Help me understand HRV

Triathlon Forum
1

As with other devices, my Fenix 7 gives me a HRV score every morning. I tried to read and understand the Whoop! description of HRV, but I’m definitely missing something. As I understand the Whoop! description, higher HRV is “better”. I’m 62 and Whoop! shows the “average range” for my age at around 45-48 for the high. My number is increasing and is now running at 50. Fenix says I’m “unbalanced”, the same as if my HRV was low.

I do understand that HRV is like HR and you can’t really compare to anyone else. Is the Fenix “problem” that it’s comparing to a large group of people and I’m no longer “average”, or is it saying something else?

2

This resource has everything to get you started: https://www.hrv4training.com/quickstart-guide.html

I can’t comment on how the Fenix defines unbalanced, but you generally want stable, but steadily increasing HRV as you work through your training plan.

3

https://simplifaster.com/articles/misconceptions-hrv-training-data/

A very common pattern, especially for endurance athletes under high load, is to show unusually high HRV numbers after a big training block. A very low resting heart rate and general tiredness/fatigue generally accompany this. It is worth remembering that when we measure HRV, we measure the strength of the parasympathetic nervous system—the body’s recovery system. It makes sense that this system is very activated after a big load as our body runs all of the rest and repair processes essential to recovering from that big load. This doesn’t, however, indicate that our body is ready for more load at that point.

In practice, tracking your HRV along with another metric that also indicates the strength of your “let’s do work” system—like heart rate—can add very useful additional context.

the hrv number on its own is nearly meaningless… higher than usual HRV could mean well-rested and ready for a hard day, or exhausted and the body is frantically trying to recover from unusual stress… have to look at the history and other indicators as well. The Fenix scores are the result of a black box algorithm and I wouldn’t pay any attention to them.

https://medium.com/@altini_marco/the-ultimate-guide-to-heart-rate-variability-hrv-part-2-323a38213fbc

Once you start collecting some data, the most important thing to remember is that you always need to interpret your HRV data with respect to your historical data, and there is no point in comparing your HRV to another person.

4

As with other devices, my Fenix 7 gives me a HRV score every morning. I tried to read and understand the Whoop! description of HRV, but I’m definitely missing something. As I understand the Whoop! description, higher HRV is “better”. I’m 62 and Whoop! shows the “average range” for my age at around 45-48 for the high. My number is increasing and is now running at 50. Fenix says I’m “unbalanced”, the same as if my HRV was low.

I do understand that HRV is like HR and you can’t really compare to anyone else. Is the Fenix “problem” that it’s comparing to a large group of people and I’m no longer “average”, or is it saying something else?

It’s saying something else. Garmin’s definition of unbalanced is that your 7-day HRV average is outside of your personal baseline. It is unrelated to other people as far as I understand it. Here’s an explanation of how Garmin handles HRV and what “unbalanced” means:

https://www.garmin.com/en-US/garmin-technology/running-science/physiological-measurements/hrv-status/

5

The Garmin implementation is only comparing you to yourself.

So you’ve got an initial 19 night calibration period. During that timeframe you won’t get any color-coded scoring (e.g. High/Balanced/etc…). You will see your nightly trends and historical trends.

Once you clear the 19 night calibration period, then you’ll start to see the balanced/etc. In the case of Whoop, the Recovery score it gives you each morning is straight-up that last night score. This makes it a bit less useful, because it’s super variable. Whereas Garmin gives your HRV value, but the actual Balanced/etc is trended over a number of days (thus one night of drinking won’t kill your score).

The Garmin implementation will take that 19 days and expand it up to the last 90 days of HRV data, as a rolling calibration period.

In any event, where you can get into challenges with any HRV baselining is if you’ve done something significantly different during that initial calibration period. For example, one person I know got COVID the same day they got their HRV-capable device. So they basically calibrated with a crazy-low HRV value baseline, which…while geeky…slaughtered their baseline for quite some time, since they deal with low HRV values for a few weeks.

Inversely, if you were to baseline while sitting on a tropical island drinking alchohol free umbrella drinks for two weeks, you’d be baselined very high/relaxed.

6

Thanks for all the answers and pointers to additional articles.

I’ve had my F7 for several months, so it should be looking back to whatever the max number of days are.

