To determine in trained females and males i) the agreement between the gas exchange threshold (GET), lactate threshold 1 (LT1), and heart rate variability threshold 1 (HRVT1), as well as between the respiratory compensation point (RCP), lactate threshold 2 (LT2), and heart rate variability threshold 2 (HRVT2), and ii) the reproducibility of HRVT1 and HRVT2 during 2-min incremental step protocols. Fifty-seven trained participants (24 females) completed a 2-min step incremental test to task failure. Nineteen participants (eight females) completed a second test to evaluate reproducibility. Gas exchange and ventilatory responses, blood lactate concentration, and RR time series were recorded to assess the oxygen consumption (V̇O 2 ) and heart rate (HR) associated with the GET, RCP, LT1, LT2, HRVT1, and HRVT2. V̇O 2 -GET versus V̇O 2 -HRVT1 and HR-GET versus HR-HRVT1 were statistically different for females (29.5 ± 4.0 vs 34.6 ± 6.1 mL·kg -1 ·min -1 ; 154 ± 11 vs 166 ± 12 bpm) and for males (33.9 ± 4.2 vs 42.7 ± 4.6 mL·kg -1 ·min -1 ; 145 ± 11 vs 165 ± 9 bpm; P < 0.001). V̇O 2 and HR at HRVT1 were greater than at LT1 ( P < 0.05). V̇O 2 -RCP versus V̇O 2 -HRVT2 and HR-RCP versus HR-HRVT2 were not statistically different for females (40.1 ± 4.7 vs 39.5 ± 6.7 mL·kg -1 ·min -1 ; 177 ± 9 vs 176 ± 9 bpm) and males (48.4 ± 5.4 vs 47.8 ± 4.8 mL·kg -1 ·min -1 ; 176 ± 8 vs 175 ± 9 bpm; P > 0.05). V̇O 2 and HR responses at LT2 were similar to HRVT2 ( P > 0.05). Intraclass correlation coefficient for V̇O 2 -HRVT1, HR-HRVT1, V̇O 2 -HRVT2, and HR-HRVT2 indicated good reproducibility when comparing the two different time points to standard methods. Whereas HRVT2 is a valid and reproducible estimate of the RCP/LT2, current approaches for HRVT1 estimation did not show good agreement with outcomes at GET and LT1.