31 Comments
User's avatar
Gabin Aguayo's avatar

Thank you for your insights Marco, always appreciate them! I recently got a Garmin watch and HRV was one of the many things I wanted the watch and its analytics to help me with. I had downloaded and used (not frequently I must admit) the HRV4Training app, since I wanted to get familiar with its use and applications to use with my athletes. However, as I mentioned, I wasn't too diligent with it (I wake up kind of in a hurry every day to coach an early class). Anyway, I think the HRV status from Garmin is pretty useful, albeit imperfect as you mention, since it may consider some "extra" information from the entire night, but the average of the whole night would still change from night to night based on suppression from stress. I realize sleeping HRV may be less useful than sitting or standing HRV after sleep, but I believe you can still notice the same patterns and signals (although they might be a little less clear because of the noise). Maybe I'm just trying to justify myself, what do you think?

Expand full comment
Marco Altini's avatar

I agree with you Gabin. While I try to highlight differences and nuances that I think are important, as otherwise we just fall pray of dubious marketing tactics, there is no doubt that both methods track similar long term trends, and are effective ways to track physiological responses. When making it a habit to measure in the morning is not possible, the data you collect with Garmin is a good alternative. Keep in mind that Garmin uses the baseline vs normal range approach, and therefore is always delayed with respect to what is happening on a daily basis (I discuss this here in the "research vs real life" section: https://marcoaltini.substack.com/p/how-to-include-heart-rate-variability - take care!

Expand full comment
Gabin Aguayo's avatar

I’m catching up on your articles, that will be the next one! But yes, the HRV status from Garmin uses the 7D average, the “baseline” values, AND also shows last night’s HRV, so you can always look at the acute response and make a decision regardless of what the 7D avg shows or what the watch recommends.

Expand full comment
Marco Altini's avatar

exactly! Thank you

Expand full comment
Michael Alston's avatar

Thank you Marco a very interesting and balanced perspective. I am fascinated by my bodies response to stressors particularly the differences created by physiological vs pathological stimulus as I try to age gracefully. It’s been an evolving journey as I have moved from a purely knowledge based perspective to a more wisdom based one creating a more appreciative interpretation. Marco you have been a valuable guide for me on that journey. Sam wearables of any brand provide so much more than HRV data, they even tell the time I wouldn’t be without mine.

Expand full comment
Daithí's avatar

Hi Marco, thanks for a very informative article. For a morning measurement, would a heart rate strap be more accurate than the HRV4 app, as it is closer to the source?

Expand full comment
Marco Altini's avatar

thanks Daithí! If you don't mind using a strap, I think that's always a great option

see also here for some considerations on this question: https://marcoaltini.substack.com/p/q-and-a-could-i-get-more-accurate

have a nice weekend

Expand full comment
Daithí's avatar

Thanks for such a quick response, just looked at that article. Polar H10 sounds good

Expand full comment
Daithí's avatar

If you don't mind: in your opinion, is HRV tested in this manner the best measurement of stress?

Expand full comment
Daithí's avatar

My apologies, I might not have been very clear: what I mean is whether HRV is the best measurement of stress, instead of alternative metrics?

Expand full comment
Marco Altini's avatar

got it.

Stressors lead to a physiological response that impact autonomic nervous system activity and hormonal activity. Hormones cannot be measured with high frequency and reliability, hence at this point there is no practical use (even though it would be as relevant to be able to capture in similar ways things like cortisol). Autonomic nervous system activity cannot be measured either, but the autonomic nervous system modulates heart rate, and as such, HRV becomes the best non-invasive marker of stress in real life contexts (assuming we measure it at the right time and all the other things discussed here). Your subjective feel is another option :)

Expand full comment
Carlos Reategui's avatar

Have you had any experience with the Frontier X2? I started using it about 4 months ago. It's a chest strap but claims to capture continuous ECG data. It appears to capture HRV for a session (sleep too) but not exactly sure how to interpret it as it is a graph. For workouts doesn't seem useful as it is mostly < 5 ms. Don't always use it at night but when I do, it seems I should look at the average but that seems to shift as the night goes on.

Expand full comment
Marco Altini's avatar

hello Carlos, I used this device in the past as a reference, when looking at the accuracy of some PPG wearables. It is okay even if not comfortable and (at that time) it would report slightly incorrect rMSSD, so I had to export the ECG and recompute everything, not sure if they fixed this. In any case, the average of the night should be used if you prefer to use night data instead of a morning measurement, otherwise the device itself can probably be used with the HRV4Training app first thing in the morning, just like a Polar strap, if it sends the RR intervals according to standard protocols.

