I have recently had to replace a faulty Garmin smart watch and chose a newer model, going from a Vivoactive 3 to a 5. This newer version provides several extra metrics to obsess over including Heart Rate Variability (“HRV”). After 2-3 weeks of round-the-clock wearing the watch will announce that it has established a Baseline and then proceeded to give you a report of your HRV Status. It tells you that everyone is different and there aren’t set values that are good or bad, but that if you are getting fitter then your HRV should trend upwards. If your 7-day average is within the baseline figures then your Status is “Balanced” but if it falls below then you are either “Unbalanced” or “Low”. You will then be advised to take things easy for a few days and ease off.
So far so good, but while you ease off in the attempt to get back to Balanced HRV Status a different bit of the Garmin experience, Training Status, will be nagging you to get out there and work harder 'cos you’re losing fitness. The net result is confusion and a certain amount of stress which is Not A Good Thing.
My question boils down to whether HRV monitoring as implemented by Garmin has some sound science behind i,t or does it just produce a collection of the Worried Well?
Yes; it’s the opposite of a placebo which is a dummy medicine that has no active ingredient but which can give benefit if the patient believes in it. A nocebo is something harmless in itself but which is detrimental to the user. Health metrics monitoring might be valuable but fretting about them can be bad for mental health.
Another example is how repeatedly hearing the same news bulletin can make you depressed if there’s bad news in it.
Nocebo aside, there is a vast amount of information available on HRV that is very interesting. Threads on this forum about it too.
I would avoid paying much if any attention to the HRV data or analysis given by a Garmin watch. It’s just not accurate enough, nor taken at the correct time in the correct way. Same for Whoop or Oura or any of these wearables.
HRV is best measured after waking waking each morning, always after following the same routine. So for example - wake, get up, go to bathroom if needed, put on HR monitor strap capable of taking accurate HRV measurements, sit down, take measurement.
Search for Marco Altini and Andrew Flatt online for more info on HRV
I understand that any wrist-based device for HR isn’t great for accuracy but have imagined that as long as it’s consistent then it’s ok to discern trends. Is that not true either?
Do you have a recommendation on a chest-strap device that IS suitable? And on how to capture the data. Thanks again. Nick
HRV is also a metric that one needs to look at over time (is the trend moving up/down or stable), and not necessarily on the day. Use it in conjunction with how you feel to get a better informed decision on how to train.
All this has been discussed in great detail in this huge thread:
You will find numerous links to trustworthy sources (surprise: Garmin is not part of that )
Some initial comments:
HR from a wrist optical sensor is already not of great quality. HRV from it is worthless.
HRV over 24hr is meaningless. It will just tell you exactly the same as Total Exercise Time or Load. Exercising is stress inducing, thus 24hr HRV will simply be lower on days that you exercised longer/harder. We already know that so it’s useless. HRV is a response parameter and must be measured at a consistent time, with a consistent protocol and a good quality sensor. Polar H10 connected to a good app like Kubios, EliteHRV, HRV4Training, etc will give you a usable result if you measure in the morning right after waking up. This will show how well your body is coping with the life-style stress and training from the former day. It will show how good you are recovering from it.
An upwards trending HRV does NOT mean that you are getting fitter. It means that you are better at coping with the actual life style stress, including training. If for some reason, you stop working out, HRV will increase because you will fully recover. But no longer exercising isn’t going to make you any fitter, right.
There are plenty of other remarks, but I’m going to stop here because all the info is available in the above mentioned thread.
Tickr has been reported in the same thread as working but ‘less reliable’. Search for Tickr in that thread, there are a few dozen posts about it.
At first sight, you could use it to see if the whole HRV story is something for you or not. If you find it useful after testing, I would advice upgrading to one of the sensors mentioned by @nasatt.
In a word, no. It’s not good enough. I have a Tickr myself and it just does not give accurate HRV data. Data from a Tickr is loaded with artifacts and tbh using a Tickr for HRV is about as useful as using an optical sensor from a watch.
To answer your earlier question re trends - again the watch is no good here. The data is inaccurate a lot of the time and inconsistent too. Somedays it’ll give readings much to high, the next day much too low and 5 days in between will be pretty OK. It just can’t be trusted though.
Dissenting opinion here. Mine doesn’t nag me to get back out there and work harder because I’m losing my fitness. In fact, it often says I need to relax or go easier because key metrics are low. Maybe that is your perception based on how you are interpreting it?
I find Garmin’s metrics pretty spot on with the Fitness and Fatigue metrics on intervals.isu once you understand them. The problem is most people (not saying you) don’t take time to understand them, the “Unproductive” message for instance. It simply means that your vo2 max measurement has gone down with what you are doing with current training. Some people don’t like to hear that, they just want constant validation regardless of what they do. You also need to make sure quality data is going in if you want quality recommendations.
Garmin’s Training Readiness is pretty spot on for me, I think it’s more accurate because I have a power meter and use a chest strap for all workouts, I suspect it’s less accurate for runners. It takes HRV, rest, recovery, sleep and a wealth of other metrics into account. It usually aligns very well with intervals.icu.
Lastly, Garmin doesn’t just analyze HRV for a 24hr period, it analyzes it over the life of your data. It even predicted when I got covid, HRV started falling greatly about 3 days out, told me to ease up, then went Unbalanced and I had my positive test.
Garmin’s overnight HRV data from a watch is garbage. It is the polar opposite of ‘quality data’. It’s this overnight sleeping HRV data that Garmin uses in it’s analysis of training readiness. It’s utterly meaningless.
