Heart Rate Variability: Why Your Number Means Less Than You Think

It’s 6:40 a.m. and the ring on your finger says 41. Last Tuesday it said 68. Nothing obvious happened in between, and now you’re standing in the kitchen doing arithmetic on your own nervous system before you’ve had a glass of water.

Heart rate variability is the most misread number in consumer health tracking, and the measurement itself is fine. The problem is the question people ask of it. They want to know whether their number is good. That question has almost no useful answer.

Think about how an engine idles. One sits at 600 rpm, another at 900, and both left the factory that way. Neither number makes one engine better. But if yours has held 700 for a year and this morning it’s sitting at 850, a mechanic gets interested. He never asks what the car parked next to you idles at.

Your reading works the same way.

What Heart Rate Variability Actually Measures

Your heart doesn’t tick like a metronome. At 60 beats per minute, the gaps between beats aren’t a clean one second each. One might be 1,010 milliseconds, the next 960, the next 1,030. Heart rate variability is the size of that wobble.

The wobble comes mostly from your vagus nerve. Breathe in and vagal outflow drops, so the heart speeds slightly. Breathe out and it returns, slowing you down again. A responsive parasympathetic system produces bigger gaps between those two states. A system stuck in sympathetic drive flattens them out.

You’ll see several abbreviations for this. RMSSD tracks beat-to-beat differences and maps closely onto vagal activity, which is why nearly every wearable reports it: the 1996 European Society of Cardiology task force preferred it to the alternatives on statistical grounds, and it stays reliable over recordings as short as a minute. SDNN captures total variability across a full 24 hours and means something different. The LF/HF ratio still shows up in app dashboards despite researchers spending two decades arguing that nobody can say cleanly what it represents.

If your device gives you one number, it’s almost certainly RMSSD in milliseconds.

Why Your Number Is Mostly Inherited

In 2021, Marco Altini and Daniel Plews published the largest free-living analysis of this measurement anyone has run. Twenty-eight thousand people. Just over nine million morning readings, collected across five years, one minute at a time, on waking.

They then asked how much of the difference between people could be explained by the obvious things. Age, sex, body mass index, and how often someone trains, all together, accounted for fifteen percent of the variation in HRV.

Fifteen. The other eighty-five percent sits somewhere else, and a large share of it appears to be genetic. Twin and family studies, including work out of the Framingham cohort, put heritability of these metrics as high as sixty percent.

Break the fifteen percent down and it gets stranger. Age was the strongest single factor, and even that was only a moderate correlation. Training frequency was weakly related, and the relationship faded with age: reasonably strong in men in their twenties, noticeably weaker by the fifties. Sex made no difference at all, which surprises people who assume it works like resting heart rate.

You can train seriously for a decade and still sit below a sedentary twenty-five-year-old on the same chart.

The authors drew the obvious conclusion, and it reads nothing like what an app will tell you: given how much of this is heritable and age-driven, treating a higher HRV as a goal to chase may not be realistic. What the measurement does well is catch day-to-day changes inside one person.

Two things about that study you should weigh. One of the authors built the app the data came from, and he says so in the paper. And everyone in the sample chose to download a tracking app aimed at athletes, so this is not a random slice of the population. Neither fact undoes the finding, but both belong in your head when you read it.

This puts HRV in a different category from something like VO2 max, where training genuinely moves the number, and closer to grip strength, where the marker predicts a great deal while the question of whether improving it changes anything stays open.

What Those “Normal HRV by Age” Charts Leave Out

Search for a reference range and you’ll find dozens of tidy tables. Ages down one side, healthy values across the top, and an implicit invitation to find your row.

Most of them trace back to a 2010 systematic review by David Nunan and colleagues, which pooled 44 studies and 21,438 healthy adults. That review is real and the numbers in it are real. Its actual conclusion was that normative data for short-term HRV do not exist. The authors were compiling what had been published in order to show how far apart the values were, not to hand anyone a lookup table.

How far apart? For the frequency-domain measures, they found interindividual variation running to 260,000 percent among people all described as normally healthy.

