Spatiotemporal Data Analysis in Sport

Sport competition and training sessions typically involve athletes interacting with each other and the environment around them. Athlete tracking technologies allow for the measurement of this interaction, by capturing an athlete’s position and space over time. For example, during team-sport matches, local positioning systems (LPS) capture the spatiotemporal data of an athlete, relative to their teammate and opponent. Optical tracking systems can also detect events that happen over time during sporting competition and the location at which they occur. Similarly, global positioning systems (GPS) can capture the position of a cyclist in the peloton during a race.



Despite spatiotemporal data being a rich source of information of where and how events happen within competition and training, working with the large volume of data and deriving meaningful information is difficult for sport scientists and analysts. This unit will introduce students to spatiotemporal data in sport and how to find meaningful patterns within matches, events and training sessions.



Students will learn how to work with common spatiotemporal sources, including athlete ball and tracking data, in R and Python programming languages. Students will understand how to derive meaning from spatiotemporal data and communicate insights for athletes, coaches and stakeholders.

Learning Outcomes

On successful completion of this unit, students will be able to:

1. Scrutinise technology sources that capture spatiotemporal data;
2. Appraise current techniques and methodologies used to investigate events during matches and training;
3. Analyse spatiotemporal data sources in R (or Python) and create impactful visualisations;
4. Implement a range of data mining techniques to meaningful interpret spatiotemporal data; and,
5. Devise reports that create insights from spatiotemporal data for athletes and coaches.