# Physical Workload Patterns in U-18 Basketball Using LPS and MEMS Data: A Principal Component Analysis by Quarter and Playing Position

## TL;DR

This study uses wearable sensors and data analysis to show how physical demands in U18 basketball vary by game quarter and player position, helping coaches tailor training and reduce injury risks.

## Contribution

The study introduces an integrative approach using PCA on LPS and MEMS data to identify key workload variables in basketball by quarter and position.

## Key findings

- High-intensity variables like accelerations and explosive distance were most prominent in early quarters and declined over time.
- Guards showed frequent accelerations and direction changes, forwards had mixed-intensity efforts, and centers experienced high impacts and jumps.
- PCA explained 61–73% of variance by quarter and 64–69% by position, revealing distinct workload profiles for different roles.

## Abstract

What are the main findings?
High-intensity variables (e.g., accelerations, explosive distance) were identified in early quarters and declined progressively, with 5–8 components explaining 61–73% of the variance.Position-specific profiles emerged: guards exhibited frequent accelerations and direction changes, forwards engaged in mixed-intensity efforts, and centers experienced a high number of impacts and jumps.

High-intensity variables (e.g., accelerations, explosive distance) were identified in early quarters and declined progressively, with 5–8 components explaining 61–73% of the variance.

Position-specific profiles emerged: guards exhibited frequent accelerations and direction changes, forwards engaged in mixed-intensity efforts, and centers experienced a high number of impacts and jumps.

What are the implications of the main findings?
LPS and MEMS data, combined with PCA, enable basketball teams to identify the most important workload parameters and specific profiles based on contextual factors.Findings support individualized training prescriptions and injury prevention by understanding the dynamic nature of basketball demands during competition.

LPS and MEMS data, combined with PCA, enable basketball teams to identify the most important workload parameters and specific profiles based on contextual factors.

Findings support individualized training prescriptions and injury prevention by understanding the dynamic nature of basketball demands during competition.

Basketball is a high-intensity, intermittent sport in which physical demands fluctuate depending on different contextual variables. Most studies addressed these demands in isolation without integrative approaches. Therefore, the present study aimed to identify key variables explaining players’ physical workload across game quarters and playing positions through principal component analysis (PCA). Ninety-four elite U18 male basketball players were registered during the EuroLeague Basketball ANGT Finals using WIMU PRO™ multi-sensor wearable devices that integrate local positioning systems (LPS) and microelectromechanical systems (MEMS). From over 250 recorded variables, 31 were selected and analyzed by PCA for dimensionality reduction, analyzing the effects of game quarter and playing position. Five to eight principal components explained 61–73% of the variance per game quarter, while between four and seven components explained 64–69% per playing position. High-intensity variables showed strong component loadings in early quarters, with explosive distance (loading = 0.898 in total game, 0.645 in Q1) progressively declining to complete absence in Q4. Position-based analysis revealed specific workload profiles: guards required seven components to explain 69.25% of the variance, with complex movement patterns, forwards showed the highest explosive distance loading (0.810) among all positions, and centers demonstrated concentrated power demands, with PC1 explaining 34.12% of the variance, dominated by acceleration distance (loading = 0.887). These findings support situational and individualized training approaches, allowing coaches to design individual training programs, adjust rotation strategies during games, and replicate demanding scenarios in training while minimizing injury risk.