: Coaches often ask players to judge their recovery status (subjective recovery). We aimed to explore potential determinants of subjective recovery in 101 male professional soccer players of 4 Italian Serie C teams and to further investigate whether the relationship between training load and subjective recovery is mediated by fatigue, sleep quality, muscle soreness, stress and mood. A complete season for each of the four teams was recorded for a total of 16,989 training sessions and matches. Every morning, players rated their perceived fatigue, sleep quality, muscle soreness, stress and mood, and judged their recovery using the Total Quality Recovery (TQR) questionnaire. Training load was obtained after each training session or match. A framework of data analytics of time series was employed to detect the factors associated with subjective recovery. Machine learning and mediation analyses suggest that TQR is primarily associated with ratings of fatigue and muscle soreness at the judgements time, and that these factors mediate most of the relationship between training load of the previous day and subjective recovery. These findings suggest that, to maximize subjective recovery, strategies minimizing fatigue and muscle soreness should be implemented. Reducing the training load of the previous day seems the most effective strategy.
Subjective recovery in professional soccer players: A machine learning and mediation approach
Simonelli, Carlo
Primo
;Formenti, DamianoSecondo
;
2025-01-01
Abstract
: Coaches often ask players to judge their recovery status (subjective recovery). We aimed to explore potential determinants of subjective recovery in 101 male professional soccer players of 4 Italian Serie C teams and to further investigate whether the relationship between training load and subjective recovery is mediated by fatigue, sleep quality, muscle soreness, stress and mood. A complete season for each of the four teams was recorded for a total of 16,989 training sessions and matches. Every morning, players rated their perceived fatigue, sleep quality, muscle soreness, stress and mood, and judged their recovery using the Total Quality Recovery (TQR) questionnaire. Training load was obtained after each training session or match. A framework of data analytics of time series was employed to detect the factors associated with subjective recovery. Machine learning and mediation analyses suggest that TQR is primarily associated with ratings of fatigue and muscle soreness at the judgements time, and that these factors mediate most of the relationship between training load of the previous day and subjective recovery. These findings suggest that, to maximize subjective recovery, strategies minimizing fatigue and muscle soreness should be implemented. Reducing the training load of the previous day seems the most effective strategy.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.