Fixture 49scheduledGW 5

Man City v Sunderland

Sun 20 Sept 13:00Zmodel prices only

Expected goals1.94 — 0.82

Expected points2.12 — 0.66

Over 2.552.2%fair 1.92

Both score48.5%fair 2.06

Most likely2-011.9%

Markets priced27

Ch 01 · Outcome

Match result

No bookmaker prices are held for this fixture, so there is no caret to compare against. Fair price is 1 / probability with no margin, and is not a price anyone is offering.

OutcomeModel probabilityProbFairΔ

MCI

63.3%1.58–

Draw

22.2%4.50–

SUN

14.4%6.92–

Ch 02 · Book

27 markets

dixon_coles xi001-20260903

Model fair prices, computed as 1 / probability from a Dixon-Coles score matrix. No bookmaker margin is included, and no bookmaker is quoting these. A real price will be shorter than every number here.

Match result

Man City63.3%1.58

Draw22.2%4.50

Sunderland14.4%6.92

Double chance

Man City or draw85.6%1.17

Man City or Sunderland77.8%1.29

Draw or Sunderland36.7%2.73

Draw no bet

Man City81.4%1.23

Sunderland18.6%5.38

A draw returns the stake, so these prices exclude it.

Win to nil

Man City37.3%2.68

Sunderland7.4%13.51

Clean sheet

Man City44.1%2.27

Sunderland14.2%7.04

Ch 03 · Reasoning

What is driving this

computed, not generated

Computed arithmetic, not generated text. Each driver is a stored quantity: the Elo difference replayed from results, points per game over the last five matches, rolling expected goals, and each absent player's share of team expected goals.

Team rating gap+192 Elo

Favours MCI. MCI 1728 against SUN 1536, computed from stored results; an Elo-driven model scores RPS around 0.2015 on this league against roughly 0.194 for de-vigged closing odds

Home advantage+60 Elo equivalent

Favours MCI. the football default is around 100 Elo points, but Premier League home advantage fell during the crowdless 2020/21 season and settled lower afterwards, so the ratings carry a smaller term

Underlying expected goals+0.25 net xG per match

Favours MCI. MCI +0.69, SUN +0.44 across the form window — results are noisier than the chances behind them

Unavailable players0% home against 0% away xG missing

Favours MCI. 1 out and 1 doubtful for MCI, 1 out and 1 doubtful for SUN; measured as the share of recent expected goals those players contributed

Recent form2.00 against 2.00 points per game

last 5 and 5 completed matches: WWLDW against WLWWD

Ch 04 · Comparison

Both sides

MCI

WWLDW

13 May – 28 Aug

SUN

DWWLW

9 May – 30 Aug

2.00Points per game2.00

11Goals for7

5Goals against4

9.65Expected goals7.59

1,728Elo1,536

Head to head over 2 meetings in the seasons held (Premier League from 2021-22, Championship from 2024-25): MCI 1, drawn 1, SUN 0. Cup ties and earlier seasons are not included, so a full-history site will show more.

Ch 05 · Market

Bookmaker prices

No bookmaker prices held for this fixture. Every price shown is the model's own fair price, meaning 1 / probability with no margin. Nobody is offering these.

Nothing held. football-data.co.uk publishes weekend Premier League prices on Friday afternoon, and The Odds API needs a key. Historical fixtures already carry closing odds, so the comparison works there.

Ch 06 · Availability

Man City

0.0% xG out

Share of attacking output is what moves the number, so it is shown rather than implied.

DokuMIDinjured0%

Calf injury - Expected back 5 Sep

Matheus N.DEFdoubtful75%

Unspecified injury - 75% chance of playing

Ch 06 · Availability

Sunderland

0.3% xG out

Share of attacking output is what moves the number, so it is shown rather than implied.

AdingraMIDinjured0%

Ankle injury - Unknown return date

DiarraMIDdoubtful75%1.1% xG

Unspecified injury - 75% chance of playing

Ch 07 · Simulation

Re-price without a player

Removing a player scales that side's expected goals by their share of recent attacking output, then rebuilds the score matrix. Every market moves together.

Ch 10 · Calibration

How this model actually scores

Dixon-Coles bivariate Poisson fitted to Premier League results with exponential time decay. Statistical model, no language model involved.

loses to closing market2,281 out-of-sample fixtures

Metric	Model	Market	Δ
Ranked probability score	0.2047	0.1965	+0.0082
Log loss	0.9882	0.9623	+0.0259
Brier	0.5883	0.5711	+0.0172
Accuracy	52.5%	54.8%	−2.3 pp

Lower is better for the first three. These probabilities are for reading a fixture, not for expecting profit against closing prices. What these numbers mean