What is HRV? How to read heart rate variability properly

You wake up. The watch says 28. Yesterday it said 44. The app has coloured the number red. Before your feet have touched the floor, your phone has decided you are in poor shape.

Then you go to work, run five kilometres in the evening and feel perfectly fine. So what exactly was red?

HRV is the most misunderstood number in consumer health tracking. It is also one of the most useful measures you can have, provided you know what it can and cannot do. This page covers both halves.

What HRV actually measures

Your heart does not beat like a metronome. One gap between beats might be 900 milliseconds, the next 940, the next 890. That difference is HRV, or heart rate variability.

The variation is governed by the autonomic nervous system, the part you do not consciously control. It has two branches. The sympathetic branch is the accelerator: it raises heart rate, mobilises energy and gets you ready to perform. The parasympathetic branch is the brake, running mainly through the vagus nerve. It lowers heart rate and handles digestion and recovery.

When the brake is engaged, it nudges the rhythm slightly between each beat. The gaps become uneven. HRV goes up. When the accelerator dominates, that fine adjustment gets pushed aside, the heart runs more evenly and HRV drops.

This is why HRV became the number every recovery app is built on. It does not measure how tired you feel. It measures how busy your nervous system is. Those two things are related, but they are not the same.

Why your number cannot be compared to anyone else's

This is the most important section on the page. It is also the part most HRV articles skip, because it makes the number harder to sell.

Two different measures share one name

There are several ways to calculate HRV from the same set of heartbeats. Two dominate.

• RMSSD looks at the difference between adjacent beats. It captures fast, beat-to-beat variation, which in turn reflects parasympathetic activity. This is the measure the training world uses.
• SDNN measures total spread across the whole recording. It comes out of clinical research on 24-hour ECG recordings, where it captures both fast fluctuations and the daily rhythm.

And here is the practical sting: Apple Watch reports SDNN. Oura, Whoop, Fitbit and Samsung report RMSSD.

So 40 on an Apple Watch is not the same thing as 40 on an Oura ring. They are different quantities that happen to share a label in the interface. Your friend's higher number may come down entirely to which device he happens to be wearing.

Practical consequence: change device and you start from zero. The old baseline does not transfer. Annoying, but not the end of the world, since a new baseline takes a couple of weeks to build.

The spread between healthy people is enormous

Published reference ranges for healthy adults are so wide that in practice they say nothing about any one individual. We are not talking about a few milliseconds either way. Two entirely healthy people of the same age can differ several-fold.

That is why "is 35 a good score?" cannot be answered. It is like asking whether shoe size 42 is a good number. Compared to what?

Age pulls the number down

HRV declines with age. In a classic survey of 24-hour recordings spanning nine decades (Umetani and colleagues, Journal of the American College of Cardiology, 1998), the beat-to-beat measures fell fastest until roughly the age of 60 and then levelled off.

For you at 45 that means two things. Sitting lower than a 25-year-old is expected, not an alarm. And comparing yourself to your own number from ten years ago is as pointless as comparing yourself to your neighbour.

Making the number useful: the baseline

If absolute numbers cannot be compared, what is left? Your own change over time. HRV is a relative measure or it is no measure at all.

The method is simple but demands patience:

1. Measure the same way every day for at least two weeks. During that time you draw no conclusions whatsoever. You are only collecting data.
2. Look at a rolling average, not at single mornings. Most apps show a seven-day average against a longer one. That comparison is the one that matters.
3. Treat isolated dips as noise. Single readings jump for reasons that have nothing to do with your condition: you slept with your arm at an odd angle, the strap was loose, a noise woke you just before the reading.
4. Look for patterns that persist. Three days running clearly below your average is a signal. One day is not.

What actually moves your HRV

After a few weeks of measuring you start to see patterns. These are the most common ones.

Alcohol deserves its own note, because it is the clearest lever most people have. Controlled trials show a dose-related drop in HRV the same evening. That does not make one glass of wine a catastrophe. But if you have been wondering why the Tuesday after a dinner always looks red, you probably have your answer.

The other big factor is sleep. What actually changes about sleep after 40 is covered in the guide to sleeping better after 40. If you also wake regularly in the middle of the night, the causes are gathered in the page on why you wake at 3am.

The decision rule: what to do on a morning below baseline

This is where the real value sits. A number that changes nothing you do is entertainment.

Research on so-called HRV-guided training has examined exactly this question. A systematic review with meta-analysis in the Journal of Science and Medicine in Sport (2021) pooled studies in which endurance athletes, both runners and cyclists, let their HRV trend decide that day's session. The effect on submaximal physiological measures was positive, in the small-to-medium range. What is interesting is how it came about: most HRV-guided groups did fewer moderate- and high-intensity sessions, not more.

Read that again, because it is the opposite of how most people use their watch. The gain did not come from training more. It came from the hard sessions landing on the right days.

Here is how that translates into a life with three or four sessions a week:

• Within baseline: train as planned. The number should not make you hesitate on a normal day.
• One day below: train anyway, but easy. Swap the hard session for one in zone 2. You lose nothing by doing so.
• Three days below in a row: drop the week's hard session entirely. "Hard session" here means intervals, hills or a tempo run, the kind of work described in our guide to running intervals. Prioritise sleep and food. Keep the easy sessions, because they cost little and keep the routine alive.
• Below baseline plus a sore throat, fever or chills: do not train. No number is needed here.

