Performance capacity in athletes depends on the ability to recover from past exercise. While evidence suggests that athletic performance decreases following (partial) sleep deprivation and increases following sleep extension, it is unclear to which extent natural variation in sleep impacts performance. Sleep quantity and, for the first time, sleep stages were assessed among 98 elite athletes on three non-consecutive nights within a 7-day monitoring period, along with performance tests that were taken on standardized times each following morning. Performance assessment included psychomotor performance (10-minute psychomotor vigilance task) and sport-specific tests of fine (e.g., accuracy) and gross motor skills (e.g., endurance, power). Mixed-effects models were employed to assess the effect of sleep quantity (total sleep time (TST), sleep onset latency (SOL), wake after sleep onset, sleep efficiency) and sleep stage duration (light, deep, REM) on performance. Average TST was 7:30 ± 1:05 hours, with a mean variation of 57 minutes across days. Longer TSTs were associated with faster reaction times (p = 0.04). Analyses indicated small and inconsistent effects of sleep quantity (TST, SOL) and sleep staging (light sleep) on gross motor performance, and no effects on fine motor skill performance.

Results indicate that natural variation in sleep quantity impacts psychomotor vigilance to a greater extent than athletic performance. Small or absent effects can be a consequence of the rather small variation in non-manipulated sleep. It is suggested that one night of compromised sleep may not be immediately problematic, but that more extreme sleep loss or accumulated sleep debt may have more severe consequences.

• Psychomotor performance is more strongly affected by sleep than athletic performance

• The effectof sleep on performance appears according to a dose-response relationship

• More extreme sleep loss or accumulated sleep debt over time may prove to be detrimental for performance and recovery.