The purpose of this study was to examine the effect of environmental temperature on variability in power output, self-selected pacing strategies, and performance during a prolonged cycling time trial. Nine trained male cyclists randomly completed four 40 km cycling time trials in an environmental chamber at 17°C, 22°C, 27°C, and 32°C (40% RH). During the time trials, heart rate, core body temperature, and power output were recorded. The variability in power output was assessed with the use of exposure variation analysis. Mean 40 km power output was significantly lower during 32°C (309 ± 35 W) compared with 17°C (329 ± 31 W), 22°C (324 ± 34 W), and 27°C (322 ± 32 W). In addition, greater variability in power production was observed at 32°C compared with 17°C, as evidenced by a lower (P = .03) standard deviation of the exposure variation matrix (2.9 ± 0.5 vs 3.5 ± 0.4 units, respectively). Core temperature was greater (P < .05) at 32°C compared with 17°C and 22°C from 30 to 40 km, and the rate of rise in core temperature throughout the 40 km time trial was greater (P < .05) at 32°C (0.06 ± 0.04°C·km^–1) compared with 17°C (0.05 ± 0.05°C·km^–1). This study showed that time-trial performance is reduced under hot environmental conditions, and is associated with a shift in the composition of power output. These finding provide insight into the control of pacing strategies during exercise in the heat.