Models
Does Blending Elo With the Market Help?
We mixed Elo forecasts into margin-free market probabilities at different weights and scored 2,660 Premier League matches. Every blend made the forecasts worse.
Read the tutorial →Poisson, Dixon-Coles, Elo and machine-learning models explained and implemented, with their assumptions and limits stated plainly.
Each model makes assumptions about how goals happen. This section implements the main ones in Python and explains what they assume, where they work and where they break.
To see the Poisson model in action, try the Poisson match predictor.
We mixed Elo forecasts into margin-free market probabilities at different weights and scored 2,660 Premier League matches. Every blend made the forecasts worse.
Read the tutorial →A tested Elo implementation: rate every Premier League team since 2015-16, add home advantage and turn the ratings into 1X2 probabilities.
Read the tutorial →