Improved determination of mechanical power output in rowing: Experimental results

In rowing, mechanical power output is a key parameter for biophysical analyses and performance monitoring and should therefore be measured accurately. It is common practice to estimate on-water power output as the time average of the dot product of the moment of the handle force relative to the oar pin and the oar angular velocity. In a theoretical analysis we have recently shown that this measure differs from the true power output by an amount that equals the mean of the rower's mass multiplied by the rower's center of mass acceleration and the velocity of the boat. In this study we investigated the difference between a rower's power output calculated using the common proxy and the true power output under different rowing conditions. Nine rowers participated in an on-water experiment consisting of 7 trials in a single scull. Stroke rate, technique and forces applied to the oar were varied. On average, rowers' power output was underestimated with 12.3% when determined using the common proxy. Variations between rowers and rowing conditions were small (SD = 1.1%) and mostly due to differences in stroke rate. To analyze and monitor rowing performance accurately, a correction of the determination of rowers' on-water power output is therefore required.