Physical activity and sleep monitoring

ActiGraph GT9X-link accelerometer (ActiGraph, Pensacola, FL, USA) was used to obtain an objective assessment of players daily PA and sleep. The monitor was initialized at a sampling frequency of 30 Hz and data were processed using Actilife software V.6.13.4 (ActiGraph, Pensacola, Florida, USA). Sixty-second epochs were used in processing the accelerometer data¹⁶. Non-wear time was defined as periods of sixty or more consecutive minutes of zero counts and was excluded from analysis⁶. The cut points used to calculate the variables associated with PA intensity were adapted from prior research³⁸ and standardized into the following levels: sedentary PA (0–180 counts/min); light PA (181–756 counts/min); moderate PA (757–1111 counts/min); vigorous PA (≥ 1112 counts/min); and MVPA (moderate PA + vigorous PA). It is noted that all PA variables were normalized according to the daily hours of accelerometer use. Players’ sleep quality was evaluated overnight (i.e., based on the information reported in the diary). The validity and reliability of wrist actigraphy were previously reported³⁹. The Sadeh algorithm, valid for younger adolescents was used to assess sleep^40,41. The variables considered in the analysis were onset latency (minutes), efficiency (AU) and SFI (AU). Sleep onset latency indicates the amount of time between bedtime and sleep onset time. Sleep efficiency specifies the percentage of time in bed spent asleep. The SFI is an index of restlessness during sleep expressed as a percentage, with higher levels of fragmentation indicating more disrupted sleep.