Modelling the Physical Performance of the Vancouver Whitecaps

Student Capstone Project

During a 90-minute plus soccer match, players can run as much as fifteen kilometers and sometimes during critical moments in a match, some might exert a massive burst of energy that can exact a large physiological toll on the body.

Similarly, there are many types of drills and exercises conducted during training sessions that can be analyzed to get a better understanding of how players respond physically to different types of stresses and loads on their bodies.

Analyzing physical performance to build a model for player fatigue and fitness was the topic of the capstone project that a group of UBC Master of Data Science (MDS) Vancouver students did with the Vancouver Whitecaps.

The Vancouver Whitecaps are Vancouver’s professional Major League Soccer franchise that has a long and storied history dating back to 1973.

For the capstone project, the Whitecaps wanted to assess two factors: 1) player endurance as measured by their capacity to sustain long duration efforts, and 2) player speed as measured by performance over shorter durations.

In order to begin this assessment, the MDS students received data collected from vests with embedded sensors that all Whitecaps players wore during training sessions and every match. The Whitecaps wanted to investigate whether the trace data obtained from the sensors can be used to obtain new insights about player fitness.

From the data collected, the students adopted an endurance-sport framework called the Critical Power Model. Their Critical Power Model looked at the maximum level of effort an athlete can sustain without accumulating fatigue and the amount of energy an athlete has available for levels of effort above their critical power.

The goal of the project was to provide an interactive and visual platform to analyze and understand training loads for each session for each player through a data pipeline that fits several advanced models on raw wearable trace data. With the model outputs, they can evaluate changes in athletes’ fitness levels throughout the competitive season. Through the platform, the Whitecaps may use the interactive platform to more closely examine athletes’ physical outputs during training sessions and matches.

Explore our Data Science Programs Explore Other Data in Action Stories