What confuses me is that the HRV says “unbalanced” (is higher than my normal), but the readiness says “peaking”. My HRV trend is upward.

I have an Oly a week from Saturday and because of my schedule this year, this is a B+ or A- race for me (not really an “A” race, but close). So I do want to be peaking or max ready for the race. I have a cycling event this Saturday. If this HRV “unbalanced”/“peaking” means I shouldn’t go for a full 80 or 100 miles, I’ll stick to something like 50 at z2 with a sprinkle of z3. But if 80 actually helps me, I’ll do a z2+ 80.

If the weather is good, my goal is to break three hours at the Oly, which will be a PR for me. That’s why I now have this interest in understanding HRV better.

7

I’ve been surprised with how my measured HRV on my recent-gen Garmin 255 (has the upgraded HRM sensor that the older models don’t have) corresponds to training fatigue and load.

I originally ignored the data for 2 months as it was mostly the same, but then I’ve been doing blocks of significantly harder training with a lot more volume, and have been amazed at how the HRV responds exactly as the papers say they should - dips downwards with acute stress, then comes black slowly with recovery days.

Importantly, at least for me, it seems to correlate with how ‘energetic’ or ‘ready to train’ I feel for the day. I wouldn’t say it’s a big surprise of a result, as I know I’m feeling tired already and don’t necessarily need a number to tell me that again, but it’s encouraging that my HRV numbers don’t seem to be wacky at all, and actually seem really consistent.

I now check my HRV every day, I just use it as an additional piece of info to help influence my daily plan decisions. I was going to do an extra swim workout this morning since my schedule permits, but since it’s my pullback week, and I’ll feeling quite ‘blah’ from high weekend volume, I checked my HRV and it’s still ‘low’ per Garmin (red), so I opted to just go with how I felt and skipped the swim workout. We’ll see how this pans out long term, but given that I tend to overtrain rather than undertrain, I suspect I’ll be seeing similar or better results with less wasted workouts.

8

Inversely, if you were to baseline while sitting on a tropical island drinking alchohol free umbrella drinks for two weeks, you’d be baselined very high/relaxed.

Not me. I start climbing the walls at those kind of “vacations.”

More seriously, does anyone have a quantitative or qualitative assessment of Garmin v. Whoop w.r.t. HRV? I gave two serious attempts at Whoop, once with the older generation, and then with the newer. I gave up both times.

9

I have a fenix 6x pro (945/245 equivalent). I’ve tracked hrv on various devices periodically for many years. I find nearly zero correlation between hrv values, and training state. That’s for both magnitude and direction of change. I have numerous personal examples of hrv moving higher in response to a stimulus one week, and moving the opposite direction (to the same workout) the next week.

Further, I find no relationship between how I feel on any given day and my hrv status.

A perfect example… For the last 28 days my hrv has ranged from 21ms to 36ms, averaging 26ms. At the end of that block (last Thursday) my last workout was 1:30 with 45min of hill sprints. My hrv Friday morning at 28ms. I rested Fri, sat, sun… With hrv of 23, 27, 25ms respectively. Yesterday, I ran 45m (short) easy (first run in 4 days)… Today my hrv was 48ms…highest it has been this year.

It is not nearly so simple to interpret.

10

Further, I find no relationship between how I feel on any given day and my hrv status.

Yeah, same thing with my Whoop. I found this quantitative comparison, and don’t hold a ton of hope given it indicates Whoop 3.0 is quite good compared to competitors, yet I couldn’t find value in it.

11

I have a fenix 6x pro (945/245 equivalent). I’ve tracked hrv on various devices periodically for many years. I find nearly zero correlation between hrv values, and training state. That’s for both magnitude and direction of change. I have numerous personal examples of hrv moving higher in response to a stimulus one week, and moving the opposite direction (to the same workout) the next week.

Further, I find no relationship between how I feel on any given day and my hrv status.

A perfect example… For the last 28 days my hrv has ranged from 21ms to 36ms, averaging 26ms. At the end of that block (last Thursday) my last workout was 1:30 with 45min of hill sprints. My hrv Friday morning at 28ms. I rested Fri, sat, sun… With hrv of 23, 27, 25ms respectively. Yesterday, I ran 45m (short) easy (first run in 4 days)… Today my hrv was 48ms…highest it has been this year.

It is not nearly so simple to interpret.

