Improving Match xG Infographics

The repository consists of the code used in my blog post about the poisson distribution of Expected Goals in single football matches.

http://roensholt-stats.com/2021-04-25-Hvorfor-expected-goals-ikke-fort%C3%A6ller-resultatet-af-en-kamp/

This project I will make my attempt to improve the common used Expected Goals (xG) race chart from a single game. This will be done by adding more context like win, loss and draw probabilities, the 3 most probable scorelines and the probability for goals for each team.

• utils contains utility functions and scripts for calculations, cleaning and metatdata
• output contains all the visualizations
• team_logos contains all Premier League team badges
• xG_Infographic.ipynb code for the complete xG Infographic
• goal_probabilities_plot.py code for goal probability viz
• win_probability_plot.py code for win probability viz
• xG_racechart_plot.py code for xG racechart viz

xG Infographic

xG infographic breakdown

xG racechart

Goal probabilities for each team

Bar chart of 0-9 goals for each team, where the actual goal(s) scored is filled.

Win probability

The probability of each team winning or a draw result based on the shots they have taken.