Tips on using HRV to adjust training or behavior

I have recently been asked how I use the data once I have collected it, in the context of adjusting training or behavior. Here are some thoughts based on my experience, after many years.

For morning measurements, I keep an eye on patterns both in the short and long term.

Acutely, I don't really stress over a single suppression if I feel well (it can happen for a number of reasons, not all of them important). This is even more common with night data, where you can get a suppression just because you had a little more food in the evening (another reason why I recommend morning measurements, especially for athletes - see here and here).

Chronically, HRV tends to track with my health, work stress, or seasonal issues (e.g. allergies), and it can be a useful reminder to take better care of myself (this can be done in different ways, prioritizing things like sleep and eating well or reducing training intensity at times). In many of these cases, the data simply reflects how I feel: I can exercise, and find benefit in it, but I am not in the right mental or physical condition for hard, structured training. The objective feedback helps me accepting the situation and giving my body the time it needs to bounce back.

It's easy to think that we can do that with 'feel', and by all means, I believe 'feel' is the most important thing, but that self-awareness can be improved by putting side by side our subjective feel (e.g. collected via a morning questionnaire as we do in HRV4Training) and using objective data. Even now, after so many years, I find that the objective data can be useful to take a step back and don't get too much into my head, so to speak. This is also why I think the questionnaire in the app is so important: these tools are not just about objective data, it is about self-assessment, and using the data to improve our self-awareness, not to dismiss it in favor of what we are measuring.

Altering training

If I have a sustained suppression, e.g. several days below my normal range, typically it reflects how I feel, hence I wouldn't really do hard training (I tend to have headaches or other issues that don't really get you in the mood for that type of work), but if I feel better than the data reflects, which can also happen, and I still have multiple days of low values, I do adjust training to reduce intensity (I do not touch volume normally, as I am trained to handle a lot of low-intensity work, which has minimal impact on autonomic activity, see here).

The principle is that not only the type of stimulus, but also the timing of it, is important, and therefore we want to apply high intensity when we are ready to assimilate that stimulus. Our ability to do so might be compromised when we are already responding poorly to stress, as highlighted by a low HRV (key to understand here that HRV does not represent stress, but our response to stress, and as such, a low value means something went wrong, because if we are under a lot of training stress and we are doing well, HRV is not suppressed (!)).

Using a more chronic approach, as opposed to an acute one (i.e. waiting for a sustained suppression that lasts a few days, instead of changing training at the first score below normal range) is more in line with recent research on HRV-guided training, which has also transitioned to this approach.

However, real life is not a study, and of course if we are not feeling well, it might be a good idea to adjust behavior right away. It is always more nuanced than simple rules.

Altering other aspects of behavior

You can always look at it from two angles, either you reduce training intensity, as just discussed, or you try to give priority to recovery (e.g. take an extra nap, and see if the day after you bounce back within your normal range). Training is a nice application for HRV because you can always modulate it, while we might not be able to address other stressors (e.g. work stress, family issues, etc.) - we should work in the long term to address these, but we cannot play with them on a day to day basis the way we can do with training.

While in studies different strategies are isolated, in real life we should probably try to attack the issue from multiple angles at the same time (e.g. reduce intensity, take the extra nap, eat well, find ways to live that work better for you in terms of mental health, etc. and see if you bounce back quickly).

Before trying any intervention, I would always collect data for a few months for an individual, just to see typical patterns and responses, e.g. what happens when you load more, when you add two back-to-back hard sessions, when you travel, when the climate changes and it gets warmer, etc. - there is only so much stress we can take.

Sometimes you don't need to change anything, but you can learn from it: e.g. I often do two hard sessions in a day, one running and one cycling, because I find it easier to be "on" for one day, than to be "on" for multiple days (this way, the other days of the week I can just do low-intensity high volume, which I can do regardless of my mental state). The first few times I would always get a suppression the morning after, while recently I seem to be able to bounce back within the day: you can adjust the stimulus, or you can wait to see if after applying it a few times, you get a better response - that's a way to track progress.

If the data looks good (i.e. HRV within normal range and a baseline that might even be increasing a bit), then you could also think of loading a bit more, but keep in mind that it all starts with proper training planning and periodiation, HRV is there to make small, meaningful adjustments, never to replace a well-thought plan (you still need your easy or rest days even when HRV is trending well).

These are just some ways in which I think the data can be helpful at different levels of athletic performance / training, and that I tend to apply myself.

I am leaving out here the obvious things everyone mentions all the time (alcohol, sickness) because hopefully we can grow up at a certain point, and I am more interested in using the data outside of situations that should be outliers for most people. What interests me is how the data changes in relation to more fundamental aspects of our lives, i.e. mental health, work-related stress, chronic health conditions, physical activity and training, and their association with health and performance, if that is something we are interested in.

I hope there are some useful pointers in there, thank you for asking!

---

Marco holds a PhD cum laude in applied machine learning, a M.Sc. cum laude in computer science engineering, and a M.Sc. cum laude in human movement sciences and high-performance coaching.

He has published more than 50 papers and patents at the intersection between physiology, health, technology, and human performance.

He is co-founder of HRV4Training, advisor at Oura, guest lecturer at VU Amsterdam, and editor for IEEE Pervasive Computing Magazine. He loves running.

Social:

• Twitter: @altini_marco.

• Personal Substack.

• Strava.

New to HRV4Training?