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
FUL
38.0%2.63−5.2 pp
Draw
30.0%3.33+4.1 pp
NEW
31.9%3.13+1.2 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
Fulham38.0%2.63
Draw30.0%3.33
Newcastle31.9%3.13
Double chance
Fulham or draw68.1%1.47
Fulham or Newcastle70.0%1.43
Draw or Newcastle62.0%1.61
Draw no bet
Fulham54.4%1.84
Newcastle45.6%2.19
A draw returns the stake, so these prices exclude it.
Win to nil
Fulham22.2%4.51
Newcastle18.4%5.44
Clean sheet
Fulham32.9%3.04
Newcastle29.1%3.43
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.
Underlying expected goals+0.83 net xG per match
Favours FUL. FUL +0.31, NEW -0.52 across the form window — results are noisier than the chances behind them
Unavailable players4% home against 12% away xG missing
Favours FUL. 2 out and 6 doubtful for FUL, 5 out and 7 doubtful for NEW; measured as the share of recent expected goals those players contributed
Home advantage+60 Elo equivalent
Favours FUL. 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
Recent form1.60 against 1.20 points per game
Favours FUL. last 5 and 5 completed matches: WWLDD against WLLWL
Team rating gap-25 Elo
Favours NEW. FUL 1517 against NEW 1542, 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
FUL
WWLDD
12 Sept – 24 Oct
NEW
LWLLW
14 Sept – 25 Oct
1.60Points per game1.20
4Goals for6
3Goals against10
7.26Expected goals6.60
1,517Elo1,542
Head to head over 8 meetings in the seasons held (Premier League from 2021-22, Championship from 2024-25): FUL 3, drawn 0, NEW 5. 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
3.30
Unibet (SE)
30.3%
42
h2h
draw
3.84
1xBet
26.0%
42
h2h
home
2.30
Betfair
43.5%
42
Ch 06 · Availability
Fulham
3.6% xG out
Share of attacking output is what moves the number, so it is shown rather than implied.
SessegnonDEFsuspended0%1.9% xG
Suspended until 7 Nov
TeteDEFinjured0%
Concussion - Expected back 10 Oct
RobinsonDEFdoubtful75%2.9% xG
Ankle injury - 75% chance of playing
AffengruberDEFdoubtful75%1.9% xG
Unspecified injury - 75% chance of playing
LarssonMIDdoubtful75%1.4% xG
Groin injury - 75% chance of playing
Smith RoweMIDdoubtful75%0.4% xG
Knock - 75% chance of playing
CairneyMIDdoubtful75%0.2% xG
Knee injury - 75% chance of playing
De FougerollesDEFdoubtful75%
Unspecified injury - 75% chance of playing
Ch 06 · Availability
Newcastle
12.4% xG out
Share of attacking output is what moves the number, so it is shown rather than implied.
J.RamseyMIDinjured0%1.9% xG
Thigh injury - Unknown return date
BurnDEFinjured0%0.9% xG
Ankle injury - Expected back 12 Oct
N.GonzalezMIDinjured0%0.5% xG
Concussion - Expected back 12 Oct
LivramentoDEFinjured0%
Knee injury - Unknown return date
JaouenGKPinjured0%
Hamstring injury - Unknown return date
OsulaFWDdoubtful50%8.4% xG
Ankle injury - 50% chance of playing
Fernandez-PardoFWDdoubtful75%5.9% xG
Illness - 75% chance of playing
ElangaMIDdoubtful75%4.6% xG
Hamstring injury - 75% chance of playing
TouréMIDdoubtful75%4.1% xG
Unspecified injury - 75% chance of playing
DedićDEFdoubtful50%1.8% xG
Unspecified injury - 50% chance of playing
and 2 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.
The model rates Fulham at 42.0%, the draw at 28.9% and Newcastle at 29.1%. Its goal estimates are 1.37 for Fulham and 1.1 for Newcastle. Both teams to score is 50.6%, while over 2.5 goals is 44.9%, pointing to a relatively tight scoring profile.
Fulham have taken 1.6 points per game across 5 matches, with 7.3 xG created and 5.7 conceded. Newcastle have 1.2 points per game, creating 6.6 xG and allowing 9.2. Newcastle’s Elo is 1542 against Fulham’s 1517, while the head-to-head record is 5 Newcastle wins and 3 Fulham wins from 8.
The market is more favourable to Fulham, at 45.6% versus the model’s 42.0%. It gives the draw 26.7% and Newcastle 27.7%, both below the model’s figures. The model does not beat the closing market on the supplied scorecard. No measured absences are listed, but there are 8 unmeasured Fulham absentees and 12 for Newcastle.
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 drawmodel 30.0% vs 3.84 (1xBet)+15.3 ppstake 1.3%medium confidence
h2h awaymodel 31.9% vs 3.30 (Unibet (NL))+5.3 ppstake 0.6%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