(11) ASSESSING THE IMPACT OF FATIGUE ON PERFORMANCE IN COLLEGIATE FEMALE SOCCER: A GPS DATA-DRIVEN ANALYSIS FOR STRATEGIC IN-GAME DECISION MAKING

Purpose: The purpose of this study was to quantify true versus trivial changes in fatigue based on the standard error of measurement (SEM) from metrics collected by global positioning system (GPS) devices throughout the game in female collegiate soccer players.

Methods: Players on a collegiate female soccer team (n = 12) wore GPS devices during games throughout the regular season. Metrics taken for each player from these devices included average speed, peak speed, and average acceleration impulse throughout each game. These metrics were sampled at 100 Hz across the entire game. For each metric and individual athlete, Z-scores were calculated within each game, which were the used to calculate an overall metric of performance changes taken as the average Z-scores from each variable during segments of the game. Metrics were then averaged and divided into the following in-game segments: first 25-min of the first half (F-25-1), last 20-min of the first half (L-20-1), last 15-min of the first half (L-15-1), last 10-min of the first half (L-10-1), last 5-min of the first half (L-5-1), entire first half (FIRST), first 25-min of the second half (F-25-2), last 20-min of the second half (L-20-2), last 15-min of the second half (L-15-2), last 10-min of the second half (L-10-2), last 5-min of the second half (L-5-2), entire second half (SECOND). True differences and trivial differences were compared across each game segment as: True Difference = Segment 2 > Segment 1 + SEM; Trivial Difference = Segment 2 ≤ Segment 1 + SEM. Metrics were collapsed across game, and all analyses were performed in custom-written R software.

Results: From F-25-1 to L-20-1, two players had true differences. From L-20-1 to L-15-1, two players had true differences. From L-15-1 to L-10-1, five players had true differences. From L-10-1 to L-5-1, differences were trivial for all players. From FIRST to SECOND, three players had true differences. From F-25-2 to L-20-2, three players had true differences. From L-20-2 to L-15-2, three players had true differences. From L-15-2 to L-10-2, four players had true differences. From L-10-2 to L-5-2, six players had true differences.

Conclusions: In general, for both the first and second half, players had more true changes in fatigue calculated from GPS metrics during the last segments of each half, while the earlier segments of each half generally had trivial changes. Nevertheless, these findings demonstrate each individual player may respond differently throughout the game. The findings underscore the importance of personalized training and recovery programs to optimize performance, reduce injury risk, and enhance team efficacy. This approach contributes significantly to sports science and athlete management, advocating for a data-driven strategy in maximizing athletic potential and strategic in-game execution.

PRACTICAL APPLICATIONS: The present results demonstrate how coaches and practitioners can use GPS devices during in-game scenarios in female soccer to monitor the physiological aspects of performance such as fatigue, which may provide coaches a method to use data-driven decision making in real-time to optimize individual and team performance. These results also provide insight for how strength and conditioning coaches can enhance needs analyses and sport specific training methods to maximize in-game performance.


Acknowledgements: None