Effects of sleep deprivation and acute supplementations on passing accuracy (non-dominant side). The mean ± SD is displayed for accuracy out of 10 passes on the non-dominant side (20 passes total per trial) for the 10 subjects under different treatment conditions (placebo; 1 or 5 mg/kg caffeine, 50 or 100 mg/kg creatine) either in non-sleep deprived or sleep deprived states. All subjects completed 20 repetitions of the passing skill per trial, alternating passing sides (10 non-dominant side). With placebo treatment sleep deprivation was associated with a significant fall in performance (a) (p < 0.001) compared to non-sleep deprivation. The 50 and 100 mg/kg creatine and 1 and 5 mg/kg caffeine doses were all associated with a significantly better performance (b) (p < 0.001) than the placebo conditions.