This study investigated the effect of flywheel moment of inertia (0.029, 0.061, and 0.089 kg·m2) on concentric and eccentric peak power and eccentric:concentric peak power ratio during unilateral flywheel leg curl and hip extension exercises. Moreover, the inter-session reliability of peak power was analyzed during both exercises. Twenty amateur male soccer athletes attended five visits—performing three sets of eight repetitions of either unilateral leg curl or hip extension (all three moments of inertias) during each visit. For the unilateral leg curl, there were no differences in any measure between moments of inertia (p = 0.479) but a higher eccentric than concentric peak power for all moments of inertia (p < 0.001). For the unilateral hip extension, differences between moments of inertia were reported for all measures (p < 0.05). Specifically, the lowest moment of inertia elicited the greatest concentric peak power (p = 0.022), there were no differences with the medium inertia (p = 0.391), and the greatest moment of inertia obtained the greatest eccentric peak power (p = 0.036). Peak power measures obtained acceptable to excellent reliability while the eccentric:concentric ratio reported unacceptable to good reliability for both exercises. A variety of moments of inertia can elicit high eccentric knee flexor demands during unilateral leg curls, whereas higher moments of inertia are needed to achieve an eccentric-overload in peak power during hip extensions. Different exercises may have different inertia-power relationships. Concentric and eccentric peak power measures should continue to inform training, while the eccentric:concentric ratio should not be used.