Patellofemoral pain (PFP) is a common injury among runners. Knee biomechanical factors associated with PFP, however, remain unclear. The purpose of this study was to determine possible associations between knee biomechanics and symptoms of PFP in recreational runners.Fifteen male and 15 female recreational runners with PFP were enrolled as the PFP group, 30 matched runners without PFP were recruited as the control group. The PFP group was tested running with and without knee pain, while the control group had only one running test. Reflective marker coordinates and ground reaction force data were collected in each test. Knee kinematics and kinetics during running were reduced and compared between groups (PFP group without knee pain and control group) and between pain conditions (PFP group with knee pain and without knee pain), as well as between sexes.Female and male participants with PFP had an increased peak knee valgus angle when running without pain compared to matched controls (P = 0.001), and to themselves when running with pain (P = 0.001). Male participants with PFP also had an increased peak knee flexion angle when running without pain compared to matched controls (P = 0.008), however did not decrease their peak knee flexion angle when running with pain (P = 0.245). No significant main effect of group or pain condition on any peak knee joint moment during running was detected (P ≥ 0.175).Increased peak knee valgus angle during running appears to be a critical biomechanical factor associated with PFP in recreational runners, while decreasing knee valgus angle during running may be an adaptation to reduce symptoms of PFP. Increased peak knee flexion angle during running appears to be another biomechanical factor associated with PFP that is sex specific for male recreational runners.