Performance analysis in SailGP: A machine learning approach

Abstract

The significance of data analysis in high-performance sports has largely increased in recent years, offering opportunities for further exploration using machine learning techniques. SailGP is a relatively new sailing event driven by a passion for speed and innovation. The racing series brings excitement and thrill to the world of sailing that has been relatively unexplored. As a pioneer work in the academic community, our work showcases the power of data-driven approaches in enhancing performance and decision-making at high-performance sailing events. We explore data mining techniques on datasets collected at high-performance sailing events in Bermuda in 2021 and Italy in 2023. By analysing race data, the study aims to gain insights into the relationship between variables such as wind speed, wind direction, foil usage, and daggerboard adjustments, and their impact on boat speed. For a comprehensive understanding, we used three distinct methodologies. Various prediction models, including Gradient Boosting, Random Forest, and a stacked model, were employed and evaluated using performance metrics like R² score and mean squared error. The results demonstrate the models’ ability to accurately predict boat speed. Anomaly detection techniques examined boat stability, adding a new chapter to the research. Additionally, we introduced a reinforcement learning approach to identify optimal settings for speed enhancement, representing another new dimension of the study. These findings refine race strategies, optimise sail and rudder settings, and improve performance in SailGP races. Future plans include collaborating with SailGP to work with larger datasets and integrate models into live racing scenarios.