Darragh Connolly1,2, Sean Stolp2, Ermanno Rampinini2,3, Aaron J. Coutts 2

  1. Sport Science and R&D Department, Juventus Football Club, Torino, Italy;
  2. School of Sport, Exercise and Rehabilitation, Human Performance Research Centre, Faculty of Health, University of Technology Sydney, Sydney, NSW, Australia;
  3. Human Performace Laboratory, Mapei Sport Research Centre, Olgiate Olona, Italy;

Abstract

Purpose: The aim of this study was to guide the selection of load variables assessed within an elite-level academy player monitoring program through the application of a conceptual framework based upon previous literature and expert opinion.

Methods: The variables identified by the framework were then subjected to a principal component analysis (PCA) to help verify the consistency of metrics retained across different age groups, the amount of variance described, and differences between absolute and relative load measures. Training and match data recorded by 145 elite youth soccer players from Under 15 to Under 19 squads were assessed across 2 full competitive seasons. Results: The conceptual framework identified 16 metrics for the absolute PCA and 14 for the relative PCA. The results identified 3 to 4 components that contained distinct themes that were consistent across the different age groups assessed and aligned closely with the constructs of load identified through the conceptual framework. The variance of load described by different components (13.8%–29.3%) was similar across each age group.

Conclusions: These findings indicate that a myriad of factors is required to describe the variance in load experienced by elite youth soccer players and supports the inclusion of both external and internal loads. The general themes identified by the conceptual framework and retained by the PCA can aid practitioners to identify variables that should be incorporated in youth soccer players load monitoring.

Keywords: player monitoring, principal component analysis, youth develoment, Soccer training

Int J Sports Physiol Perform. (Ahead of print) 2026 Feb 19

PMID: 41722558                                                  DOI: 10.1123/ijspp.2025-0253