Manchester City vs Sunderland prediction & probabilities (2026-09-20)

Premier League

Manchester City vs Sunderland Correct Score

2026-09-20 · 14:00
Correct Score1-0Most likely · 12.1%
Best predictionMost likely · 12.1% · 1-0No single outcome dominates, read the distribution. Calibrated, not a tip.
ConvictionOpen

Expert AI analysis

The model makes Most likely · 12.1% at 1-0, its strongest read on the Correct Score market. The goals picture is built on form: Manchester City have averaged 1.8 scored and 0.8 conceded, Sunderland 1.0 and 1.2 over their recent games. That fits a balanced goals environment: 51.2% for over 2.5 and 76.0% for at least two goals. Their last 5 meetings have averaged 1.8 goals a game. These are the model’s own calibrated numbers, not betting advice.

Correct Score

Most likely1-0 12.1%
Scoreline Probability
1-0 12.1%
2-0 12.0%
1-1 10.7%
2-1 9.7%
3-0 7.6%
0-0 7.1%
3-1 6.2%
0-1 4.8%

Score probability grid

Manchester City \ Sunderland 0 1 2 3 4 5
0 7% 5% 2% 1% · ·
1 12% 11% 4% 1% · ·
2 12% 10% 4% 1% · ·
3 8% 6% 3% 1% · ·
4 4% 3% 1% · · ·
5 1% 1% · · · ·

Rows = Manchester City goals, columns = Sunderland. The most likely scoreline is 1-0 at 12.1%. No exact score is ever truly likely; the value is in the full distribution the model assigns. The brightest cells are where the match is most likely to land.

Goals distribution

The model’s probability for the total number of goals, the shape behind every goals market. The 5+ bar groups every result with five or more goals.

7.1%

0

16.9%

1

24.8%

2

22.1%

3

15.0%

4

12.7%

5+

In-depth analysis

66.5%Favoured: Manchester City
47.8%Both Teams to Score
51.2%Over 2.5 goals

On the goals side, the model sees 51.2% for three or more goals and 76.0% for at least two, painting a balanced picture. Given Manchester City favoured at 66.5%, a one-sided game can cap the goals at one end.

Every figure is the model’s own calibrated read, not advice. We grade every call after the match.

Model lenses

Different analytical lenses our model applies to the same calibrated data, not separate tipsters, and never a tip.

The CalibratorRaw model read: 1-0 on Correct Score, graded against real out-of-sample results, no odds attached.
The TacticianSees a balanced game from the 51.2% over 2.5 read.
The ScoutManchester City are the model’s pick on the result at 66.5%, context for how the goals may flow.
The RealistThis one sits near a coin-flip, the honest read is low conviction, no strong lean.

Key players

E

Erling HaalandManchester City · 3.61 shots/90
To score 48.6%

W

Wilson IsidorSunderland · 3.53 shots/90
To score 17.5%

This season’s squad only, rated on recent form; a watch list, not a confirmed lineup.

Form & head-to-head

Manchester CityLWWWW
SunderlandWLWDL

Last 5 meetings: Manchester City 4 · 1 draws · Sunderland 0 · 1.8 goals/game.

FAQ

What is the most likely score in Manchester City vs Sunderland?
The model's most likely scoreline for Manchester City vs Sunderland is 1-0 at 12.1%.

How is this probability calculated?
We fit a goals and markets model (Dixon-Coles plus an xG strength model and an ensemble) on real match data, then calibrate the output against out-of-sample results. It is a calibrated model probability, not an odds-derived tip.

Is this a betting tip?
No – these are calibrated model probabilities, graded against real results. We publish our full track record, including the misses, and never claim to beat the closing market.

Overall model accuracy

We don’t yet publish a graded Correct Score-specific track record, so this is our overall accuracy across every 1X2 call, graded out-of-sample, updated daily.

0.0096Calibration error (1X2)
47.7%Top-pick hit rate (1X2)
55,405Matches graded

Predictions & track record updated 16 Sep 2026. See the full, dated track record →

More Premier League, Correct Score

All Correct Score predictions →

Related

Model probabilities, calibration-graded, not a tip. We publish our track record, including the misses. 18+. For information only, not betting advice. If gambling is a problem, visit BeGambleAware.