Skip to content

Latest commit

 

History

History
36 lines (25 loc) · 1.73 KB

README.md

File metadata and controls

36 lines (25 loc) · 1.73 KB

FIFA-World-Cup-Simulation

Model Predicting who will be the winner of FIFA World Cup 2022

In this notebook I implemented a model that predicts who will be the winner of the 2022 FIFA World Cup.

It includes:

  • Calculating xG of a team by avering their score in the latest matches
  • Adjusting the xG of a team according to the previous opponent's ranking
  • Adjusting the xG of a team according with who they are playing in the next match
  • Adding time decay (latest matches have heavier weights)
  • Simulating matches based on the xG of two teams
  • Implementing the whole structure of the tournament (group stage and knockout stage)

Model is fully modifiable, that means:

  • We can use that model in future tournaments - we only need to download the dataset with newest matches and ranking
  • We can predict the winner of one match - with or without draws
  • We can change parameters like date from which matches are taken into account
  • We can modify the amount of iterations

We can simulate the whole tournament:

alt text

Or just the knockout stage:

alt text

Or just one match:

alt text

In the notebooks everything is clearly explained. If you have any questions don't hesitate to ask.

Used datasets: