Read on Twitter
Just published my first preprint: https://www.preprints.org/manuscript/202008.0085/v1 In this article, we (me, @jesperbencke, @Merete_Moller and Henrik Sørensen) describe our (quite long) efforts to use a wearable device (similar to the picture) to determine throwing speed in handball. Thread.
We find that it works reasonably well, with mean absolute estimation errors ranging from 1.32 to 1.82 m/s. In comparison, we find that the mean difference in speed between a short pass and a penalty is 6.2 m/s.
We first worked on this idea back when I was doing my bachelor's thesis (in 2014!
) but found that the accelerometers we were using at that time could not measure high enough accelerations. I presented this data at @SCHMS2018 in 2016.
) but found that the accelerometers we were using at that time could not measure high enough accelerations. I presented this data at @SCHMS2018 in 2016.
So we went on to build our device, which was much less pretty, but capable of measuring +/- 200g, which is enough to reliably measure peak accelerations at the wrist. Almost exactly three years ago, we finished collected data from 19 participants = 475 throws recorded.
Then came other and more interesting/important project, I had to finish my masters degree (which included putting an LED-display on the device and wrapping it in silicone molding paste, somehow making it even uglier) and had my second child.
Finally, in March this year, we submitted the article for the first time! And got desk rejected. So we submitted it again in April! And got desk rejected again. Thankfully, the editors provided some valuable feedback and since then the manuscript have improved markedly IMHO.
The manuscript is now submitted to a third journal, but simultaneously we are releasing it as a preprint and would any feedback on how to improve the manuscript further.
Simultaneously, we are working with @behnam_liaghat and http://ccathletics.dk to improve the device packaging and further develop the method to also be capable of automatically distinguishing throws from non-throws, so we can count the number of throws. The end.
