Barbosa, LF, Montagnana, L, Denadai, BS, and Greco, CC. Reliability of cardiorespiratory parameters during cycling exercise performed at the severe domain in active individuals. J Strength Cond Res 28(4): 976–981, 2014—The purpose of this study was to determine the test–retest reliability of cardiorespiratory parameters during cycling exercise performed at severe domain in active individuals. Thirteen active males (24.5 ± 4.5 years) performed the following tests: (a) an incremental test to determine V̇o2max and the intensity associated with V̇o2max (IV̇o2max); and (b) 4 repetitions of square-wave transitions from rest to a power corresponding to 95%IV̇o2max to determine the parameters of V̇o2 kinetics and time to exhaustion (Tlim). Participants performed only 2 transitions on any given day. The interval between the 2 experimental sessions was 48–72 hours. The intraclass correlation coefficient (ICC) and typical error as the coefficient of variation were used to assess reliability. Although the 2 measures of Tlim were moderately related (ICC = 0.78; p < 0.01), Tlim from the second session (545.2 ± 103.1 seconds) was significantly higher than that of the first (492.5 ± 100.9 seconds; p = 0.02). Moderate to high reliability (ICC = 0.76–0.93) for the amplitudes of the V̇o2 kinetics responses was found. Poor reliability, however, was found for time constants and time delays of the V̇o2 kinetics responses. Thus, in nonfamiliarized individuals, Tlim shows a relatively low within-subject coefficient of variation. However, the second score in a series of 2 Tlim tests may be significantly greater than the first. We have also demonstrated that the amplitudes of the V̇o2 response have significantly moderate to high reliability. The time-based parameters, however, present an important day-to-day intraindividual variation. Therefore, several transitions are recommended to monitoring changes in an individual over any time frame.