Advancing athlete development: How Percentile Comparison Methods (PCMs) can identify youth developmental profiles

Shaun Abbott; Clorinda Hogan; Giovanni Postiglione; Gary Barclay; Stephen Cobley

doi:10.1016/j.jsams.2024.03.004

Advancing athlete development: How Percentile Comparison Methods (PCMs) can identify youth developmental profiles

J Sci Med Sport. 2024 Mar 21:S1440-2440(24)00081-1. doi: 10.1016/j.jsams.2024.03.004. Online ahead of print.

Authors

Shaun Abbott¹, Clorinda Hogan², Giovanni Postiglione², Gary Barclay³, Stephen Cobley⁴

Affiliations

¹ Discipline of Exercise & Sport Science, Faculty of Health Sciences, The University of Sydney, Australia. Electronic address: https://twitter.com/ShaunAbb0tt.
² Discipline of Exercise & Sport Science, Faculty of Health Sciences, The University of Sydney, Australia.
³ Swimming Australia Ltd, Australia.
⁴ Discipline of Exercise & Sport Science, Faculty of Health Sciences, The University of Sydney, Australia. Electronic address: stephen.cobley@sydney.edu.au.

PMID: 38594115
DOI: 10.1016/j.jsams.2024.03.004

Abstract

Objectives: Inter-individual developmental differences confound the capability to accurately evaluate youth athletic performance, highlighting the need for considerate methodology and analytical approaches. The present study demonstrated how Percentile Comparison Methods (PCMs) were developed, tested, and applied to identify athlete developmental profiles in Australian youth swimming.

Design: Cross-sectional.

Methods: Participants were N = 866 female 100-metre (m) Front-Crawl swimmers, aged 9-15 years, competing at 36 Australian regional-national level long course events. At respective events, swim performance time was collated alongside, age, date of birth, and anthropometric measures to identify age group, relative age, and maturity status. Quadratic relative age and maturity status with 100-m performance regression trendlines were generated. Then, individual swim performances at a given relative age or maturity status were converted into percentile rank distributions and compared with raw (unadjusted) annual age-group performance percentile ranks.

Results: At a cohort level, initial testing confirmed relative age and maturity-adjusted percentile rankings were associated with general rank improvements for relatively younger and later maturing swimmers compared to raw ranks (and vice versa). When assessing individual swimmer plots, where three percentile rank scores were compared and rank change threshold criteria applied, five Percentile Comparison Method profile types were identified, namely: 'Early Developing' (19 %); 'Later Developing' (18 %); 'Consistent' (15 %); 'Mixed' (38 %) and 'Counteracting' (10 %). Percentile Comparison Method plots helped identify developmentally (dis-)advantaged swimmers; specific factors leading to (dis-)advantage, and likely onward development trajectories.

Conclusions: Overall and with practical considerations, Percentile Comparison Methods can improve the validity of youth athletic performance evaluation as well as inform athlete development programming.

Keywords: Maturation; Relative age effects; Skill evaluation; Sport performance; Talent identification.