There’s a second problem underneath the first. Fred Shaffer and Jay Ginsberg’s review of HRV norms makes the point that 24-hour values, five-minute values, and sub-five-minute values are not interchangeable, because they measure different things over different windows. A chart built from clinical 24-hour recordings tells you very little about the sixty-second reading your ring took while you were half asleep.

So when your app shows you a percentile, treat it roughly the way you’d treat a stranger’s opinion of your handwriting. Mildly interesting. Not actionable.

What Actually Moves It, and By How Much

The same nine-million-measurement dataset answers this one properly, because it tracked what happened to individual people after specific events rather than comparing strangers to each other.

What happened the day before	Change in HRV	Change in resting heart rate
High alcohol intake	12% lower	6% higher
Being sick	10% lower	6% higher
Hard training session	4.8% lower	1% higher
Full rest day	1.6% higher	0.6% lower

Read the top row against the third one. A heavy night hit these people’s readings roughly two and a half times harder than a hard training session did. If you’ve ever logged an easy week, slept fine, and still watched your numbers sink, check what you drank rather than what you lifted. The same pattern shows up in how alcohol wrecks sleep quality while leaving sleep duration untouched.

The resting heart rate column stays nearly flat by comparison. Across every stressor, HRV moved three to four times more than heart rate did. That sensitivity is the whole argument for tracking it.

Sensitive, But Not Specific

Sensitivity comes with a cost, and the researchers named it directly: HRV is a sensitive marker of stress without being a specific one.

Your reading dropped nine points. That could be Tuesday’s intervals. It could be the two beers. It could be a cold arriving in about thirty-six hours, an argument you haven’t finished having, a flight, a late meal, a warm bedroom, or the fact that you woke up twenty minutes earlier than usual and measured yourself standing rather than lying down. Your nervous system responds to all of it and reports back with one number that can’t tell you which.

Back to the engine. An idle sitting 150 rpm high tells the mechanic something is off. It doesn’t tell him whether it’s a vacuum leak, a dirty sensor, or a cold morning. He uses it to decide where to look next.

How to Measure It So the Number Means Something

Because you’re comparing yourself to yourself, consistency in how you measure matters more than which device you own.

1. Same time, same position. On waking, still lying down, before you stand up or reach for coffee. Standing up changes the reading substantially, so a sitting measurement and a supine one are two different data series.
2. Don’t pace your breathing. Slow deliberate breathing inflates the number. If you do a breathing drill before measuring on some mornings and not others, you’ve added noise you’ll later mistake for recovery.
3. Give it a week before it means anything. A single morning is mostly noise. A seven-day rolling average is the standard approach for a reason.
4. Know your error bars. The typical measurement error for HRV runs around 12 percent, and the smallest change considered practically meaningful is roughly 3 percent. A reading that bounces around inside that range is telling you nothing.
5. Don’t switch devices mid-baseline. Different sensors, different measurement windows, different algorithms. Your history resets when your hardware does.

Thirty days of consistent measurement gives you something worth reading. Thirty days of measuring whenever you remember gives you a graph.

Does Training by HRV Actually Work?

The pitch is appealing. Wake up, check your reading, go hard if it’s up and go easy if it’s down. Several apps are built entirely around this idea.

Two meta-analyses have tested it against ordinary preplanned training, and the results are more modest than the marketing.

Peter Düking’s group pooled eight studies covering 198 participants. HRV-guided training produced a moderate benefit on submaximal physiological markers. On actual performance and on peak oxygen uptake, the effects were small and not statistically significant. A second review by Agustín Manresa-Rocamora and colleagues, working through the methodological differences between studies, found the same shape: better vagal HRV indices, no significant edge in aerobic capacity or endurance performance. Their own summary was that if the approach beats preplanned training, it does so by a small margin.

One finding held up across both, and it’s the practical one. Fewer people responded badly. HRV-guided groups had fewer non-responders, largely because the method quietly removes some hard sessions from weeks when the athlete was already cooked.

