The purpose of this study was to evaluate a novel procedure, based on the perceived exertion threshold (PET) concept, involving non-exhaustive tests (PET14-17) to estimate critical power (CP), as well as to examine the reproducibility of these indices. Twenty young men performed 2 trials (Trial 1 and Trial 2) of 4 exhaustive predictive tests to estimate CP, PET, and PET14-17. The slope coefficients of the regression lines corresponding to the ratings of perceived exertion (RPE) versus time relationship (y axis) obtained during the predictive tests were plotted against the power outputs (x axis) to estimate PET. PET was calculated as the intersection point of the regression line in the power axis. The 14-17 RPE range of each predictive test was modeled using the same mathematical procedures used to estimate PET14-17. CP was derived from the power-time hyperbolic equation. Analysis of variance revealed no significant differences among CP (189-194 W), PET (190-191 W) and PET14-17, (191-195 W). The correlations between CP and PET (r = 0.87), CP and PET14-17 (r = 0.89), and PET and PET14-17 (r = 0.88) were all strong. Additionally, the bias +/- limits of agreement when plotting CP and PET was -2.16 +/- 31.60 W, and -5.70 +/- 31.21 W when comparing CP and PET14-17. Relatively high reproducibility levels of CP (ICC = 0.96), PET (ICC = 0.85) and PET14-17 (ICC = 0.83) were recorded in the test-retest fashion. It can be concluded that the PET14-17 can be utilized with relatively high accuracy and reproducibility to estimate CP without causing exhaustion in the subjects.