Guide · updated 2026-11-20

The Dixon–Coles model, explained

The Dixon–Coles model gives every team an attack and a defence rating, turns them into expected goals for each side, and from those builds a probability for every possible score. Adding up the scores gives home, draw and away probabilities.

Interactive score model explorer

Explore how expected goals shape the scoreline matrix

Adjust pre-match expected goals and the Dixon–Coles low-score dependence term (ρ) to see how a statistical score model builds exact scoreline, 1X2, over/under 2.5 and clean-sheet probabilities.

Try:
Manchester City win
59.5%
Draw
23.3%
Fulham win
17.2%
Over 2.5 goals
53.3%Under 46.7%
Both teams score
51.7%No 48.3%
Clean sheet
40% · 15%Manchester City · Fulham

Scoreline probability grid

1–1 11.1% · most likely

Columns: fulham goals → · Rows: manchester city goals ↓
012345
06.64.92.50.80.20.0
110.811.14.81.50.30.1
210.89.94.61.40.30.1
36.86.22.90.90.20.0
43.22.91.40.40.10.0
51.21.10.50.20.00.0

Draws (diagonal) Most likely scoreBelow the diagonal: manchester city wins · above: fulham wins

Cells show percent. The 0–5 grid covers 98.7% of the probability mass; the outcome totals above sum over the full 0–12 goal grid so they add to 100.0%. Darker cells are more likely.

Want to see this on a real fixture? Open the Manchester City vs Fulham match page for that match's published probabilities. To see how a result changes the title race or relegation fight, use the season simulator.

From ratings to expected goals

Each side's goals are modelled as a Poisson count. The home side's expected goals rise with its attack rating, the away side's defensive weakness and a league-wide home advantage; the away side's expected goals work the same way without the home term.

The low-score correction

Dixon and Coles added a term that nudges the probabilities of 0–0, 1–0, 0–1 and 1–1 up or down, because real football produces these results at rates a pure Poisson model misses. Setting the term to zero gives back the independent Poisson model.

How our version is set up

  • Training data: Premier League full-time results since 2005-06, with only matches finished before the prediction.
  • Time weighting: a one-year half-life, so last season counts half as much as this one.
  • Shrinkage: ratings are pulled towards average when a team has little data, and newly promoted clubs start below average, as promoted sides historically have.
  • Player availability: expected goals are adjusted for injured, suspended and doubtful players listed by the Fantasy Premier League feed.

Full details are on the methodology page.

Example: Manchester City vs Fulham

The model expects 1.89 goals for Manchester City and 0.92 for Fulham. Its most likely single score is 1–1 at 11.0%, and summing every scoreline gives Manchester City 59.5%, draw 23.1% and Fulham 17.3%. See the full match page.

How good is it?

On 1,140 held-out matches the model's most likely result happened 51.8% of the time and its log loss was 0.993. On the same matches, simply backing the club higher in the table was right 49.0% of the time: the model gets about 3 more results right per 100 matches. The model record explains the comparison.

Common questions

What is the Dixon–Coles model?

A football score model published by Mark Dixon and Stuart Coles in 1997. It treats each team's goals as Poisson counts driven by attack and defence ratings and home advantage, with a correction for low scores such as 0–0 and 1–1.

Why not a plain Poisson model?

Independent Poisson counts tend to misprice low-scoring results. The Dixon–Coles correction adjusts the probabilities of 0–0, 1–0, 0–1 and 1–1, which matters because those are among the most common scores.

How often is the model refitted?

Ratings are refitted every time the site updates, using only matches that had already finished. Its settings are checked once a month on held-out matches.

Free to use: one small ask

The data, method and results on this page are free. If they helped you, please link to or cite this page so readers can check the work themselves.

Football Edge Lab (2026). The Dixon–Coles model, explained. https://footballedgelab.com/guides/dixon-coles-model

Get new forecasts and studies as they are published: RSS feed.

More guides

Probabilities come from statistical score models and describe uncertainty, not guaranteed outcomes.