That’s a real benefit and a limited one. If you already respect easy days and keep most of your volume aerobic, as in a standard polarized split, HRV monitoring will mostly confirm what you’d have done anyway. If you tend to hammer every session regardless of how you feel, it gives you a number to argue with. Evidence quality here is moderate at best: eight studies and fewer than two hundred people is a thin base.

When a Low Reading Is a Medical Question

Everything above assumes a healthy man reading a consumer device. Some patterns belong with a doctor instead.

Speak to a physician if your baseline has fallen steadily over weeks or months without an explanation you can point to, or if a low reading arrives alongside symptoms: chest discomfort, breathlessness that’s new, fainting or near-fainting, palpitations, persistent unexplained fatigue, or a resting heart rate that has climbed and stayed climbed. Reduced HRV is associated with several conditions that need proper assessment, including sleep apnea, thyroid disorders, diabetes, and cardiac problems. Your ring cannot distinguish between them and neither can an article.

You’ll notice this piece hasn’t given you a target number, and that’s deliberate. Since no normative range exists for short-term HRV, any figure quoted as “good” was invented by whoever quoted it. The same reasoning applies to why an article about ApoB should describe what the guidelines say rather than hand you a threshold to hit. Numbers that sound clinical belong to clinicians.

What the Number Is Actually For

Stop asking whether your heart rate variability is good. The comparison you keep reaching for, you against a chart, is the one the underlying research says cannot be made.

The comparison that works is narrower and more useful: you this week against you last month. Measured the same way, at the same hour, in the same position, averaged across seven days. That series will tell you when alcohol is costing you more than you assumed, when a training block has stopped being productive, and occasionally when you’re getting sick before you feel it.

Two engines idle at 700 and 900 and both are running fine. What matters is that yours has held 700 since spring. The morning it reads 850, you have something worth investigating, and it was never the other car’s business.

References

1. Altini M, Plews D. What Is behind Changes in Resting Heart Rate and Heart Rate Variability? A Large-Scale Analysis of Longitudinal Measurements Acquired in Free-Living. Sensors. 2021;21(23):7932. https://doi.org/10.3390/s21237932
2. Nunan D, Sandercock GRH, Brodie DA. A Quantitative Systematic Review of Normal Values for Short-Term Heart Rate Variability in Healthy Adults. Pacing and Clinical Electrophysiology. 2010;33(11):1407-1417. https://doi.org/10.1111/j.1540-8159.2010.02841.x
3. Shaffer F, Ginsberg JP. An Overview of Heart Rate Variability Metrics and Norms. Frontiers in Public Health. 2017;5:258. https://doi.org/10.3389/fpubh.2017.00258
4. Laborde S, Mosley E, Thayer JF. Heart Rate Variability and Cardiac Vagal Tone in Psychophysiological Research. Frontiers in Psychology. 2017;8:213. https://doi.org/10.3389/fpsyg.2017.00213
5. Düking P, Zinner C, Trabelsi K, et al. Monitoring and adapting endurance training on the basis of heart rate variability monitored by wearable technologies: A systematic review with meta-analysis. Journal of Science and Medicine in Sport. 2021;24(11):1180-1192. https://doi.org/10.1016/j.jsams.2021.04.012
6. Manresa-Rocamora A, Sarabia JM, Sánchez-Meca J, et al. Heart Rate Variability-Guided Training for Enhancing Cardiac-Vagal Modulation, Aerobic Fitness, and Endurance Performance. International Journal of Environmental Research and Public Health. 2021;18(19):10299. https://doi.org/10.3390/ijerph181910299
7. Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology. Heart Rate Variability: Standards of Measurement, Physiological Interpretation, and Clinical Use. Circulation. 1996;93(5):1043-1065. https://doi.org/10.1161/01.CIR.93.5.1043
8. Singh JP, Larson MG, O’Donnell CJ, et al. Heritability of Heart Rate Variability: The Framingham Heart Study. Circulation. 1999;99(17):2251-2254. https://doi.org/10.1161/01.CIR.99.17.2251

FAQ