I wonder if your Fenix 6 (wikipedia says released 2019) doesn’t have the most recent updated optical built-in HRM.

That could definitely account for your meaningless HRV numbers.

I wrote about this before, but I’ve been super impressed with the built-in optical HRM on my current-gen Garmin 255 (released June 2022, has an upgraded HRM that’s shared in modern models.)

On my Garmin 945 which I was using right before I upgraded, the optical HRM was almost trash. It was ONLY useful at rest, and barely if that. I always used a scosche+ for run workouts and strap for bike. It didn’t have the HRV stat when I got it, but if they added it in as a software update, I’m almost sure it would be useless given how unreliable that sensor was.

In comparison, I’ve never worn my scosche since getting the 255 since the built-in actually is BETTER than my scosche, with less errors on the initial phase. I strongly suspect having this modern-gen HRM is really key for getting good, reliable HRV data.

It’s actually shocking to me how strongly and reliably my HRV correlates to training load and ‘beat-down feeling’ I get.

12

I have tracked hrv with chest straps for all the years before I got the f6. I’ve used garmin connect iq apps, and hrv4training on a phone both paired with a chest strap. The results were always very similar to my f6 ohr.

So, perhaps the f6 ohr isn’t great, I’m not a huge fan of hrv using ohr anyway. I find the rmssd values to be very low compared to what I have gotten with a chest strap previously—somewhat expected given the blunted qrs from optical pulse rate.

Regardless, the random directional trends with respect to stimulus has been very typical for me for many years.

13

I have tracked hrv with chest straps for all the years before I got the f6. I’ve used garmin connect iq apps, and hrv4training on a phone both paired with a chest strap. The results were always very similar to my f6 ohr.

So, perhaps the f6 ohr isn’t great, I’m not a huge fan of hrv using ohr anyway. I find the rmssd values to be very low compared to what I have gotten with a chest strap previously—somewhat expected given the blunted qrs from optical pulse rate.

Regardless, the random directional trends with respect to stimulus has been very typical for me for many years.

I think you have to try the new device. It’s unlikely that guys like Dan Plews (who posts a bunch of his data in one of the articles linked above) is getting inconsistent results like yours - my results look a lot like his, with big drops after big days and slow recovery to normal when I’m recovering.

Doesn’t sound like it’ll work with your current equipment, but def don’t forget to take a look at it again if you get a recent-gen Garmin; I suspect you’ll be surprised with the results.

14

The gold standard for HRV is electrical (ecg) measurement of the r-r intervals between successive qrs waves. Not any optical device measuring pulse intervals, at all. I would be quite suspicious of any ohr device that showed data differently than an electrical measurement.

I’m aware that it works for some people… My point is it doesn’t work for all. My autonomic system seems to react differently than yours. However, I’ve had similar conversations with other reputable individuals who know how to take controlled data under repeatable conditions—eith observations similar to mine.

15

The gold standard for HRV is electrical (ecg) measurement of the r-r intervals between successive qrs waves. Not any optical device measuring pulse intervals, at all. I would be quite suspicious of any ohr device that showed data differently than an electrical measurement.

I’m aware that it works for some people… My point is it doesn’t work for all. My autonomic system seems to react differently than yours. However, I’ve had similar conversations with other reputable individuals who know how to take controlled data under repeatable conditions—eith observations similar to mine.

I see your point, but unless you have done the ECG-measurement yourself, I don’t think you can rule yourself out as an HRV-nonresponder.’

Similarly, the optical sensor in the current-gen Garmin seems to be giving me and others very plausible readings. Sure, I haven’t gold-standard ECG’d it, but it’s certainly correlating well with strain and recovery thus far - really well.

16

I see your point, but unless you have done the ECG-measurement yourself, I don’t think you can rule yourself out as an HRV-nonresponder.’

That’s what a quality chest strap provides. You don’t need multi-lead ecg for hrv measurement. You just need a precisely measured qrs peak.

Similarly, the optical sensor in the current-gen Garmin seems to be giving me and others very plausible readings. Sure, I haven’t gold-standard ECG’d it, but it’s certainly correlating well with strain and recovery thus far - really well.

Strain (and recovery) is just a transformed version of hrv trend analysis. So, it would be weird for it to NOT be correlated…regardless of any relationship to the real world.

