Projects

This is a collection of my best work and open source projects I have contributed to.

(PS. The title’s are links. Click on them!)

R Masterclass

Conducted an online class via Training Ground Guru, a site focused on the latest developments in the professional football world.

The class introduced participants to R and RStudio and showed how they can be used with football event data.
Interactive follow-along session over Zoom that lasted over 2 hours and 30 minutes.
Live class attended by several individuals including hobbyists and data professionals working in the football industry.

An analysis and visualization R package that works with publicly available soccer data.

Authored 5 functions and contributed additional features to an extra 3 functions while working with other contributors as an open source project. Functions are compatible with JSON data and specific APIs while also supporting the use of web scraped data.
Helped create an official landing page with GitHub Pages that experiences an average of 800+ views and 100+ unique visitors per week. Initial release and updates invite upwards of 1000+ views on the package.
Created the ggshakeR library, a secondary website with articles and a helper code base pertaining to the usage of R in sports analytics.

Interactive web app that generates a dashboard of visualizations that can be useful in getting an overview of a football player’s finishing ability.
Deployed using shinyapps.io on free tier.
Link to the GitHub repo so that the app can be deployed on local system.

Interactive web app that creates a visualization depicting the composition of squads in top European leagues.
Deployed using shinyapps.io on free tier.
Link to the GitHub repo so that the app can be deployed on local system.

Clustered football player passes using the k-means algorithm. The data from the StatsBomb open dataset.
Created two different designs of plots, inspired by work seen on twitter.

Created an xG model in R using tidymodels.

xG Model using Understat data. Variables used included shot distance, angle as well as the body part used to take the shot.

A project to quantitatively determine the closeness and centrality of a shot to the centre of the goal.
This works on Understat shot event data and uses the method of standard deviation from a set point to derive the values.

I have been a regular member of the football/soccer analytics community on twitter, regularly making my code to make visualizations open source.
Here are two repositories with code to create football specific visualizations :-

Here is a repository that shows the implementation of machine learning techniques in R.