attack residual+1.451

Opponent-adjusted defense residual (higher is better)+0.408

Opponent-adjustment weight0.000

Model attack signal+1.499

Model defense signal+0.456

Opponent blend0.000

Form matches 5 / 105 / 10

Last 5 xGF / xGA / xGD3.02 / 1.00 / +2.02

Last 10 xGF / xGA / xGD2.78 / 0.94 / +1.83

Rangers

Away attack strength1.295

Away defense weakness0.822

Base lambda away1.11

Attack form xGF level2.530

Defense form xGA level (lower is better)1.051

Opponent-adjusted attack residual+1.011

Opponent-adjusted defense residual (higher is better)+0.308

Opponent-adjustment weight0.000

Model attack signal+1.092

Model defense signal+0.386

Opponent blend0.000

Form matches 5 / 105 / 10

Last 5 xGF / xGA / xGD2.63 / 0.99 / +1.63

Last 10 xGF / xGA / xGD2.35 / 1.16 / +1.19

Historical data coverage

Celtic

Total matches158

Home / away matches79 / 79

xG matches44 / 158

xG home / away coverage22 / 79 · 22 / 79

Effective total23.97

Effective attack / defense23.97 / 23.97

Effective home matches11.99

Effective home attack / defense11.99 / 11.99

Rangers

Total matches158

Home / away matches79 / 79

xG matches44 / 158

xG home / away coverage22 / 79 · 22 / 79

Effective total23.91

Effective attack / defense23.91 / 23.91

Effective away matches12.01

Effective away attack / defense12.01 / 12.01

Historical data coverage

73.12%

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 928 historical matches.

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

Total goals log-loss vs baseline-0.03%

1X2 log-loss vs baseline+3.84%

BTTS Brier vs baseline+0.59%

Over 2.5 Brier vs baseline+0.21%

Full backtest metrics

Model

MAE total goals1.3056

Exact score log-loss3.0349

Total goals Brier0.8278

Total goals log-loss1.8853

1X2 Brier0.6143

1X2 log-loss1.0247

BTTS Brier0.2503

Over 1.5 Brier0.1702

Over 2.5 Brier0.2495

Over 3.5 Brier0.2211

League baseline

MAE total goals1.3056

Exact score log-loss3.0905

Total goals Brier0.8281

Total goals log-loss1.8848

1X2 Brier0.6438

1X2 log-loss1.0656

BTTS Brier0.2518

Over 1.5 Brier0.1700

Over 2.5 Brier0.2500

Over 3.5 Brier0.2205

Calibration diagnostics

Home win · ECE 10.44%

10- 20%  n=20   avg= 16.0% actual= 25.0%
20- 30%  n=38   avg= 25.9% actual= 18.4%
30- 40%  n=245  avg= 37.1% actual= 29.0%
40- 50%  n=536  avg= 42.9% actual= 54.7%
50- 60%  n=39   avg= 53.2% actual= 43.6%
60- 70%  n=33   avg= 63.8% actual= 75.8%
70- 80%  n=13   avg= 75.3% actual= 84.6%
80- 90%  n=4    avg= 84.8% actual=100.0%

Draw · ECE 5.62%

 0- 10%  n=1    avg=  9.0% actual=  0.0%
10- 20%  n=22   avg= 16.9% actual=  9.1%
20- 30%  n=862  avg= 28.4% actual= 22.7%
30- 40%  n=43   avg= 31.2% actual= 34.9%

Away win · ECE 7.60%

 0- 10%  n=18   avg=  7.0% actual=  5.6%
10- 20%  n=55   avg= 15.7% actual= 20.0%
20- 30%  n=454  avg= 27.5% actual= 20.7%
30- 40%  n=341  avg= 32.6% actual= 42.5%
40- 50%  n=32   avg= 44.9% actual= 46.9%
50- 60%  n=16   avg= 54.8% actual= 50.0%
60- 70%  n=11   avg= 63.3% actual= 72.7%
70- 80%  n=1    avg= 70.9% actual=  0.0%

BTTS · ECE 3.55%

30- 40%  n=3    avg= 38.1% actual=  0.0%
40- 50%  n=66   avg= 46.6% actual= 47.0%
50- 60%  n=823  avg= 55.0% actual= 51.5%
60- 70%  n=36   avg= 62.6% actual= 55.6%

Over 2.5 · ECE 6.16%

30- 40%  n=15   avg= 36.8% actual= 26.7%
40- 50%  n=575  avg= 48.7% actual= 55.0%
50- 60%  n=295  avg= 52.0% actual= 58.3%
60- 70%  n=38   avg= 64.3% actual= 63.2%
70- 80%  n=5    avg= 74.1% actual= 60.0%

Walk-forward - latest periods

Period	N	Model 1X2 LL	Baseline 1X2 LL	Improvement
2025-09	15	1.1264	1.1263	-0.01%
2025-10	23	0.9993	1.0516	+4.97%
2025-11	22	0.9961	1.1143	+10.61%
2025-12	34	0.9362	1.0301	+9.11%
2026-01	25	0.8692	1.0779	+19.36%
2026-02	29	0.9587	1.0372	+7.57%
2026-03	16	1.1289	1.0485	-7.67%
2026-04	18	0.9330	1.0215	+8.66%
2026-05	24	1.0772	1.0639	-1.26%
2026-07	1	1.5515	1.2725	-21.93%
2026-08	18	0.9447	1.0669	+11.46%
2026-09	17	0.9311	1.0794	+13.74%

Football is data.

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

Back to Matches Follow on X

Celtic vsRangers

Match probabilities.

Most likely scorelines.

Strength and recent form.

Celtic

Rangers

Historical matchup context.

Total goals profile.

How the model reached the estimate.

Celtic

Rangers

Celtic

Rangers

Historical model performance.

Home win · ECE 10.44%

Draw · ECE 5.62%

Away win · ECE 7.60%

BTTS · ECE 3.55%

Over 2.5 · ECE 6.16%

Football is data.

Celtic vs
Rangers