17

I see your point, but unless you have done the ECG-measurement yourself, I don’t think you can rule yourself out as an HRV-nonresponder.’

That’s what a quality chest strap provides. You don’t need multi-lead ecg for hrv measurement. You just need a precisely measured qrs peak.

Similarly, the optical sensor in the current-gen Garmin seems to be giving me and others very plausible readings. Sure, I haven’t gold-standard ECG’d it, but it’s certainly correlating well with strain and recovery thus far - really well.

Strain (and recovery) is just a transformed version of hrv trend analysis. So, it would be weird for it to NOT be correlated…regardless of any relationship to the real world.

The new Garmins measure HRV during sleep, overnight. It takes a lot of readings across the entire night and plots the entire timeline overnight.

Are you wearing your HRM overnight and having your device take HRV measurements across the evening? Upon looking at my data again, it does seem that if you focus only on small 1-hr segments, the info can seem random, but by using the whole overnight average, it smoothes out to consistent daily patterns.

This is a sample screenshot of someone’s HRV page (not mine) that was posted on Reddit so you can see the ‘nightly timeline’

18

Taking more data that varies even more wildly and then averaging it down does not, in and of itself give you “better”, more meaningful data.

Yes, I have done all kinds of various measurements over the years. Yes, I wear my fenix6 24/7. Yes, I have the same nightly graph as any other garmin forerunner/fenix user. I’ve done the same overnight measurements with a chest strap. I’ve done 5m and 1m measurements under controlled conditions… To compare with overnight samples.

I never found anything actionable in hrv… Ie, it would have prevented me (correctly) from doing something I otherwise would have done. And it was wrong at least as often as it was right.

I really did want it to provide me with useful data and I spent a LOT of time trying to make sense of it and relate it to something physical. It wasn’t even as useful as simple rhr for me. The only meaningful use I found was that generally hrv trends up with fitness. So, you can see fitness trends in a long term increase in average hrv. But, you know… I can see that in run pace, too.

It WILL tell you when you are about to be deathly ill. Every time I have gotten the flu, my hrv drops out the floor (eg, by like half).

19

Taking more data that varies even more wildly and then averaging it down does not, in and of itself give you “better”, more meaningful data.

Yes, I have done all kinds of various measurements over the years. Yes, I wear my fenix6 24/7. Yes, I have the same nightly graph as any other garmin forerunner/fenix user. I’ve done the same overnight measurements with a chest strap. I’ve done 5m and 1m measurements under controlled conditions… To compare with overnight samples.

I never found anything actionable in hrv… Ie, it would have prevented me (correctly) from doing something I otherwise would have done. And it was wrong at least as often as it was right.

I really did want it to provide me with useful data and I spent a LOT of time trying to make sense of it and relate it to something physical. It wasn’t even as useful as simple rhr for me. The only meaningful use I found was that generally hrv trends up with fitness. So, you can see fitness trends in a long term increase in average hrv. But, you know… I can see that in run pace, too.

It WILL tell you when you are about to be deathly ill. Every time I have gotten the flu, my hrv drops out the floor (eg, by like half).

Well, I guess it sounds like it doesn’t work for you, unfortunately. If you get a new Garmin device, try it again - I suspect you’ll be surprised with the results, if you’d get anything similar to what I’m getting.

20

Taking more data that varies even more wildly and then averaging it down does not, in and of itself give you “better”, more meaningful data.

Yes, I have done all kinds of various measurements over the years. Yes, I wear my fenix6 24/7. Yes, I have the same nightly graph as any other garmin forerunner/fenix user. I’ve done the same overnight measurements with a chest strap. I’ve done 5m and 1m measurements under controlled conditions… To compare with overnight samples.

Throw away the overnight measurements, unlikely to be useful.

As Alan Couzens says in the article I mentioned above,

in my experience, you will find far better and more easily interpreted value in sticking with a single heart rate variability test: each morning, at the same time, under the same conditions.
It is challenging to interpret all-day HRV meaningfully. It is far better to have a controlled, resting test where your heart rate is relatively similar to assess significant changes in HRV.

I use the app HRV4Training and have found it reliable, it has kept me out of the bigger training/exhaustion holes since I started using it.
The author has a PhD in data science and MSc in sport science, says,

always measure at same time after waking up in morning.
measurements taken at other times can still be interesting, but for a different purpose