Longitudinal performance trajectories for youth female and male soccer players: 10m-sprint percentile curves adapted to biological age

Abstract

Introduction Longitudinal performance monitoring is essential in sport science for accurate talent identification and forecasting future performance. The present study applied linear mixed-effects models (LMM) to analyze the longitudinal performance development of youth soccer players, both female and male. In addition to percentile curves based on chronological age, a novel approach incorporating biological age, which accounts for maturation variability during puberty, is introduced. The incorporation of biological age, as compared to chronological age, is intended to enable a more equitable assessment of physical potential and to reduce selection bias. The goal was to create an evidence-based tool for coaches and researchers to establish realistic benchmarks, model predictions, a support athlete development.

Methods A total of 33’647 10-meter sprint test results were analyzed, derived from 11’752 male and female soccer players (f = 1’112; m = 10’640) aged 9 to 20 years. To estimate biological age, the Mirwald test was used, with 18’748 measurements utilized for the analysis. These data were extracted from the Swiss Football Association’s online database, covering the period from 2014, respectively 2019 to 2024. LMM were utilized to generate performance trajectories, establish benchmarks, and produce individualized performance predictions. A practical software tool, developed as a Web-Application, was created and made accessible to facilitate individual performance forecasting based on 10-meter sprint times from previous seasons.

Results The mixed model approach identified individualized longitudinal performance developments and estimated predictions of future performance based on both chronological and biological age. Percentile curves based on chronological and biological age were developed, differences could be shown and trajectories for early and late developers were established. A Web-Application tool was created and can be used on the soccer pitch to evaluate current and predict future performance.

Discussion/Conclusion LMM were utilized to analyze longitudinal sport performance data, enabling the establishment of performance benchmarks and the prediction of future outcomes. By incorporating biological age into the model, the predictions account for individual differences in maturation, enhancing accuracy during critical developmental stages such as puberty. A software tool was developed to support coaches in setting realistic training goals and identifying promising soccer players, providing a more tailored and maturity-sensitive approach to athlete development and identification.