Expand full comment
Carlos Reategui's avatar

Yeah comfort is a bit of an issue for sleep tracking which is why I don't sleep with it regularly. And if you side sleep I doubt it will get good readings.

How can one tell if it is reporting bad rMSSD?

I do have an H10 which I have used with your app. My biggest struggle is consistency in measurement protocol. I find it very hard not to think about my breathing, heart rate, etc while taking a measurement and subconsciously trying to will a good result, lol. Doing a few back to back I sometimes get fairly different results.

Maybe this is where there is room for one of these types of devices/wearbles to at least provide consistent (relative to itself) results ??

Expand full comment
Marco Altini's avatar

yes for sure. There is no best protocol for everyone, we need to find what works for us and understand the differences in the various approaches, so that we are also aware of possible limitations or simply of the implications wrt actionability, sensitivity to stress, etc.

Regarding rMSSD, I had recomputed it from the raw ECG and it differed from what it was provided by the device.

Expand full comment
Shiong Tan's avatar

Thank you, Marco. I’ve used the HRV4Training app for a few months with my Oura ring and Garmin watch. I’m happy to say that the methods for measuring HRV provide the same/similar guidance for the upcoming day’s planned activities. That said, the HRV4Training app is more aligned with how I feel. Because it measures HRV first thing in the morning, it considers the end-state restorative impact of my sleep versus the average impact (which I assume the Oura ring and Garmin watch are doing when measuring HRV during the night and taking an “average”). So I find that if I had an excellent night’s sleep, I would have more confidence in doing a harder scheduled training session even though the Oura and Garmin statistics may be more borderline/circumspect about my HRV statistic vs the historical baseline when giving its guidance. Bottom line, as you said, if used correctly, at the right time in the proper context (i.e., in the morning), your HRV statistic can provide a very accurate steer on how to modulate your training - both in terms of the actual physical work you intend to do, but equally, psychologically, being kind to yourself to accept that some days aren't meant for going 100%.

Expand full comment
Marco Altini's avatar

Great take, thank you so much

Expand full comment
Chris Cocuzzo's avatar

Great post Marco. I don't think you're saying "never measure" but that is roughly what I take away.

What I mean is simply that you need lots of awareness and discipline to do HRV right, and without it, you aren't measuring anything worth knowing. So, it might be a better use of your time to avoid it unless you are committing to the process.

Broadly, I look at HR trends over time. Is my HRV rising over time? Is my resting HR falling over time? And I try to connect those data trends back to how I feel and what I'm doing (running, lifting, sleeping, eating, etc.). Not a doctor and not medical advice but I look across a minimum of 6 months, and often longer.

Usually don't need the data to confirm what I already know.

Expand full comment
AR Mokha's avatar

Thank you Marco for this blog. It's very useful read.

Expand full comment
patrick's avatar

Why do you push for hrv then? ;)

Expand full comment
Marco Altini's avatar

a very useful metric when used correctly :)

Expand full comment
Anna's avatar

Grateful for this blog, Marco.

Expand full comment
Steve's avatar

Thanks. Really helpful.

Expand full comment
Sam Tobias's avatar

This post has convinced me to pass on the purchase of a $350+ Garmin watch that captures continuous HRV data.

Expand full comment
Rui Pereira's avatar

Garmin captures HRV during sleep and shows the average of the night, the protocol used is a good substitute for a morning reading and is aligned with Oura ring for example. During the day apparently it captures HR and HRV for measuring “stress”, but this data is not computed for night HRV. I see no reason to pass on a Garmin, in fact I think it’s one of the best (and cheaper) options for capturing HRV during the night.

Expand full comment
Sam Tobias's avatar

Thanks Rui. I’ve been very happy with HRV4Training to monitor HRV. Its a great product. I’ll be purchasing a new Garmin, my Vivofit HR died in Mexico while swimming in the ocean. So much for being waterproof. I just won’t buy a Garmin that includes HRV & save myself $100. 👍🏼

Expand full comment
CecilRhodes's avatar

Informative as ever on the stress response & recovery. Will you do any additional work on the "DFA alpha 1" for workout monitoring?

Expand full comment
Marco Altini's avatar

thank you. I'll try to put something together on DFA as well in the next months.

Expand full comment
Brian Dickert's avatar

In the subheading should it read (flawed) instead of (falwed)?

Expand full comment