The aim of this study was to compare stride length, and peak knee and ankle moments during over-ground running performed barefoot, in minimalist and in maximalist shoes. Fifteen (10 male, 5 female) recreational endurance runners who habitually wore conventional-cushioned shoes participated. Stride length, as well as knee and ankle moments, were recorded during running on an indoor runway at a self-selected comfortable speed while barefoot, in minimalist and in maximalist shoes. Each condition was performed on a different day and the order of conditions was randomised and counterbalanced. Differences in stride length, and peak knee and ankle moments between conditions were examined with ANCOVA with speed as the covariate. After adjusting for speed, there was a significant increase in stride length from barefoot (1.85 ± 0.01 m) to minimalist (1.91 ± 0.01 m) to maximalist shoes (1.95 ± 0.01 m). Peak knee flexion moment also increased significantly from barefoot (2.51 ± 0.06 Nm⋅kg−1) to minimalist (2.67 ± 0.06 Nm⋅kg−1) to maximalist shoes (2.81 ± 0.06 Nm⋅kg−1). Results then showed peak dorsiflexion moment was lower in the maximalist condition (2.34 ± 0.04 Nm⋅kg−1) than both the barefoot (2.57 ± 0.04 Nm⋅kg−1) and minimalist condition (2.66 ± 0.03 Nm⋅kg−1). Results suggest that stride length and peak knee flexion moment increase from barefoot to minimalist to maximalist shoes, and ankle moment significantly changes as a function of footwear. This indicates that footwear can influence self-selected stride length and peak lower limb loads that are a risk factor for running-related knee injury.