We used three techniques of precooling to test the hypothesis that heat strain would be alleviated, muscle temperature (Tmu) would be reduced, and as a result there would be delayed decrements in peak power output (PPO) during exercise in hot, humid conditions. Twelve male team-sport players completed four cycling intermittent sprint protocols (CISP). Each CISP consisted of twenty 2-min periods, each including 10 s of passive rest, 5 s of maximal sprint against a resistance of 7.5% body mass, and 105 s of active recovery. The CISP, preceded by 20 min of no cooling (Control), precooling via an ice vest (Vest), cold water immersion (Water), and ice packs covering the upper legs (Packs), was performed in hot, humid conditions (mean +/- SE; 33.7 +/- 0.3 degrees C, 51.6 +/- 2.2% relative humidity) in a randomized order. The rate of heat strain increase during the CISP was faster in Control than Water and Packs (P < 0.01), but it was similar to Vest. Packs and Water blunted the rise of Tmu until minute 16 and for the duration of the CISP (40 min), respectively (P < 0.01). Reductions in PPO occurred from minute 32 onward in Control, and an increase in PPO by approximately 4% due to Packs was observed (main effect; P < 0.05). The method of precooling determined the extent to which heat strain was reduced during intermittent sprint cycling, with leg precooling offering the greater ergogenic effect on PPO than either upper body or whole body cooling.