Thread: Swim Computers
View Single Post
Old 11-21-2014
machelett machelett is offline
Join Date: May 2009
Posts: 68

Originally Posted by Polish TI fan View Post
But for an argument sake, I would point out that:

1. Garmin Swim promotional video suggests it counts every stoke (each hand movement) - watch the video at - look at 0:54 sec. of the movie and text displayed.
2. I thought this piece of kit was so clever that it would double the count itself...
3. I would have thought, the watch would actually add odd strokes when involuntary movements, not reduce them...
4. Doubling 9no. strokes up gives me 18 strokes, while I was doing 14... [edit: for a lap (2x lengths) I would have 28 strokes.., again not 18]
5. No one complained before, so I thought it had to be my problem. Does your watch count correctly? Does any other watch (forerunner?) count correctly?
Sure, I like getting to the bottom of things. I embrace every opportunity to look at things from a different perspective to find out whether I'm wrong or if I got it right. :)

First off, when I said "lap", in fact I meant "length". I fell into that common trap once again so I'll try to be more precise in the future. The Garmin Swim has no notion of a "lap" as in going down the pool and coming back. All it knows are individual lengths.

1. That part of the video is an idealized/artistic representation; some people might call it bogus. That's simply not how it works and Garmin should worry about being sued over false claims. ;)
There is no immediate display and how should the watch be able to count the other hand's stroke? If it were sensitive enough, it would count every bump in the road so to speak.
2. You have to be aware of the fact that not every length is freestyle. It might be fly or whatever else. The watch tries to guess the stroke type but gets it wrong quite often. Doubling the number based on that unreliable guess would aggravate the problem, for example, when you were actually swimming breaststroke and suddenly saw an outrageously high stroke count.
3. You assumption is correct.
4. It determines the strokes per length. So for one length you have 2 x 9 = 18, which makes it 36 for two lengths.
5. Correctness lies in the eye of the beholder in this case. I realize that there are technological limitations and therefore I'm aware that it tells me what movements it registered and not necessarily what I think it should tell me.
I'm not aware of any swim watch that uses a vastly different counting algorithm. The Forerunner series basically runs a similar software and you'll get the same results there.
I've been informed that my new 920XT is in the mail so I'll be able to compare that. From what I have read, I don't expect anything different when it comes to the stroke count.

I have noticed that the variation in stroke count gets smaller when I swim at tempo. It seems that the watch has more trouble distinguishing and interpreting slow, flowing, movements than fast and crisp ones.
For me, stroke count isn't anything that I closely monitor. I'll check from time to time but normally I'm indifferent. Sometimes, however, on a particularly good or bad day, I'll notice afterwards when looking at the average that it is significantly higher or lower than usual. So let's say that an "8" would be my normal average and suddenly it is "7" or "9".
That simply serves to confirm what I already felt, that it was a great day or a crap day, respectively. :)

Last edited by machelett : 11-21-2014 at 12:10 PM. Reason: Typo
Reply With Quote