ABSTRACT Background: Supramaximal constant work rate tests (CWR) elicit intense hyperventilation, thus potentially up-shifting ventilation (⩒E)-to-carbon dioxide (CO2) responses when compared to graded exercise tests (GXT) in athletes. We predicted higher ventilatory efficiency on supramaximal CWR using a new method, challenging the classic orthodox interpretation of an increased ⩒ E -⩒CO2 as ventilatory inefficiency. This misinterpretation could make difficult to differentiate between physiological hyperventilation from heart disease conditions in athletes. Methods: On different days, a GXT and a CWR at 110% of the maximal velocity achieved in the GXT were performed. Twenty-seven athletes completed the two tests and were compared for usual (linear regression) and log-transformed new variables for ventilatory efficiency through paired t-Student statistics. Results: The ⩒ E -⩒CO2 slope (31.4 ± 4.9 vs. 26.2 ± 3.4, p < .001), ⩒ E -⩒CO2 intercept (7.2 ± 7.5 vs. 2.8 ± 4.2, p < .007), ⩒ E /⩒CO2 nadir (33.0 ± 3.6 vs. 25.4 ± 2.2, p < .001), ⩒CO2-log⩒ E slope (10.8 ± 2.9 vs. 6.9 ± 2.2 L*logL−1, p < .001), and η⩒ E (36.0 ± 12 vs. 22.8 ± 8.1%, p < .001) values were all significantly higher in the CWR compared to the GXT. We registered a bi-modal nadir response for ⩒E/⩒CO2 on CWR for 22 out of 27 subjects for the first time. A weak association was observed between ⩒ E /⩒CO2 nadir (coefficient of determination ~ 27%) and time to exhaustion. Conclusions: The new method allows us to improve the quantification and interpretation of ventilatory efficiency in athletes, avoiding misinterpretation due to the up-shifting elicited by the usual ⩒ E -⩒CO2 slope and ⩒ E /⩒CO2 nadir indices, which may be confounded with ventilatory inefficiency. This study suggests that ventilatory changes underpin better ventilatory efficiency during CWR.