Online Program Home
My Program

Abstract Details

Activity Number: 451 - Getting Shots Inside the Box-Cox -- Transformational Soccer Analytics
Type: Invited
Date/Time: Wednesday, August 1, 2018 : 8:30 AM to 10:20 AM
Sponsor: Section on Statistics in Sports
Abstract #326888 Presentation
Title: Interpretable Analysis of Team Performance in Soccer Using Tracking Data: a Hybrid of Supervised and Unsupervised Methods.
Author(s): Paul David Power*
Companies: STATS
Keywords: Tracking; Soccer; Expected Goals; Decision Making; Deep Learning; Supervised Learning
Abstract:

Johan Cruyff, arguably the godfather of modern day soccer once said "when you play a match, it is statistically proven that players actually have the ball 3 minutes on average . So, the most important thing is: what do you do during those 87 minutes when you do not have the ball. That is what determines whether you're a good player or not."

This highlights the current issues with existing data analysis and why the application of tracking data is vital to capture the full context of players and teams actions in soccer.

In this talk we demonstrate how we use supervised and unsupervised methods to better reconstruct the game and answer key questions and myths asked by coaches and pundits including:

- How do teams change formation based on game state and opposition style? - Who is the best passer and destroyer in the EPL? - What is the most effective defensive set up at set-pieces. Man to man or zonal marking? - How does the oppositions playing style effect player workload? - Predicting players performance in a fatigued state.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program