Spatiotemporal pattern analysis of soccer player trajectories

• Main Author: Thadani, Sailesh Gobindram
• Other Authors: Cai Wentong
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2022
• Subjects: Engineering::Computer science and engineering
• Online Access: https://hdl.handle.net/10356/156388
• Institution: Nanyang Technological University

Data analytics have been increasingly used in the field of sports for various purposes. In the game of soccer, there has been an increasing trend in the adoption of data analytics for improving the performance of teams, through methods ranging from scouting new players to analysing opponent’s tactics. As teams have a lot to prepare for and play multiple games in short periods of time, the use of analytics can help teams overcome time constraints to draw insights from past performances in order to strategically improve future performances. This project aims to explore methods to improve post-match analysis of soccer players’ individual and team performance as well as analysis of future opponents’ prior performances. The project focuses on spatiotemporal trajectories of players and the ball during a game with positional and events data from a synthetic dataset that was generated through simulated gameplay. This report showcases the various visualisations and analyses that teams could utilise in efforts to meet the team’s objective of winning. The visualisations allow teams to gain insights into their performances at a glance, from different aspects, such as possession, areas where the ball was played, passes, and shots. Furthermore, the analyses cover individual players’ speed throughout the game, as well as the teams’ centroid and pitch control. Visualisation is an effective tool for data analysts when conveying insights from games to managers. Moreover, the analyses that were conducted yielded positive results and interesting insights. Overall, this project has been successful in replicating some of the analyses that can be utilised by soccer clubs based on theories discussed in several research papers. In the long run, teams can continue to improve their technological and analytics capabilities to enhance their strategy and performances. Some of the possibilities include a 3-dimensional tracking device, to collect data of the height of players and the ball, automating the entire process from data collection to generating the results of the analyses, and extending the analysis to a longer duration, up to an entire season.