Purpose: To assess the reliability of power-output measurements of a Wahoo KICKR Power Trainer (KICKR) on 2 separate occasions separated by 14 mo of regular use (∼1 h/wk). Methods: Using the KICKR to set power outputs, powers of 100–600 W in increments of 50 W were assessed at cadences of 80, 90, and 100 rpm that were controlled and validated by a dynamic calibration rig. Results: A small ratio bias of 1.002 (95% limits of agreement [LoA] 0.992–1.011) was observed over 100–600 W at 80–100 rpm between trials 1 and 2. Similar ratio biases with acceptable limits of agreement were observed at 80 rpm (1.003 [95% LoA 0.987–1.018]), 90 rpm (1.000 [0.996–1.005]), and 100 rpm (1.002 [0.997–1.007]). The intraclass correlation coefficient with 95% confidence interval (CI) for mean power between trials was 1.00 (95% CI 1.00–1.00) with a typical error (TE) of 3.1 W and 1.6% observed between trials 1 and 2. Conclusion: When assessed at 2 separate time points 14 mo apart, the KICKR has acceptable reliability for combined power outputs of 100–600 W at 80–100 rpm, reporting overall small ratio biases with acceptable LoA and low TE. Coaches and sport scientists should feel confident in the power output measured by the KICKR over an extended period of time when performing laboratory training and performance assessments.