The novel self-paced, cycle-based maximal oxygen uptake (V̇O2max) test (SPV) has been shown to produce higher V̇O2max values than standard graded exercise test (GXT) protocols. This study sought to ascertain whether these observations would also be apparent in a self-paced, treadmill-based test design. Fourteen trained male runners performed a standard GXT on a motorised treadmill and a self-paced V̇O2max test on a nonmotorised treadmill in a counter-balanced design. The GXT included a plateau verification and was designed to last between 8 and 12 min. The self-paced test included 5×2 min stages and allowed participants to set their own running speed based on fixed increments in rating of perceived exertion. Significantly higher V̇O2max values (t = 3.71, p = 0.003) were achieved in the self-paced test (64.4 ± 7.3 mL·kg-1.min-1) compared with the GXT (61.3 ± 7.3 mL·kg-1.min-1), and 13 of the 14 participants achieved the same or higher V̇O2max values in the self-paced test. Higher (p = 0.01) maximum heart rates were observed in the GXT (191 ± 10 beats·min-1 vs. 187 ± 7 beats·min-1), but no differences were observed in any other recorded variables. The self-paced V̇O2max test may provide a more valid means of measuring V̇O2max than the GXT and suggests that a V̇O2 plateau during a GXT does not always signify achievement of a definitive V̇O2max. These results provide further support that self-paced V̇O2max testing produces higher values for maximal oxygen uptake.