Sprint cycling performance is heavily dependent on mechanical peak power output (PPO) and the underlying power- A nd torque-cadence relationships. Other key indices of these relationships include maximum torque (T MAX), cadence (C MAX) and optimal cadence (C OPT). Two common methods are used in the laboratory to ascertain PPO: Isovelocity and isoinertial. Little research has been carried out to compare the magnitude and reliability of these performance measures with these two common sprint cycling assessments. The aim of this study was to compare the magnitude and reliability of PPO, T MAX, C MAX and C OPT measured with isovelocity and isoinertial sprint cycling methods. Two experimental sessions required 20 trained cyclists to perform isoinertial sprints and then isovelocity sprints. For each method, power-cadence and torque-cadence relationships were established, and PPO and C OPT were interpolated and T MAX and C MAX were extrapolated. The isoinertial method produced significantly higher PPO (p<0.001) and T MAX (p<0.001) than the isovelocity method. However, the isovelocity method produced significantly higher C OPT (p<0.001) and C MAX (p=0.002). Both sprint cycling tests showed high levels of between-session reliability (isoinertial 2.9-4.4%; isovelocity 2.7-4.0%). Functional measures of isovelocity and isoinertial sprint cycling tests were highly reliable but should not be used interchangably.