Sie sind hier: Startseite › Teaching › SS2026 › Soccer Analytics Lab

Soccer Analytics Lab

 soccer-analytics.png

General Information

 

Overview

This is a practical lab course based on Soccermatics. 

This lab course introduces students to the field of football analytics through a hands-on, self-directed learning experience. Based on the open Soccermatics course by David Sumpter, students will develop practical skills in data analysis, statistical modelling, and machine learning — all applied to the context of professional football. The course covers how to understand the game using mathematics, statistics, and machine learning, drawing on techniques from data visualisation, logistic regression, probability distributions, geometry, and AI methods. 

 

Learning Objectives

By the end of this course, students will be able to:

  • Work with professional-grade football event and tracking data in Python
  • Create meaningful visualisations of player and team performance
  • Build and evaluate statistical models such as Expected Goals (xG)
  • Apply machine learning methods to player scouting and action evaluation
  • Communicate data-driven findings clearly to a technical and non-technical audience
  • Collaborate on an independent research project using real football datasets

 

Prerequisites

  • Basic programming experience in Python
  • Familiarity with fundamental statistics (distributions, regression)
  • No prior football analytics knowledge is required

 

Format

This is a self-study lab course. There are no lectures. Students are expected to work through the Soccermatics online materials independently and at their own pace, bringing both conceptual understanding and practical coding skills to each assessment.

Mid-Term Exams (3 × on-site)

Three on-site mid-term exams are held over the course of the semester, roughly corresponding to the progression through the course material. Each exam tests both theoretical understanding (concepts, models, and interpretation) and practical coding ability (working with Python under exam conditions).

⚠️ Progression requirement: Students must achieve at least 50% of the available points in each mid-term exam in order to continue to the next stage of the course. Students who do not meet this threshold are not eligible to proceed.

Group Project

After passing all three mid-term exams, students form small groups and undertake an independent data analysis project using football game data. Groups conduct their analysis, and produce a written report. Projects are expected to go beyond reproducing course examples — students are encouraged to apply course methods to new questions, datasets, or contexts.

Final Presentation Session (on-site)

At the end of the semester, all groups present their project findings in a public on-site session attended by all course participants. Each group is expected to present their methodology, results, and conclusions, and to answer questions from the audience and instructors.