Search

Introduction: A promising approach to optimizing recovery in youth football has been the use of machine learning (ML) models to predict recovery states and prevent mental fatigue. This research investigates the application of ML models in classifying male young football players aged under (U)15, U17, and U19 according to their recovery state. Weekly training load data were systematically monitored across three age groups throughout the initial month of the 2019-2020 competitive season, covering 18 training sessions and 120 observation instances. Outfield players were tracked using portable 18-Hz global positioning system (GPS) devices, while heart rate (HR) was measured using 1 Hz telemetry HR bands. The rating of perceived exertion (RPE 6-20) and total quality recovery (TQR 6-20) scores were employed to evaluate perceived exertion, internal training load, and recovery state, respectively. Data preprocessing involved handling missing values, normalization, and feature selection using correlation coefficients and a random forest (RF) classifier. Five ML algorithms [K-nearest neighbors (KNN), extreme gradient boosting (XGBoost), support vector machine (SVM), RF, and decision tree (DT)] were assessed for classification performance. The K-fold method was employed to cross-validate the ML outputs., Results: A high accuracy for this ML classification model (73-100%) was verified. The feature selection highlighted critical variables, and we implemented the ML algorithms considering a panel of 9 variables (U15, U19, body mass, accelerations, decelerations, training weeks, sprint distance, and RPE). These features were included according to their percentage of importance (3-18%). The results were cross-validated with good accuracy across 5-fold (79%)., Conclusion: The five ML models, in combination with weekly data, demonstrated the efficacy of wearable device-collected features as an efficient combination in predicting football players' recovery states., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. The author(s) declared that they were an editorial board member of Frontiers, at the time of submission. This had no impact on the peer review process and the final decision., (Copyright © 2024 Teixeira, Encarnação, Branquinho, Ferraz, Portella, Monteiro, Morgans, Barbosa, Monteiro and Forte.)

The aims of this study were 1) to analyze the influence of chronological age, relative age, and biological maturation on accumulated training load and perceived exertion in young sub-elite football players and 2) to understand the interaction effects amongst age grouping, maturation status, and birth quartiles on accumulated training load and perceived exertion in this target population. A 6-week period (18 training sessions and 324 observation cases) concerning 60 young male sub-elite football players grouped into relative age (Q1 to Q4), age group (U15, U17, and U19), and maturation status (Pre-peak height velocity (PHV), Mid-PHV, and Post-PHV) was established. External training load data were collected using 18 Hz global positioning system technology (GPS), heart-rate measures by a 1 Hz short-range telemetry system, and perceived exertion with total quality recovery (TQR) and rating of perceived exertion (RPE). U17 players and U15 players were 2.35 (95% CI: 1.25-4.51) and 1.60 (95% CI: 0.19-4.33) times more likely to pertain to Q1 and Q3, respectively. A negative magnitude for odds ratio was found in all four quartile comparisons within maturation status (95% CI: 6.72-0.64), except for Mid-PHV on Q2 (95% CI: 0.19-4.33). Between- and within-subject analysis reported significant differences in all variables on age group comparison measures ( F = 0.439 to 26.636, p = 0.000 to 0.019, η 2 = 0.003-0.037), except for dynamic stress load (DSL). Between-subject analysis on maturity status comparison demonstrated significant differences for all training load measures ( F = 6.593 to 14.424, p = 0.000 to 0.037, η 2 = 0.020-0.092). Interaction effects were found for age group x maturity band x relative age ( Λ Pillai's = 0.391, Λ Wilk's = 0.609, F = 11.385 , p = 0.000, η 2 = 0.391) and maturity band x relative age ( Λ Pillai's = 0.252, Λ Wilk's = 0.769, F = 0.955 , p = 0.004, η 2 = 0.112). Current research has confirmed the effects of chronological age, relative age, and biological maturation on accumulated training load. Perceived exertion does not seem to show any differences concerning age group or maturity status. Evidence should be helpful for professionals to optimize the training process and young football players' performance., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2022 Teixeira, Alves, Ferraz, Forte, Leal, Ribeiro, Silva, Barbosa and Monteiro.)