Background: It has been shown that endurance performance after one night of sleep deprivation is not compromised despite the feeling of fatigue and that, in contrast, performance in the heat deteriorates even though people may feel good. However, it is essentially unknown how the estimation of performance capabilities relate to actual performance. We hypothesized that endurance performance in the heat would be overestimated and performance after sleep deprivation would be underestimated. We also hypothesized that jumping performance will be underestimated in the heat. Methods: There were 11 fit (V̇ O2peak 52.0 ± 3.7 ml. kg -1. min -1 ) men, familiar with cycling, who performed a 20-min all-out cycling test (AO) and a vertical jump test (VJ) under four different conditions: a test trial at 24°C, at 11°C ambient temperature without (C) and with one night of sleep deprivation (CS), and at 31°C (H). The subjects estimated the performance prior to exercise in CS, C, and H. Results: AO performance was less for H (6.95 ± 0.36 km) than for C (7.68 ± 0.29 km) and CS (7.62 ± 0.33 km). The subjects underestimated AO performance for CS by 1.11 km and C by 0.42 km, but not for H. VJ was higher in the H condition, in contrast with subjects ' assessment. Discussion: We conclude that subjective estimates of performance are not in line with actual performance for endurance exercise after sleep deprivation and for explosive exercise in the heat. © by the Aerospace Medical Association, Alexandria, VA.