In this study, we assessed the effectiveness of a mobile feedback system on transitioning runners to a non‐rearfoot strike (NRFS) pattern and prospectively compared injury incidence rates at one year between rearfoot strike (RFS) and NRFS runners. 128 RFS runners participated in a 2‐hour training session to learn a NRFS pattern with 114 completing the 1‐year follow‐up. Participants were randomized into a control group (CON) with no additional training and a biofeedback group (BFG) where they received equipment to provide real‐time biofeedback to augment the transition to a NRFS pattern. Foot strike patterns (FSP) were assessed at baseline, post‐training, 6‐months, and 1‐year. Injury data were collected through weekly email surveys over one year. Eighty percent of runners demonstrated a NRFS pattern following the training session (91/114, P < .001). The percentage of NRFS runners remaining at the one‐year follow‐up decreased slightly in both groups, but was not significantly different between groups (CON = 69%, P = .29; BFG = 75%, P = .36). Injury rates were similar between RFS runners (37% injured) and NRFS runners (30% injured) after one year (P = .47). The relative risk for knee injury in RFS runners was 5.64 (95% CI: 1.90‐16.8; P = .02). In conclusion, both groups had a significant number of participants transition to a NRFS immediately after training and maintain NRFS at the 1‐year follow‐up. However, compliance with the sensor in the BFG group was very poor due to limitations of the sensor. Regardless of FSP, runners experienced 1‐year injury incidence rates between 30%‐37%. RFS runners had nearly a six times greater risk for developing a knee injury than NRFS runners.