The present study was conducted to assess test-retest reproducibility of explosive strength measurements during single-joint isometric knee extension using the IsoMed 2000 dynamometer. Thirty-one physically active male subjects (mean age: 23.7 years) were measured on two occasions separated by 48–72 h. The intraclass correlation coefficient (ICC 2,1) and the coefficient of variation (CV) were calculated for (i) maximum torque (MVC), (ii) the peak rate of torque development (RTD_peak) as well as for (iii) the average rate of torque development (RTD) and the impulse taken at several predefined time intervals (0–30 to 0–300 ms); thereby explosive strength variables were derived in two conceptually different versions: on the one hand from the MVC-trial (version I), on the other hand from the trial showing the RTD_peak (version II). High ICC-values (0.80–0.99) and acceptable CV-values (1.9–8.7%) could be found for MVC as well as for the RTD and the impulse taken at time intervals of ≥100 ms, regardless of whether version I or II was used. In contrast, measurements of the RTD_peak as well as the RTD and the impulse taken during the very early contraction phase (i.e. RTD/impulse_0–30ms and RTD/impulse_0–50ms) showed clearly weaker reproducibility results (ICC: 0.53–0.84; CV: 7.3–16.4%) and gave rise to considerable doubts as to clinical usefulness, especially when derived using version I. However, if there is a need to measure explosive strength for earlier time intervals in practice, it is, in view of stronger reproducibility results, recommended to concentrate on measures derived from version II, which is based on the RTD_peak-trial.