Football Prediction Model 4.0

A python machine learning model to predict the outcomes of EPL, La Liga, BundesLiga, Ligue One, and Serie A games.

Overview

A machine learning model designed to predict the outcome of football (soccer) matches. The model has been trained on data (140 data points) from the last 5 years, 2018-2023, predicting the expected goals for and against. Using a combination of those predictions and ELO ratings, frm current data, I then calculate offensive and defensive strengths for each team, with which I can predict the outcome of upcoming games, and determine whether my odds have an edge over Vegas.

This script also contains code for appending these predictions to a firebase real-time database.

(back to top)

Built With

• Python

(back to top)

Dependencies

• numpy
• pandas
• selenium
• webdriver_manager
• BeautifulSoup
• seaborn
• matplotlib.pyplot
• sklearn
• firebase_admin

(back to top)

Acknowledgements

Data sourced from TheOddsAPI, FiveThirtyEight and FBref.

(backop)