This study aimed to identify the minimum increment duration required to accurately assess 2 distinct lactate thresholds. A total of 21 elite rowers (12 women and 9 men) participated in this study, and each performed 8 or 9 rowing tests comprising 5 progressive incremental tests (3-, 4-, 5-, 7-, or 10-min steps) and at least three 30-min constant-intensity maximal lactate steady-state assessments. Power output (PO) at lactate threshold 1 was higher in the 3- and 4-min incremental tests. No other measures were different for lactate threshold 1. The PO at the second lactate threshold was different between most tests and was higher than the PO at maximal lactate steady state, except for the 10-min incremental test. Lactate threshold 2 oxygen consumption was higher in the 3-, 4-, and 5-min tests, but heart rate (HR) and rating of perceived exertion were not different between tests. Peak PO in the incremental tests was inversely related to the step durations (r 2 = .86, P ≤ .02). Peak oxygen consumption was higher in the shorter (≤5 min) than the longer (≥7 min) incremental tests, whereas peak HR was not different between tests. These data suggest that for the methods used in this study, incremental exercise tests with step durations ≤7 min overestimate maximal lactate steady-state exercise intensity, peak physiological values are best determined using incremental tests with step durations ≤4 min, and HR measures are not affected by step duration, and therefore, prescription of training HRs can be made using any of these tests.