Mechanical power output in rowing should not be determined from oar forces and oar motion alone

Mathijs J. Hofmijster*, Lotte L. Lintmeijer, Peter J. Beek, A. J. Knoek van Soest

*Corresponding author for this work

Research output: Contribution to JournalArticleAcademicpeer-review


Mechanical power output is a key performance-determining variable in many cyclic sports. In rowing, instantaneous power output is commonly determined as the dot product of handle force moment and oar angular velocity. The aim of this study was to show that this commonly used proxy is theoretically flawed and to provide an indication of the magnitude of the error. To obtain a consistent dataset, simulations were performed using a previously proposed forward dynamical model. Inputs were previously recorded rower kinematics and horizontal oar angle, at 20 and 32 strokes∙min−1. From simulation outputs, true power output and power output according to the common proxy were calculated. The error when using the common proxy was quantified as the difference between the average power output according to the proxy and the true average power output (P̅residual), and as the ratio of this difference to the true average power output (ratiores./rower). At stroke rate 20, P̅residual was 27.4 W and ratiores./rower was 0.143; at stroke rate 32, P̅residual was 44.3 W and ratiores./rower was 0.142. Power output in rowing appears to be underestimated when calculated according to the common proxy. Simulations suggest this error to be at least 10% of the true power output.

Original languageEnglish
Pages (from-to)2147-2153
Number of pages7
JournalJournal of Sports Sciences
Issue number18
Early online date8 May 2018
Publication statusPublished - 17 Sept 2018


  • biomechanics
  • energetics
  • Forward dynamical simulation
  • rowing dynamics


Dive into the research topics of 'Mechanical power output in rowing should not be determined from oar forces and oar motion alone'. Together they form a unique fingerprint.

Cite this