Technical vs Biological variability in velocity-measuring devices, creating energy from snow, and resiliency
📝 Weekly paper summary
Context
Researchers and practitioners are leveraging technological advances to improve the physical preparation of athletes. For example, velocity-measuring devices are becoming popular tools to monitor how people move, provide feedback, set goals, or enhance motivation and competitiveness in training environments.
However, the benefit this technology provides is contingent somewhat on its validity, but certainly its reliability. Most studies to date assessing the validity and reliability of various velocity-measuring tools have done so by "lumping together" the biological and technical variability into a single measure. However, as discussed in multiple previous newsletters, movement is inherently variable. Therefore, it is crucial when assessing these devices that the technical versus biological variability is differentiated to understand the device's capacity better. This investigation aimed to determine the technical and biological variability in the FLEX velocity-measuring device. Understanding the differences between the two is imperative when using this data to drive training decisions.
Correctness
The study itself was relatively straightforward, given that it was a validation study. The study's main strength is that it used a calibration rig and humans to validate the device. This methodology is unique relative to most of the literature.
I would like to have seen this study (and other validation studies in this space) do differently to base their validation analysis on the movement velocity rather than the relative load lifted. Although the relative load lifted will likely correspond to different velocities, the movement velocities themselves may influence the device's validity.
Another potential limitation of this study is that the derivation of the p-values in their mixed-effects models compared to Flex's velocities and motion capture system wasn't made clear. Although not impossible, it's also unclear what degrees of freedom are in mixed-effects models. Thus there are different ways of generating p-values, with some being better than others (consider reading Evaluating significance in linear mixed-effects models in R by Steven Luke (2017)). Since their validation analysis interpreted p-values, knowing how they converted their mixed-effects model outputs is essential.
Contributions
- There were no significant differences between Flex and the motion capture system during the barbell bench press when lifting loads of 40, 60, 80, and >=90%1RM. For the back squat, there were no significant differences between the two systems when lifting 20, 80, and >=90%1RM. Where there were significant differences, Flex's velocities were significantly slower than the motion capture system.
- The overall interunit reliability (i.e., the coefficient of variation of velocities measured with two Flex devices) was about 3.96% when tested in a calibrated rig (i.e., technical variability in the measurement).
- The overall interunit reliability (i.e., the coefficient of variation of velocities measured with two Flex devices) with the addition of human movement was 9.82% for the back squat and 9.83% for the bench press.
- The between-day reliability of FLEX was 2.71-5.93%.
- Although this is seemingly a low CV for the device, researchers often claim that practitioners can use velocity data to estimate %1RM in the literature. Suppose we used %1RMs to assign a 100kg training load. Would we expect our weight plates to have errors of +/- 4kg (i.e., a CV of about 4%) every time we load 100kg on the bar? How about +/- 9.8kg (i.e., CV of about 9.8%)? Likely not. Therefore, while these data suggest that velocity data from FLEX is reliable enough to be used for several training purposes (e.g., enhancing training motivation), we need to be cautious about using these devices to prescribe specific %1RM training loads.
🧠 Fun fact of the week
Tis the season for winter-related fun facts! Did you know there is a device that can generate energy from the snow? Snow is positively charged and gives up electrons. Silicon, on the other hand, is negatively charged. When falling snow contacts the surface of silicone, a device known as a snow-based triboelectric nanogenerator (TENG) extracts the charge produced between the two materials and creates electricity! You can check it out for those interested in the paper at this link.
🎙 Podcast recommendation
A short but good podcast!
🗣 Quote of the week
"It's your reaction to adversity, not adversity itself, that determines how your life's story will develop."
- Dieter F. Uchtdorf