Skip to content

Monksc/EnglishPremierLeaguePredictions

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

28 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

🚀 Overview

This project ingests historical match data and generates predictive insights for football matches, including:

  • Win / Draw / Loss probabilities
  • Team performance rankings
  • Match outcome predictions across multiple leagues

The goal is to build a system that can evaluate team strength and produce actionable probability estimates based on real-world data.


🧠 Approach

The system combines multiple techniques:

  • Statistical modeling (including ranking systems and probability estimation)
  • Machine learning models (TensorFlow-based)
  • Data-driven feature generation from historical match results
  • Evaluation against test datasets to measure prediction accuracy

📊 Data Pipeline

  • Collects match data across multiple leagues (EPL, Bundesliga, La Liga, Serie A)
  • Processes and normalizes data into structured formats
  • Generates training and test datasets
  • Outputs predictions in JSON format for downstream use

⚙️ Features

  • Multi-league support
  • Automated data processing and model training
  • Prediction output for match outcomes
  • Error evaluation and model validation tools

📈 Example Output

  • Probability of win/draw/loss for a given match
  • Team strength rankings
  • Model performance metrics on test data

🛠 Tech Stack

  • Python
  • TensorFlow
  • Data processing pipelines
  • JSON-based output for integration with applications

Use Cases


Page Rank

To do a page rank to see predictions for ranking of teams and predictions for end of season type in

python3 epl.py <csv file for schedule> <week>

Predicting Games

Currently it doesnt predict score. You can do a page rank to see who is the better team and finally adding in what percent chance does a team to have to win.


Neural Network

Uses several variations to page rank, wins, tie, loss, from home to away games to see the chance a team has to win. You may have to change around some code in predictions_tensorflow.py.

python3 predictions_tensorflow.py

Folder 'old_data'

This folder contains data from other leagues from previous years. The name format usually goes by "{name of the league}-{year season started}"

May need to run command below

cd old_data/jackpotdata/all/
tar -xf all.csv.tar.gx

To Create the data

python3 make_data_like_my.py

To Train Model

You may have to change around some code in predictions_tensorflow.py.

python3 predictions_tensorflow.py

About

Predicts Matches for the English Premier League. Can be changed for other leagues.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages