The variability of torque produced during maximal and subject selected submaximal isokinetic knee extension efforst was compared. Subjects were 40 healthy volunteers (20 men and 20 women). Following three warm-up efforts subjects randomly performed one set of both four repeated maximal and submaximal knee extension efforts at 60°/s. Torque was measured at 30 and 45° of flexion and corrected for gravity. Variability of torque was examined by first calculating for each subject the mean, standard deviation (s.d.), and coefficient of variation (CV) for the four trials under the maximal and submaximal conditions. The s.d. was used to represent the absolute variation from the mean, whereas the CV was used to represent the relative variation from the mean. The CVs and s.d.s were compared between the maximal and submaximal efforts using analysis of variance. Variability, as demonstrated by s.d.s, was not consistently greater during submaximal than during maximal efforts. Variability, as demonstrated by CVs, was significantly greater during submaximal than during maximal efforts. In spite of finding statistical differences in the variability of maximal and submaximal efforts, no clear cut-off was identified above and below which performance could be dichotomized into the submaximal and maximal groups based on variability. Relevance Although the variability of angle specific knee-extension torques is greater during submaximal than during maximal efforts, the range of variability associated with maximal and submaximal efforts does not appear to be mutually exclusive. Therefore clinicians should exercise caution in judging the sincerity of an individual's maximal effort based solely on isokinetic measurement variability.