Crystal Palace v Liverpool

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.

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.

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.

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.

Written by gpt-5.6-terra from the numbers above, which it cannot add to or override.

Liverpool have the model’s highest win probability at 52.7%, with the draw at 37.0% and Crystal Palace at 10.2%. The goal model is restrained: 0.33 expected goals for Palace and 1.09 for Liverpool. Over 2.5 goals is 17.2%, while both teams scoring is 19.3%. Recent form supports Liverpool’s stronger rating. They have 2.0 points per game across five matches, compared with Palace’s 0.8, and their Elo is 1644 against 1494. Palace have conceded 11 goals and allowed 10.4 xG in those five games. Liverpool have conceded four, with 5.6 xG against. Liverpool also lead the last 10 meetings with five wins to Palace’s two, alongside three draws. The market is more favourable to Palace and Liverpool than the model, assigning 19.8% to a Palace win and 57.9% to Liverpool, versus the model’s 10.2% and 52.7%. It is far less favourable to the draw at 22.3%. No named absentees have measured output shares, but there are 12 unmeasured Palace absentees and 15 for Liverpool, so their effect is unknown. The scorecard does not beat the market.

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