INTRODUCTION: Repeated-sprint training in hypoxia (RSH) seems to be particularly effective in increasing anaerobic performance of athletes in various sports. However, belief in the positive effects of altitude is widespread among coaches and athletes, raising the issue of a putative placebo effect of such training. The purpose of this study was to determine the respective effects of normobaric hypoxia and placebo combined to a repeated-sprint training. METHODS: Thirty moderately trained participants [age (mean ± SD) 20.7 ± 3.1 years] were randomly allocated to 3 groups: normoxia (RSN; inspired oxygen fraction (FiO2) 20.9%) placebo (RSN-P; FiO2 20.9%) or hypoxia (RSH; FiO2 14.5%). Participants from the RSH and RSN-P groups were connected to a hypoxic generator during training and were told they trained under simulated altitude (between 2500 and 3500 m). Participants from the RSN group knew they were training at sea level. Training consisted in six cycling sessions comprising three sets of 8 x 6-s sprint with 24-s over three weeks. Power outputs were measured during a Wingate (30 s) and a repeated-sprint ability test (RSA; 10 x 6-s sprint with 24-s recovery) before, one and two weeks after training. RESULTS: None of the participants in the RSN-P and RSH groups estimated being in normoxia during training, and mean estimated altitude was not different between groups (2500 ± 445 and 2605 ± 425 m in RSN-P and RSH, p = 0.60 respectively). As expected, the RSH group was the only one with SpO2 values that differed from those of the RSN group (85.5 ± 2.6%, 96 ± 1.4%, and 95.8 ± 0.6%; p < 0.001 for RSH, RSN-P, and RSN, respectively). The relative intensity during training (power output normalized by the mean power output sustained during the pre-RSA test) was not different between the 3 groups (91 ± 3.9, 91.2 ± 6.28, and 87.2 ± 3.7%; p = 0.14 for RSH, RSN-P, and RSN, respectively), whereas the subjective rating of perceived exertion was higher in the RSH group (15.3 ± 0.9, 14.3± 0.8, and 14.4 ± 1; p = 0.04 for RSH, RSN-P, and RSN, respectively). Mean and peak power outputs on RSA and Wingate tests were significantly increased after training in all groups. However, no group has progressed more than another. Regarding the ability of participants to achieve a maximum number of sprints more powerful than 85% of the best performance of the RSA pre-test, no evolution exists (p = 0.91). CONCLUSION: These data show that repeated-sprint training is effective in improving anaerobic performance and that hypoxia exposure during training did not provide additional benefit.