Comparison of on-water rowing with machines

Did you know that… recently, biomechanical characteristics of on-water single sculling were compared with those observed during rowing on six different types of machines: Concept2 RowErg, PowerFun PF-580, TechnoGym, FirstDegree MegaPRO XL, RP3 Model S, Concept2 Dynamic?

To determine the effects of rowing intensity and external resistance factors, the testing protocol consisted of nine 1-minute trials performed under three resistance levels and three stroke rates. Ten junior rowers of international level (five males and five females) repeated the protocol seven times on each ergometer and in the single scull. On the machines, the BioRow system measured handle force and three displacements: at the handle, seat, and shoulders. On water, handle force was derived from measured oar bend, handle displacement was calculated from oar angles and inboard length, seat and shoulders movements were measured similarly.

At all resistance levels, the force curve in the boat increased more rapidly after the catch, while handle velocity increased more gradually than on rowing machines. This can be explained by the heavier dynamic oar gearing, which is not replicated in all machines.

Power values reported by the machines were systematically 1–2% lower than those recorded by the BioRow system, because the handle sensor measured total force, whereas ergometer systems detect only the force contributing to flywheel acceleration and do not account for shock chord force.

On-water rowing exhibited a distinctly different HDF pattern: it increased rapidly after the catch, peaking at approximately 10% of stroke length. In contrast, on all ergometers, the highest resistance occurred later, at 30–50% of stroke length.

DF recorded from each machine’s monitor had extremely high correlations (r > 0.999) with the average HDF over the stroke cycle measured with BioRow sensors, indicating perfect validity of the measurements. Furthermore, this supports HDF as a universal measure of external resistance for both on-water rowing and rowing machines.

The features of each rowing machine can be summarized as follows:

Concept2 RowErg is the de facto standard in rowing conditioning, supported by extensive global performance data and established training models. It provides accurate power calculation and smooth, comfortable HDF dynamics throughout the drive, approximating those of a single scull (at damper settings 8–10). However, it does not replicate on-water rowing in terms of force/velocity curves and HDF pattern.

PowerFun is nearly identical to the Concept2 mechanically and reproduces its positive characteristics. It shares the same limitations and additionally reports power values approximately 1.5% lower than on RowErg, requiring adjustments in results and loads.

TechnoGym is mechanically similar to the Concept2 and shares its positive attributes. However, it demonstrates low reliability in power calculations, which were approximately 8.5% lower than on RowErg.

RP3 Model S quite well resembles on-water rowing in terms of force curve and shows acceptable reliability in power calculation. However, damper settings have minimal influence on HDF resistance factor (corrected in the T model), and power values displayed approximately 1.5% higher than on RowErg, requiring correction in training load prescription.

FirstDegree MegaPRO XL is the only hydro-resistive machine tested (all others were aero-resistive) and provides the most similar force curve to on-water rowing. However, it has an unreliable and inaccurate power calculation system, incompatible with the Concept2 standard, and exhibits the highest HDF among all machines, resulting in a substantially heavier rowing sensation than in the boat.

Concept2 Dynamic is a highly specific machine and the only device in which handle velocity decreases toward the finish, producing a sensation of a “viscous” stroke lacking acceleration. Power values were reported approximately 14% lower than on RowErg.

Acknowledgements: Thanks to Russian Rowing Federation for organising this study and personally to Alexander Kleshnev for his extensive work with these measurements.