During exhaustive incremental pedaling exercises, root mean square or amplitude of integrated electromyographic values exhibits a nonlinear increase, i.e., the so-called electromyographic threshold (EMG(Th)). As proposed by various authors, this EMG(Th) could be used as a complementary indicator of the aerobic-anaerobic transition in physiological evaluations. However, most of these studies used visual detection for the EMG(Th) and to date no previous study has shown the reliability of this type of EMG(Th) detection. We aimed to compare a visual and a mathematical method for EMG(Th) detection in each of 8 lower limb muscles during incremental cycling exercise. Our results showed an overestimation in the number of cases in which EMG(Th) was detected when using visual inspection (n = 45) compared with the mathematical method (n = 32). However, no significant differences were observed between the 2 methods concerning the power output at which EMG(Th) occurred. These results suggest that EMG(Th) should be mathematically detected. In this context, coaches can easily perform such measurements in order to evaluate the impact of their training programs on the neuromuscular adaptations of their athletes. For example, an automatic mathematical detection of EMG(Th) could be performed during a pedaling exercise in order to detect neuromuscular fatigue. Furthermore, this index could be used during test or training sessions performed either in a lab or in ecological situations. Moreover, the use of EMG(Th) to predict ventilatory threshold occurrence could be an interesting tool for trainers who cannot use the very expensive devices needed to analyze respiratory gas exchanges. ; SIN FINANCIACIÓN ; 1.336 JCR (2006) Q2, 28/73 Sport sciences ; UEM