Notice what is missing from that list: "rest completely". It is rarely the right answer. Changing intensity beats cancelling almost every time, because one cancelled day easily becomes a cancelled week. How to hold a routine together when the days fall apart is covered in the guide to training when you have no time.

HRV or resting heart rate: which should you trust?

Almost everyone measuring HRV is also measuring resting heart rate. The two numbers are related but behave differently, and that difference is worth understanding.

Resting heart rate is the sluggish one. It moves slowly across weeks as fitness changes. It is also less sensitive to how you lay in bed or how the strap sat. That makes it easier to trust day to day.

HRV is more sensitive but noisier. It reacts faster to last night. The price of that sensitivity is that more of its swings are measurement error rather than reality.

The practical stance is to let them vote together:

• HRV down, resting heart rate up: the clearest signal there is. Your body is working on something. Worth easing off the intensity.
• HRV down, resting heart rate unchanged: probably noise or one poor night. Train as planned.
• Both steady for weeks while training increases: you are absorbing the load. Carry on.
• Resting heart rate creeping up over two weeks: take it seriously even if HRV looks normal. Its sluggishness makes that signal more meaningful, not less.

If you also want to understand why heart rate drifts upward during a session, the explanation is in the guide to zone 2, in the section on cardiovascular drift.

What HRV cannot tell you

An honest chapter, because this is where people go wrong.

HRV cannot separate causes. A low morning after a genuinely good hard session looks exactly like a low morning after a night of worrying about work. The screen shows load, not where it came from. Only you hold that knowledge.

HRV is not a health grade. It is a snapshot of autonomic balance, sensitive to sleeping position, temperature, breathing and how your strap sat. Reading it as a verdict on you as a person gives a millisecond figure a job it cannot do.

HRV is not a diagnostic tool. Low values show up across a long list of conditions, but the route from number to diagnosis runs through a clinician, never through an app.

And the number itself can become the problem. Sleep research has a name for measurement making the thing being measured worse. If the morning figure makes you anxious, if knowing you are being tracked costs you sleep, or if you start avoiding training you enjoy, the tool has stopped serving you.

If the number sits low week after week and your training does not explain it, the thing to adjust is not the tracker. What to do about the underlying load is covered in our guide to stress management.

How to measure so the number means something

HRV is sensitive to measurement conditions. The difference between measuring lying down on waking and sitting up after coffee can be larger than the difference between a good week and a bad one. So standardise the following:

• Same time. Either let a watch or ring measure automatically during sleep, or measure manually the moment you wake. Do not mix the two.
• Same body position. Lying down all week or sitting all week. Posture affects the value directly.
• Before coffee, phone and news. All three move the number before you have had a chance to read it.
• Breathe normally. Deep breathing raises HRV acutely. Pleasant in the moment, but it wrecks comparability if you only do it some days.
• One device at a time. See the section on RMSSD and SDNN above.
• Note the context. Late wine, late dinner, a bad day at work. Without the note, the number becomes a riddle in hindsight.

One last piece of advice: look at the number in the evening, not in the morning. Read it before you have got out of bed and it gets to decide how the day will feel. Read it once the day has already been lived and it becomes what it should be: an entry in a log.

When to seek medical care

• Contact a doctor promptly: heartbeats that feel irregular, that skip, or that race without exertion. HRV does not measure rhythm disorders and must not be used to rule them out.
• Contact your doctor: if you wake up tired and stay tired for most of the day, and a few days of rest do not help. Sweden's 1177 health service counts sleep problems as long-term once they have lasted more than four weeks and recur at least every other night. For symptoms you believe are stress related, the advice is to seek help early, because recovery is usually quicker then.
• Call emergency services: for chest pain that does not ease within a few minutes of rest, particularly alongside breathlessness, cold sweats or feeling faint.
• Check with your doctor before training by the numbers: if you have known heart disease or take medication affecting heart rate. Beta blockers affect both heart rate and its variability.

This is general health information, not a diagnosis or personal medical advice. HRV is not a tool for self-diagnosing burnout. If you are unsure, speak to a clinician.

Frequently asked questions

What is a good HRV score?

There isn't one. Published ranges for healthy adults are so wide that they say nothing about any individual, and two equally healthy people of the same age can differ several times over. The only meaningful comparison is your own number against your own average from the past few weeks.

Why does my Apple Watch show a completely different HRV than an Oura ring?

Because they calculate it differently. Apple Watch reports SDNN, which measures total variation. Oura, Whoop, Fitbit and Samsung report RMSSD, which measures the difference between adjacent beats. The same number on two such devices does not mean the same thing. Stick to one device and compare only against yourself.

Can I raise my HRV?

Partly. Regular easy cardio, better sleep and less alcohol tend to move the number up over weeks to months. But age pulls the other way, and a large share of your level is simply individual. Treat a rise as a likely side effect of good habits rather than a target in itself.

Does low HRV mean I am stressed?

Not necessarily. Low HRV means your body is working on something. That could be psychological stress, but it could equally be yesterday's hard session, short sleep, alcohol, a late heavy meal or an infection on its way. The number shows the load, not the cause. Only you know the cause.

Should I skip training when my HRV is low?

Rarely skip it entirely. A single low morning is usually noise. Swap a planned hard session for an easy one rather than cancelling. If the number sits below your baseline three days running, it is worth dropping the week's hard session. With a fever or a sore throat you do not train at all, whatever the screen says.

How long do I need to measure before the number means anything?

At least two weeks of measurements under identical conditions before you draw a single conclusion. Before that you have no baseline to compare against, only isolated numbers. Expect a month before the trend is clear enough to make decisions on.