Hi! My name is Judd Jones, and I am happy to be one of the initial users for Instabeat! I am a competitive masters swimmer, based in Knoxville, TN, USA, and I train for pool events (not open water or triathlon). I swim nearly every day, and I’m trying to keep careful notes about my Instabeat experience. I have a Ph.D. In bioengineering, and a lot of experience with physiological measuring instruments, so I am trying to be careful and systematic.

Step 1: Goggle Fitting

For me, the Instabeat (which I’ll call “IB”) didn’t fit well on my Speedo Vanquishers, so I went to a local store and tried various different goggles. The goggles with flexible gaskets seemed to fit the IB much better than goggles with rigid gaskets. I picked a pair of TYR Nest Pros, which have a more rounded shape, and seemed to me to fit the more rounded shape of IB the best. It looked like the edge of the goggles might be interfering with the IB sensor, so using scissors, I trimmed a little silicone away from the edge. The photos show the goggles before and after trimming.


goggle fitting judd 1 goggle fitting judd 2


Step 2: Resting HR measurement

The IB web site suggests measuring your resting heart rate (HR) by hand, but I figured, why not use the Instabeat? So the first thing I did to test the IB was to put it on immediately after waking up. Not only should this tell me my average resting HR, it should also give an idea of HR variability. The screenshot shows the dashboard session for my resting HR. After an initial, likely incorrect, spike to 80 bpm, the IB reports that my HR stayed at about in the high 40s / lower 50s for half an hour, with an average of about 48, with some variability. This looks pretty good to me. It’s about what I expect resting heart rate to look like. I actually fell asleep during this session. The Nest Pro goggles + Instabeat were so comfortable, I even fell asleep wearing them!

step 2 judd resting HR


Step 3: Initial pool testing

The next thing I wanted to do was to test IB in the pool. For an initial test, I selected 10 x 100 freestyle on a 2:00 minute interval, descend 1-4 to 400 pace & hold. The idea was to get my HR high enough (400 pace), but allow enough time between 100s allow my HR to recover somewhat. Then I should get a nice up and down pattern. I also carefully synchronized my pace clock to the beginning of the IB session, so I could tell how the times on the dashboard session correlate with the times on my pace clock.

In the screenshot, you can clearly see 10 peaks corresponding to the 10 intervals. The max HR at the peaks increases from 1 to 4 or  5, and then stays pretty constant, which corresponds nicely to the set design, “descend 1-4 and hold”. At the end of each interval, I took my heart rate by hand via the carotid artery. By hand, I am getting  around 130 BPM, and that’s roughly what the IB reports at the peaks, so I think IB is getting about the right HR at the peaks. The IB held onto my HR signal for almost the entire set. That’s the good news, and it’s very good indeed.

The not-so-good news is that during the 9 visible troughs, the HRs go to unrealistically low values. I’ve indicated a point where it says HR is 34. That’s too low. Moreover, it happens to a greater or lesser extent during all the troughs, so something is wrong there. Also, the peaks and troughs were not occurring when I expected them to in time, so that suggested another test.

step 3 judd 1 step 3 judd 2



step 3 judd graph


Step 4: A more challenging test

For the next test, I selected 5 x 50 freestyle, all-out, on a 3:00 minute interval. This test should drive my HR high for a short time, and then give it plenty of time to recover. Again, I carefully synchronized my pace clock to the beginning of the IB session, so I could tell how the times on the dashboard correlate with the times on my pace clock.

The screenshot shows the dashboard session for this set. The first 5 minutes shows my HR recovery from the previous set. Then you can see exactly 5 deep troughs, followed by 5 peaks, corresponding to the 5 x 50. The troughs happen right at the beginning of each interval; curiously, the peaks happen after I stop swimming!

I think this test and the previous test shows that IB is doing a good job at tracking the maxima in HR, and also at tracking HR recovery. But there are spurious, and quite reproducible, unrealistically low minimums. There’s also probably some kind of delay between when the HR actually hits a maximum, and when the IB reports it. I’ll have to do some further tests to verify this. Also, the lap counting is not particularly working for me.

step 4 judd graph

Step 5: A note on maximum HR

To calculate your training ranges, you’ll want to know your maximum HR. Most sources give you some kind of formula for calculating maximum HR, given your age. The most commonly used formula “220 – age”, is known to be wrong for some age ranges. Moreover, there’s a lot of individual variability in maximum HR, which isn’t accounted for by the formulas.

The graph below shows the correlation of HR with age in over 3000 healthy adults in a recent study from the University of Trondheim. If you want to use a formula, the ones given here are a bit better,

Men: max HR = 213 – 0.65 * age

Women: max HR = 210 – 0.62 * age

But, as you can tell from the graph, your true max HR could be much different. The standard deviation at each age is about 10  beats per minute. You can use a formula as a guide, but don’t take it too literally.

step 5 judd graph




My wishlist:

  • I suggest is to doubling the time before Instabeat turns itself off to give it a better chance to re-find the HR signal.
  • Improved the HR accuracy, particularly in the troughs.
  • Reduced delay
  • Improved lap counting




FacebooktwitterlinkedinmailFacebooktwitterlinkedinmailby feather