New analysis of variability of rower’s technique
Our previous analysis of the variability in rowers' motions (RBN 2024/06) was based on variations between stroke cycles at each data point for each biomechanical variable. While that method provided valuable insights, it did not allow for the analysis of between-channel variation. This limitation made it impossible to assess the variation of derived indicators (such as Catch and Rowing Style Factors) and the synchronization of rowers within a crew (e.g., how consistently rowers match the timing of direction changes at the catch).
Recent BioRow developments have addressed this gap, providing comprehensive information on both the variability of individual rowing techniques and crew synchronization. 20 indicators were selected for the analysis of variation. Across every sample taken during a test or race, four primary measures of variation were calculated for each indicator:
- Mean (average) value (Vav)
- Standard deviation (SD or σ)
- Minimum value (Vmin)
- Maximum value (Vmax)
Since the magnitudes of these indicators vary significantly, it would be inappropriate to evaluate their variation using only the standard deviation. Therefore, a coefficient of variation (CV) was used, which was defined as the ratio of SD to the absolute mean value:
The coefficient of variation (CV) works well for most indicators, except for timing-related ones (e.g., Catch Factor and others mentioned on pp. 16–20 above). This is because their absolute mean value could be zero or close to zero, making the CV undefined or approaching infinity. For these indicators, it is more appropriate to use the standard deviation (SD) or the absolute range (Rabs) as measures of variation.
An initial analysis of the collected data revealed that.
The new method of variation analysis has already been implemented in the routine BioRow technique assessment, with the results incorporated into testing reports.
Key variables were compared during a standard 2000m step-rate test for of two M1x scullers of different levels: a national-level junior athlete (left) and an elite, Olympic-level sculler (right). Although the magnitude of oar angles was similar for both scullers, the junior athlete's variation (mean 1.2% across all samples) was twice as high as that of the elite sculler (0.6%).
It was found that oar angles were much more consistent than both average force and WpS, which showed significantly higher variation across all rowers
The junior athlete applied nearly half the force and power of the elite sculler but exhibited more than double the variation: force variation was 5.1% compared to 2.3%, and WpS variation was 4.7% and 1.3%.
The newly developed features of the BioRow software have a range of practical and scientific applications, including:
- Providing a more comprehensive evaluation of rowers’ technique and skill levels.
- Assisting in crew selection, particularly when choosing a stroke rower, which is a frequent concern for coaches.
- Monitoring race or training performance in both short- and long-term perspectives.
- Enabling scientific studies on the correlation between rowing variability and factors such as rower’s level, age, gender, and psychological characteristics of rowers.
As always, feedback, questions, and suggestions are welcome
©2024 Dr. Valery Kleshnev
This is a short version of the Newsletter. To access the full text, please subscribe to BioRow membership here: https://biorow.com/membership/