The bar is the model's probability; the caret on the same scale is the market's, with the margin removed. The distance between them is the disagreement.
OutcomeModel fill · market caretProbFairΔ mkt
SUN
16.3%6.14−9.8 pp
Draw
38.0%2.63+12.0 pp
CHE
45.7%2.19−2.1 pp
Ch 02 · Book
27 markets
dixon_coles xi001-20261102
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
Sunderland16.3%6.14
Draw38.0%2.63
Chelsea45.7%2.19
Double chance
Sunderland or draw54.3%1.84
Sunderland or Chelsea62.0%1.61
Draw or Chelsea83.7%1.20
Draw no bet
Sunderland26.3%3.80
Chelsea73.7%1.36
A draw returns the stake, so these prices exclude it.
Win to nil
Sunderland13.0%7.71
Chelsea37.3%2.68
Clean sheet
Sunderland38.0%2.63
Chelsea62.3%1.60
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.
Unavailable players22% home against 34% away xG missing
Favours SUN. 6 out and 5 doubtful for SUN, 9 out and 5 doubtful for CHE; measured as the share of recent expected goals those players contributed
Recent form0.20 against 1.40 points per game
Favours CHE. last 5 and 5 completed matches: DLLLL against LWWLD
Underlying expected goals-0.52 net xG per match
Favours CHE. SUN -0.92, CHE -0.39 across the form window — results are noisier than the chances behind them
Home advantage+60 Elo equivalent
Favours SUN. 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
Team rating gap-56 Elo
Favours CHE. SUN 1510 against CHE 1566, 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
Ch 04 · Comparison
Both sides
SUN
DLLLL
12 Sept – 25 Oct
CHE
DLWWL
12 Sept – 24 Oct
0.20Points per game1.40
3Goals for9
11Goals against8
7.17Expected goals5.53
1,510Elo1,566
Head to head over 2 meetings in the seasons held (Premier League from 2021-22, Championship from 2024-25): SUN 2, drawn 0, CHE 0. Cup ties and earlier seasons are not included, so a full-history site will show more.
Ch 05 · Market
Bookmaker prices
Real bookmaker prices, collected from published market data.
Market
Selection
Best
Book
Implied
N
h2h
away
2.15
1xBet
46.5%
40
h2h
draw
3.80
Betfair
26.3%
40
h2h
home
3.85
Betfair
26.0%
40
Ch 06 · Availability
Sunderland
21.7% xG out
Share of attacking output is what moves the number, so it is shown rather than implied.
BallardDEFinjured0%2.3% xG
Hamstring injury - Unknown return date
DansoDEFsuspended0%2.3% xG
Suspended until 31 Oct
MundleMIDinjured0%1.3% xG
Knee injury - Unknown return date
ReinildoDEFsuspended0%0.2% xG
Suspended until 10 Oct
AldereteDEFsuspended0%0.2% xG
Suspended until 21 Nov
AdingraMIDinjured0%
Ankle injury - Unknown return date
BrobbeyFWDdoubtful75%26.1% xG
Hamstring injury - 75% chance of playing
IsidorFWDdoubtful50%12.7% xG
Knock - 50% chance of playing
MukieleDEFdoubtful75%6.2% xG
Concussion - 75% chance of playing
MeunierDEFdoubtful75%3.7% xG
Groin injury - 75% chance of playing
and 1 more
Ch 06 · Availability
Chelsea
34.3% xG out
Share of attacking output is what moves the number, so it is shown rather than implied.
João PedroFWDinjured0%22.7% xG
Thigh injury - Unknown return date
LacroixDEFinjured0%3.0% xG
Hamstring injury - Unknown return date
JamesDEFinjured0%2.5% xG
Knee injury - Unknown return date
ChavarriaDEFinjured0%0.3% xG
Unspecified injury - Expected back 31 Oct
HatoDEFinjured0%0.1% xG
Unspecified injury - Unknown return date
MartinezGKPinjured0%
Head injury - Unknown return date
AnselminoDEFinjured0%
Unspecified injury - Unknown return date
EmeghaFWDinjured0%
Hamstring injury - Unknown return date
GittensMIDinjured0%
Ankle injury - Unknown return date
PalmerMIDdoubtful75%18.7% xG
Muscular injury - 75% chance of playing
and 4 more
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 08 · Briefing
Written summary
generated text
Written by gpt-5.6-terra from the numbers above, which it cannot add to or override.
Chelsea have the highest model win probability at 46.1%, with the draw at 38.1% and Sunderland at 15.8%. Expected goals are low at 0.47 for Sunderland and 1.0 for Chelsea. The model puts over 2.5 goals at 18.2% and both teams scoring at 24.2%, pointing to a low-scoring profile.
The market rates Chelsea at 48.5%, Sunderland at 25.8% and the draw at 25.7%. It is more favourable to Sunderland and Chelsea than the model, while the model is markedly more favourable to the draw. The scorecard says this model does not beat the closing market.
Sunderland’s last five brought 0 wins, 1 draw and 4 losses, with 3 goals scored and 11 conceded. Chelsea have 2 wins, 1 draw and 2 losses in five. Sunderland won both listed head-to-head meetings, 2-1 at home and 2-1 away. There are no measured missing players, but 11 Sunderland and 14 Chelsea absentees have unmeasured output shares, so their attacking impact is not known.
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