Introduction The analysis of position and event data for match analysis in football has seen an increasing interest in research and practice in recent years (Memmert, 2022, 2023). However, the barriers to enter this field can be high for newcomers. Sports scientists are often not trained programmers but (especially position-) data require complex parsing, filtering and manipulation routines before further analysis. These are often incompatible across data providers. Another reason is the limited public availability of high-quality data. No data set of synchronized event- and position data from elite football matches exists to this day. We tackle these challenges by introducing the software package floodlight (Raabe et al., 2022) for Python along with making seven matches of professional football available under open source. Methods The software package floodlight for Python was developed to provide a convenient solution for data driven match analysis with an emphasis on scientific computing. At the current state, the package includes parsing modules for nine commercial data providers and two public datasets. In addition, it allows the user to directly manipulate and filter the data. Models to calculate intensity are implemented along models of geometry, space control, and collective behavior. The accompanied data set consists of two matches from the German Bundesliga season 2022/23 1st division and five matches from the respective 2nd division. For all matches, position data from every player and the ball is available in 25 Hz, along with de tailed event data and meta information. Results The dataset contains information about 10 teams with 207 players, 11.137 events and a total of 1.002.644 frames of position data. Using floodlight, we show a pipeline to create advanced data analysis and visualization that is interchangeable, reproducible, and requires minimal prior coding experience. Discussion Lowering the entry bar into match analysis in football, will allow newcomers from various fields to contribute to this field. Additionally, we emphasize the idea of open science. Here, floodlight can serve as a platform for reproducible and open research projects in the future.
© Copyright 2024 15. Symposium der dvs-Sektion "Sportinformatik und Sporttechnologie": Zwischen Geistesakrobatik und praktischer Anwendung: Innovationen in der Sportinformatik und Sporttechnologie - Abstractband. Veröffentlicht von Technische Universität Dortmund. Alle Rechte vorbehalten.