The Garmin HRV taken during a morning ‘Health Snapshot’ can be close to correct a lot, if not most of the time. But even that still throws out some crazy readings at times. However, even if it was as accurate as a good HR strap, that snapshot data is not used in training readiness analysis by Garmin.
The overnight HRV is just one tiny component of Training Readiness. And it’s not just overnight that is measured, it’s analysis over time. My personal data, compared to intervals.icu, anecdotally looks like it much better than garbage or meaningless.
Maybe they have science behind it. If they do, it doesn’t hold much water tbh. The world’s leading experts on HRV, or at least those who publish data, are consistent in their assertion that effective HRV analysis requires morning readings which are taken after following the same routine time after time after time. Nightime measurements can not, and will not ever, be able to be taken under the same circumstances every night. Further to that, the best optical HR sensors available to consumers today are just about able to record nearly accurate HR, and even at that, they can’t do it all the time. HRV data from a watch is worse than useless. Maybe in time, some years down the line, it might become accurate enough to be useful.
When you say ‘compared to Intervals.icu’ what are you referring to? Intervals in itself doesn’t calculate or give a readiness score. Do you mean the ithlete chart?
The patients in the study you linked to had the data recorded under strict conditions and were almost certainly sitting as still as possible during the readings. They were not taken as an average over a night time sleep.
Point taken, and yes, I mean that my Training Readiness on Garmin tracks my Fitness, Form and Fatigue on intervals.icu very closely. When my Form dips into the red “High Risk” area, Garmin is always in a similar spot for Training Readiness. The Recovery time Garmin recommends usually corresponds very closely to the timeframe intervals.icu shows to get back to Optimal. Garmin’s FTP and LTHR estimates line up pretty closely to intervals.icu and TrainerRoad. But I give it quality data and I make sure any data I’m giving it is analyzed (data from Wahoo, MyWhoosh, IndieVelo, etc is not, Zwift is). One blogger I saw complaining about Garmin metrics was using a Wahoo bike computer, of course the metrics will be off then as Garmin doesn’t use that data. Whether they should is another story, I think they should as their metrics are then all wrong if they just ignore that activity.
I use both intervals.icu and Garmin. I mostly use Garmin for quick glancing and just to make sure I’m on track. I think it varies between watches, but the FR955 takes into account Sleep Score, Recovery time, HRV status, acute load, chronic load, Sleep History and Stress History into the Training Readiness, so it’s far more than just overnight HRV. The more data you feed it, and the more time it has, the more accurate I feel it is.
I accept pretty much all of what you’re saying. It’s just that the OP was specifically asking about Garmin HRV and that, in isolation, is rubbish and should be ignored.
I have a Garmin watch myself. It’s handy to have. It provides a backup of all activities, if nothing else! Useful for any jogging or walking I do so I don’t need to wear a HR strap. HR data recorded during walks or runs is remarkably accurate.
Sleep start and wake times are remarkably accurate also. Not so sure about the sleep stages data (deep sleep, REM etc). Overnight HRV, as we’ve established, is a non runner. The data is handy to glance through and keep an eye on but I’d be hesitant to pay too much attention to it or over analyse it, at least on an acute basis.
Why bother measuring a supplemental parameter if it returns the same result as other parameters that are way easier and less complex to measure? The way HRV is analysed here results in a value that is very closely mirroring Fitness/Fatigue. Simply because HRV is low during any stress moment. If you measure over 24hr, the result is that any day with a long training/race/event results in a low HRV number. That’s exactly the same as a high Fatigue number… As a consequence, you take one or more rest days resulting in an increasing HRV. Nothing to wonder about because rest days have far less stress moments and it is pure logic that HRV over the full day will be higher. What did you learn out of this? Nothing that you didn’t already know.
A Readiness score is effectively the result of a bunch of different parameters, thrown together, to brew some sort of score that should tell how you’re doing. Completely ridiculous because a long hard training will reduce HRV and on top of that, the readiness score throws in your load/volume of the day before, making the score even worse. You’re double punished this way. Mixing up inputs and outputs to calculate such a score is wrong. The Fitness /Fatigue chart shows you what could/should/might happen with the inputs you’re giving it. A HRV spotcheck in a consistent way shows you what the result is. It shows you the outcome of the impact those inputs have on your body. If you do a long training today, the HRV result tomorrow will show how well you’re coping with that load. That’s what we’re interested in and not a twin parameter of load that tells us nothing new. Taking HRV measurements the way it is intended to do, will give you more insight in what kind of training will need more or less recovery. That is valuable information.
So, no, that is false. It should read: the more it is an alternative to the PMC chart (=Fitness chart).
Stick to the correct way of doing things. Perform actions and look at what kind of reactions they cause. Don’t just predict that action A will result in reaction B.
The exact same logic goes for Sleep. It’s not because you had a bad night that Readiness must be low. It can, but it surely isn’t always like that. This too is a mix of inputs and outputs and results in a meaningless number that isn’t giving us any valuable information.
Overnight HRV with a good sensor can be more meaningful but still has some flaws. If you train late in the evening, a big part of the night will have low HRV because your body is healing itself during the night. That healing process can be entirely done by the morning and give you a positive HRV score in the morning. While overnight HRV will be a lower score, simply because it has a lot of data while the healing process was in full action. It is also more prone to movement artefacts and records through all sleep stages which can have very dissimilar HRV.
If you want to read more about all that, check out Marco Altini’s blog here:
)