1. Simplifying External Load Data in NCAA Division-I Men's Basketball Competitions: A Principal Component Analysis
- Author
-
Jason D. Stone, Justin J. Merrigan, Jad Ramadan, Robert Shaun Brown, Gerald T. Cheng, W. Guy Hornsby, Holden Smith, Scott M. Galster, and Joshua A. Hagen
- Subjects
sport science ,wearables ,inertial measurement unit (IMU) ,collegiate athletics ,athlete monitoring ,Sports ,GV557-1198.995 - Abstract
The primary purpose was to simplify external load data obtained during Division-I (DI) basketball competitions via principal component analysis (PCA). A secondary purpose was to determine if the PCA results were sensitive to load demands of different positional groups (POS). Data comprised 229 observations obtained from 10 men's basketball athletes participating in NCAA DI competitions. Each athlete donned an inertial measurement unit that was affixed to the same location on their shorts prior to competition. The PCA revealed two factors that possessed eigenvalues >1.0 and explained 81.42% of the total variance. The first factor comprised total decelerations (totDEC, 0.94), average speed (avgSPD, 0.90), total accelerations (totACC, 0.85), total mechanical load (totMECH, 0.84), and total jump load (totJUMP, 0.78). Maximum speed (maxSPD, 0.94) was the lone contributor to the second factor. Based on the PCA, external load variables were included in a multinomial logistic regression that predicted POS (Overall model, p < 0.0001; AUCcenters = 0.93, AUCguards = 0.88, AUCforwards = 0.80), but only maxSPD, totDEC, totJUMP, and totMECH were significant contributors to the model's success (p < 0.0001 for each). Even with the high significance, the model still had some issues differentiating between guards and forwards, as in-game demands often overlap between the two positions. Nevertheless, the PCA was effective at simplifying a large external load dataset collected on NCAA DI men's basketball athletes. These data revealed that maxSPD, totDEC, totJUMP, and totMECH were the most sensitive to positional differences during competitions. To best characterize competition demands, such variables may be used to individualize training and recovery regimens most effectively.
- Published
- 2022
- Full Text
- View/download PDF