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
COV
24.8%4.03+2.7 pp
Draw
25.9%3.85+1.4 pp
BHA
49.2%2.03−4.2 pp
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.
Thin sample
Coventry City has only 2 completed matches in the training window, so its fitted strength is imprecise. Treat the long prices in this book as arithmetic rather than information: a real bookmaker would quote a much shorter price on the outsider.
Matches behind each fit: Coventry City 2 · Brighton 149
Match result
Coventry City24.8%4.03
Draw25.9%3.85
Brighton49.2%2.03
Double chance
Coventry City or draw50.8%1.97
Coventry City or Brighton74.1%1.35
Draw or Brighton75.2%1.33
Draw no bet
Coventry City33.5%2.98
Brighton66.5%1.50
A draw returns the stake, so these prices exclude it.
Win to nil
Coventry City12.8%7.81
Brighton27.5%3.64
Clean sheet
Coventry City20.0%4.99
Brighton34.7%2.88
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 players0% home against 14% away xG missing
Favours COV. 3 out and 0 doubtful for COV, 4 out and 3 doubtful for BHA; measured as the share of recent expected goals those players contributed
Underlying expected goals-0.77 net xG per match
Favours BHA. COV -0.11, BHA +0.65 across the form window — results are noisier than the chances behind them
Team rating gap-76 Elo
Favours BHA. COV 1486 against BHA 1562, 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
Recent form1.80 against 1.20 points per game
Favours COV. last 5 and 5 completed matches: LLWWW against LWLLW
Home advantage+60 Elo equivalent
Favours COV. 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
Ch 04 · Comparison
Both sides
COV · incl. Championship
LLWWW
21 Apr – 29 Aug
BHA
WLLWL
9 May – 30 Aug
1.80Points per game1.20
12Goals for10
6Goals against8
1.74Expected goals (2 of 5)10.95
1,486Elo1,562
Ch 05 · Market
Bookmaker prices
Real bookmaker prices, collected from published market data.
Market
Selection
Best
Book
Implied
N
h2h
away
1.83
1xBet
54.6%
33
h2h
draw
4.00
Betfred (UK)
25.0%
33
h2h
home
4.70
Tipico
21.3%
33
Ch 06 · Availability
Coventry City
0.0% xG out
Share of attacking output is what moves the number, so it is shown rather than implied.
Kesler-HaydenDEFinjured0%
Hamstring injury - Unknown return date
WoolfendenDEFinjured0%
Knee injury - Unknown return date
WrightFWDinjured0%
Thigh injury - Unknown return date
Ch 06 · Availability
Brighton
14.4% xG out
Share of attacking output is what moves the number, so it is shown rather than implied.
FergusonFWDinjured0%
Ankle injury - Expected back 10 Oct
MintehMIDinjured0%
Leg injury - Expected back 28 Nov
MitomaMIDinjured0%
Hamstring injury - Unknown return date
TzimasFWDinjured0%
Knee injury - Unknown return date
HinshelwoodMIDdoubtful50%27.0% xG
Unspecified injury - 50% chance of playing
AyariMIDdoubtful75%2.1% xG
Ankle injury - 75% chance of playing
GeorginioFWDdoubtful75%1.5% 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 09 · Divergence
Where the model differs from a price
Shown with the model's own record, because that is what decides how much these are worth.
h2h homemodel 24.8% vs 4.70 (Tipico)+16.7 ppstake 1.1%medium confidence
h2h drawmodel 25.9% vs 4.00 (Betfred (UK))+3.8 ppstake 0.3%low confidence
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