Heart-Rate Accuracy Revisited

I’ve gotten some odd results, especially when working out at higher heart rates, from my Whoop strap: it can give what seems to be spurious high heart readings. It is especially bad when doing exercise with moving arms, like using my kayak trainer or walking with hand weights. But it even happens while bicycling. I had this issue in the past with wrist-based heart rate measurements (with a Fitbit). It’s pretty important for me, as someone who had Afib in the past, to monitor HR accurately. So I bought a Polar H9 chest strap, which is highly regarded for accuracy. This was an inexpensive solution at $55. There is a free app “Polar Beat” that pairs with the strap, and turns your phone into a fitness tracker, using both GPS and heart rate. I used that to compare vs. the readings from my Whoop..

This first is a simple 5 minute easy spin on my recumbent stationery bike trainer. Max HR is 104, but Whoop shows a spurious spilke to 130 BPM:

This is about 50 Minute walking with hand weights, so arms swinging about waist high. WHOOP has lots of spurious variation, and overpredicts Max HR by 39 beats:

I’ve done multiple test since and the results are similar: the Whoop often throws in spurious spikes to high heart rate, even while bicycling. Unfortunately, this severely limits the usefulness of Whoop as a training app. It still is good for predicting heart-rate variability, for recovery purposes. But it’s kind of pricy to pay $30 a month just for that. You’re supposed to also be able to use the app for training suggestions (using its “strain coach”). And these just can’t be trusted because they are based on heart rate. For a good example of how spuriously high heart rate readings can be misleading, consider the last paragraph of this previous post, where I went for a long comfortable bike ride but was misled into thinking I had gone too hard. I now know the readings were probably incorrect and my pace was probably just fine.

I will continue to use the Whoop until December because I’m prepaid for this year, and I want to continue monitoring to see if my heart rate variability keeps improving over time, which is a good sign my training is good for my health. After that I’ll switch to a much cheaper alternative (see below).

And for monitoring the intensity of my training, I am happy to now have my new strap and Polar Beat App.

Inexpensive Alternative for Monitoring Heart Rate Variability and Recovery- Hrv4Training

Recently I discovered the website Hrv4Training. They have an app for monitoring heart rate variability, all you need is a smart phone, with a one-time cost of $12. Hrv4Training was founded by Dr. Marco Altini, who has a PhD and two MS degrees in relevant fields and has published extensively on heart rate variability in the peer-reviewed literature. So the science behind the app and the website is sound. Also, the accuracy of the app’s results has been verified vs Electrocardiogram methods. Amazingly, all that is needed is putting your finger over the smartphone’s camera. Heart rate and HRV are then measured with the camera with a technique called “Photoplethysmography” (quite a tongue twister). However, if you have another way to measure HR like a chest strap, the app accepts that also. So I connected my Polar strap to the Hrv4Training app. For the last week and a half I’ve been comparing its results with Whoop. The exact values cannot be compared directly because the proprietary algorithm used by Whoop for calculating HRV differs from that used by Hrv4Training. But the trends agree very nicely: On days when one app says “you’re recovery is great, you can go hard today”, or “you’re recovery is poor, take it easy today”, the other agrees. So I’m now confident I can continue on with the Hrv4Training app after my Whoop subscription expires. The combination of recovery info from Hrv4Training and training intensity from Polar Beat will tell me all I need to know.

I have one minor enhancement I suggested for the Polar Beat app. They use the inaccurate formula 220-age to estimate max heart rate. They then use that to give feedback on what zone you are training in, which can be highly misleading. In my case, for example, the 220-age estimate is 18 beats too low: my max HR is over 170, not 152. The app should allow the user to enter a more accurate number for max HR if it is known.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s