Abstract Purpose The objective was to investigate if performing a sub-peak or supra-peak verification phase following a ramp test provides additional value for determining 'true' maximum oxygen uptake ($$ {{\dot{\text{V}}}} $$ V ˙ O2). Methods 17 and 14 well-trained males and females, respectively, performed two ramp tests each followed by a verification phase. While the ramp tests were identical, the verification phase differed in power output, wherein the power output was either 95% or 105% of the peak power output from the ramp test. The recovery phase before the verification phase lasted until capillary blood lactate concentration was ≤ 4 mmol·L−1. If a $$ {{\dot{\text{V}}}} $$ V ˙ O2 plateau occurred during ramp test, the following verification phase was considered to provide no added value. If no $$ {{\dot{\text{V}}}} $$ V ˙ O2 plateau occurred and the highest $$ {{\dot{\text{V}}}} $$ V ˙ O2 ($$ {{\dot{\text{V}}}} $$ V ˙ O2peak) during verification phase was < 97%, between 97 and 103%, or > 103% of $$ {{\dot{\text{V}}}} $$ V ˙ O2peak achieved during the ramp test, no value, potential value, and certain value were attributed to the verification phase, respectively. Results Mean (standard deviation) $$ {{\dot{\text{V}}}} $$ V ˙ O2peak during both ramp tests was 64.5 (6.0) mL·kg−1·min−1 for males and 54.8 (6.2) mL·kg−1·min−1 for females. For the 95% verification phase, 20 tests showed either a $$ {{\dot{\text{V}}}} $$ V ˙ O2 plateau during ramp test or a verification $$ {{\dot{\text{V}}}} $$ V ˙ O2peak < 97%, indicating no value, 11 showed potential value, and 0 certain value. For the 105% verification phase, the values were 26, 5, and 0 tests, respectively. Conclusion In well-trained adults, a sub-peak verification phase might add little value in determining 'true' maximum $$ {{\dot{\text{V}}}} $$ V ˙ O2, while a supra-peak verification phase adds no value.