While it is often reported that muscular coactivation increases with age, the mechanical impact of antagonist muscles, i.e., the antagonist torque, remains to be assessed. The aim of this study was to determine if the mechanical impact of the antagonist muscles may contribute to the age-related decline in the resultant torque during maximal voluntary contraction in knee flexion (KF) and knee extension (KE). Eight young (19–28 years old) and eight older (62–81 years old) healthy males participated in neuromuscular testing. Maximal resultant torque was simultaneously recorded with the electromyographic activity of quadriceps and hamstring muscles. The torque recorded in the antagonist muscles was estimated using a biofeedback technique. Resultant torques significantly decreased with age in both KF (−41 %, p < 0.005) and KE (−35 %, p < 0.01). Agonist and antagonist torques were significantly reduced in KF (−44 %, p < 0.05; −57 %, p < 0.05) and in KE (−37 %, p < 0.01; −50 %, p < 0.05). The torque elicited by double twitch stimulation (−37 %, p < 0.01) and the activation level (−12 %, p < 0.05) of quadriceps was significantly lower in older men compared to young men. This study showed that antagonist torques were not responsible for age-related declines in KF and KE resultant torques. Therefore, decreased resultant torques with age, in particular in KE, can primarily be explained by impairments of the peripheral factors (excitation–contraction coupling) as well as by decreased neural agonist activation.