Issues with night data and made-up scores based on sleep

for actionability, focus on the physiology and measure in the morning

In this post, we look at a few important aspects associated to measuring resting physiology and stress responses:

• Measurement timing (morning vs night).

• Measurement position (sitting vs standing).

• Actual physiology (HRV) vs made-up scores (e.g. readiness).

It is not my intention to pick on any brand, but it is my intention to make sure people can use physiological data effectively.

In the context of training, and with the goal of daily actionablity (i.e. we want to make changes for the day ahead, not only assess behavior for the previous day) the way to do this is really simple: HRV must be measured in the morning.

Here is an example: the other night I didn't sleep great, maybe because of eating out. Wearables gave me the usual “just stay in bed today”. Why? because night physiology was altered (e.g. higher resting heart rate).

All wearables will tell you the same when this happens (Oura, Whoop, Garmin, whatever). Why?

Night data comes too soon and is too close to the stressors to be useful for readiness / recovery assessment. HRV is useful when measured far from the stressors, to assess if you have re-normalized (i.e. responded well), or not.

It is perfectly normal for physiology to be changing during stress or shortly after, otherwise you would be dead.

Wearables are translating any normal physiological change into something pathological. You can't keep getting derailed mentally because of a bad meal, it’s nonsense.

As you can see, my physiology re-normalized towards the morning and was all good during my morning measurement with HRV4Training.

The same happens if the wearable detects poor sleep quality (what does that even mean?). Keep in mind that poor sleep, acutely, has zero impact on your performance.

A few days later, I was getting a bit sick, nothing too serious, a throat ache and a bit of tiredness. I spent more hours in bed, as I felt fatigued. Here is my data:

Resting physiology, measured while seated, in the morning, could capture well my suboptimal state (see also this article). On the other hand, made up scores tend to always give you better numbers just because you spent more time in bed.

Please ignore made-up scores.

If you are interested in your body’s response to stress, you need to look at your morning physiology in comparison to your normal range, and that’s all. For training purposes, there is no better way.

No continuous data, no made-up numbers, no unnecessary stress or poor guidance.

The only meaningful data to add is your subjective feel, first thing in the morning, which I discuss in more detail here.

Have a good one!

---

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?