domingjm wrote:
I don't have any experience with the start/stop detection algorithms. But pressing a button takes a fraction of a second before and after. We're talking less than 1% error on a typical set of 100s. Is that really not accurate enough for you?
I do agree though that there's something to be argued for start and finish technique if you're overly focused on the watch though, so there's that.
I think the time it takes to stop a wrist watch after a proper finish takes the better part of a second, if not more. Furthermore, the variability is probably pretty substantial. Surely there's a big difference in how quickly you can get it stopped when you touch with one hand vs. the other. Add in the delay from start button push to actual start (which is probably more consistent, but is still introducing further inaccuracy), and your getting data that's not very clean.
On a set of 300's, that margin of error probably isn't that critical. On 100's, it is, never mind 75's, 50's, and 25's.
If you get something useful out of a swim watch, with or without automatic stop-start detection, great for you. I wouldn't tell you not to use it. Just know that the data isn't very granular. And because it's not very granular, it's not valuable to a high-level swimmer.
"They're made of latex, not nitroglycerin"