They're trying to to detect heart issues in people who don't otherwise have any reason to suspect they have them. Given that the heart rate monitor isn't even accurate enough to be useful for exercising[1], it seems unlikely you'd want to rely on this if you already know you have an issue.

http://www.mdpi.com/2075-4426/7/2/3/htm

Curious why you come to that conclusion? That study itself says the watches were within acceptable error for their purposes. Why would you need more accuracy than the roughly 2% error on the Apple Watch?

The error-rate for walking was up to 3.8%, which is roughly 8bpm for someone with a 200bpm max HR. And that error rate is averaged over a minute, so presumably the instantaneous error rate is even greater. If I'm trying to keep my HR at a constant 150 or whatever, I don't want to be speeding up or slowing down on the basis of getting garbage data from my HR monitor. This is an issue even with the Polar chest straps, and those are significantly more accurate than the Apple Watch.

Also, if you look at the protocol the vast majority of participants wouldn't have even broken a sweat, let alone have been sweating significantly.

In what health scenario do you need your HR to stay at a constant 150 so stringently?

If you're trying to do UT1 (70%-80% of max HR) training and instead you're doing UT2 (55%-70%) or AT (80%-85%), then you're not going to get the benefits you were going for.

Further, the only way you can accurately measure your rate of improvement is by comparing average watts across time at a fixed HR. So if your HR isn't at your target level then not only are you not going to be making progress at an acceptable rate, but you're not even going to know that you're not on track until it's already too late.

Agree with everything you said. But if you're running for 90 minutes let's say, checking in on the last 1 minute average heart rate should be quite sufficient to set your pace.

And I certainly expect the app would easily be able to show you average watts vs heart rate over time. Given weight, height, pace via GPS, and heart rate for every run you could definitely do the necessary analytics.

The watch isn't showing you the one-minute average though, it's showing the instantaneous value but the researchers averaged it for the purpose of the study.

Sounds like a pretty easy software fix for a running app. Which means it must already exist, right?

Your assuming that your other measurement devices don't have any significant error rates.

And that the gobblygook of UT1, UT2, etc, is actually meaningful.

It's not gobblygook, it's training at a serious level. Those levels correspond well with lactic acid levels in the blood, for various types of training. I've both rowed at a high level and done tech research work with an Olympic cycling team, and this (or variants) are used frequently. HR for power (not pace etc) is a good measure of personal fitness and wellbeing (e.g. high resting heart rate means you're possibly falling ill or not well rested). However, all this doesn't matter for anyone who just wants to be healthy. That said, I defer to Garmin For proper HR smart watches and accessories, which are tailored for specific sports.

If you don't want to wait, we launched a study with UCSF Cardiology which showed 97% accuracy at detecting atrial fibrillation in May:

https://itunes.apple.com/us/app/cardiogram/id1000017994?ls=1... cnbc.com/2017/05/11/apple-watch-uses-cardiogram-to-screen-for-heart-rhythm-abnormalities.html

So the Apple Watch is in an accurate heart rate monitor? Or is high precision not needed for testing for this abnormality? I want highly accurate rate monitoring throughout the day, especially on long hikes.

I've found mine is accurate as long as it is snug. But if it's not snug, it can be way off.

Also, when I first tried it I had wrist pain, and learned that I actually was "wearing it wrong". Moving it further up my arm fixed it, and now it's literally my favourite device.

Hrmm. I'll await the in depth reviews. For now, my chest strap will keep monitoring for me.

I'm talking about Series 2. Series 3 isn't available yet. But Series 3 is supposedly same size as Series 2, except the thing that sticks out of the back is "2 sheets of paper thicker", to use Apple's new contribution to the imperial system :)

So my guess is it will be the same: just as good as my Polar chest strap when snug, and off when, well, off.