This study aimed to examine how the power output changes while running at a continuous comfortable velocity on a motorized treadmill by comparing running power averaged during different time intervals. Forty-nine endurance runners performed a running protocol on a treadmill at self-selected comfortable velocity. Power output (W) was estimated with the Stryd™ power meter, and it was examined over six recording intervals within the 3-min recording period: 0–10 s, 0–20 s, 0–30 s, 0–60 s, 0–120 s and 0–180 s. The ANOVAs showed no significant differences in the magnitude of the power output between the recording intervals (p=0.276, F=1.614, partial Eta 2 =0.155). An almost perfect association was also observed in the magnitude of the power output between the recording intervals (ICC≥0.999). Bland-Altman plots revealed no heteroscedasticity of error for the power output in any of the between-intervals comparisons (r 2<0.1), although longer recording intervals yield smaller systematic bias, random errors, and narrower limits of agreement for power output. The results show that power data during running, as measured through the Stryd™ system, is a stable metric with negligible differences, in practical terms, between shorter (i. e., 10, 20, 30, 60 or 120 s) and longer recording intervals (i. e., 180 s).