Exercise endurance time as a function of percent maximal power production.

To develop and validate a mathematical model of the relationship between endurance time (T) and power production, 15 male subjects were first tested for maximal power on an instrumented cycle ergometer at 60 rpm. On subsequent days, they were tested for T at various percentages of maximal power. Curves of T as a function of percent maximal power were curvilinear, and could be made to overlap among subjects by individual abscissa scaling, which resulted in the appearance of horizontal stretching or compression of the curves. The degree of stretching-compression was defined by a statistically obtained scaling factor (F) which served to quantify each subject's endurance ability at fractions of maximal power. F was used to transform percent of maximal power to a scaled power variable (Psc). A curve of the form T = a(Psc)b was developed on 10 of the subjects and validated on the remaining five. Correlation between predicted and actual T was 0.967 for the fitting group and 0.980 for the validation group. A maximal power test and a single endurance test at 40 to 50% of maximal power were found to establish individual endurance-power curves fairly well, with a correlation of 0.828 between actual and predicted T. The combination of F and maximal power for a given physical activity provide a useful profile of an individual's ability to perform at constant exercise intensity.