Implementation of the following paper :A profitable model for predicting the over/under market in football , Edward Wheatcroft
We implemented a model based on Elo Rating of the Football teams (General Attacking Performance), The Ratings may take into different measure of performance which made the model simple and robust
We used the datasets from https://www.football-data.co.uk/ on The 2021/2022 and 2022/2023 Premier League Seasons
The Class dataset is implemented in dataset.py
The GAP ratings class and updates were implemented in team.py
Two Betting Strategies were implemented the level stakes betting and the Kelly criterion betting.
We first initialize GAP parameters with (1,0.5,0.5) for making predictions on the 2021/2022 Season,then optimizing the GAP ratings parameters with minimazing the ignorance score(Log Loss) over the past season, and then making predictions for 2022/2023 Season.
The model yields 7% net return profit on the 2023 season and this without optimizing the parameters, and without using maximum odds.
Using the Bivariate Poisson Process for predicting Game results
Implementing "Forecasting Football Match Results in National League Competitions Using Score-Driven Time Series Models(Koopman)" for more robust predictions
Improving the betting Strategies using risk minimalization