HRVHeart Rate Variability
The variation in time between heartbeats, used as a readiness and recovery signal.
Heart rate variability is the variation in time between consecutive heartbeats. A heart beating 60 times a minute is not beating precisely once a second: the gaps differ by milliseconds, and that irregularity is normal and healthy. HRV measures how much they differ.
The variation exists because two branches of the autonomic nervous system are constantly adjusting the interval. The parasympathetic branch, acting through the vagus nerve, slows the heart and can do so beat to beat; the sympathetic branch speeds it up more slowly. When you are rested and unstressed, vagal influence dominates and the intervals fluctuate freely. Under stress, illness, hard training or poor sleep, sympathetic drive rises, vagal influence is withdrawn, and the rhythm becomes more metronomic. HRV is therefore an indirect window onto autonomic balance, which is what makes it interesting, and also what limits it.
There are several ways to quantify it and they are not interchangeable. RMSSD, the root mean square of successive differences, is what most wearables report: it responds mainly to vagal activity and can be derived from short recordings, which is why it dominates consumer devices. SDNN captures overall variability but needs longer recordings, ideally 24 hours, to mean what it is supposed to mean. Frequency-domain measures such as high- and low-frequency power split the signal differently again. Comparing an RMSSD from one app to an SDNN from another is comparing different quantities, and a good deal of online confusion comes from exactly that.
The single most important thing to understand is that absolute values are close to meaningless between people. Healthy adults span an enormous range, one person's entirely normal resting RMSSD can be several times another's, and the figure declines with age. There is no target, no good score, and no useful comparison with a friend or a population average. HRV only works as a signal measured against your own baseline, under consistent conditions, over time.
That consistency is not optional, because HRV is easily perturbed by measurement conditions. Body position, breathing rate, time of day, hydration, a recent meal, caffeine and alcohol all shift it: alcohol in particular produces a large, reliable overnight suppression that is often someone's first convincing demonstration that the metric is measuring something real. So measure at the same time, in the same position, ideally overnight or immediately on waking. Chest straps reading electrical activity are more accurate than wrist optical sensors, which infer beat timing from blood flow and lose precision when you move.
Read it as a trend, not a verdict. A single low morning means very little; a rolling average drifting below your own baseline over a week or more is worth attention. Large deviations are the genuinely useful signal: a sharp drop often precedes noticeable illness by a day or two, and reliably follows a heavy training block or a bad night. Small daily wobbles are noise and acting on them is a good way to make training decisions worse rather than better.
It is not a fitness test and higher is not automatically better. HRV correlates loosely with aerobic fitness at population level, but VO₂max and resting heart rate tell you far more about capability. An unusually high reading can also accompany deep fatigue in heavily trained athletes, where excessive parasympathetic dominance is a sign of accumulated strain rather than freshness, which is one more reason to treat the number as one input into how you feel and train, and not as a score to chase.
Sources
Frequently asked questions
- What is a good HRV for my age?
- There is no meaningful answer to this, and it is the most commonly asked question about HRV. Healthy people of the same age differ several-fold in their normal values, and the metrics reported by different devices are not the same quantity. HRV declines with age at population level, but that tells you nothing useful about whether your own figure is good. The only comparison worth making is against your own baseline, measured consistently over weeks.
- Why is my HRV lower than my friend's?
- Almost certainly because you are different people. Absolute HRV varies enormously between healthy individuals and is influenced by genetics, age, body position, breathing, measurement method and the device's algorithm. A lower number does not mean worse health or worse fitness. Comparing HRV between people is not a meaningful exercise.
- What makes HRV drop?
- Poor or short sleep, alcohol, a hard training session or accumulated training load, illness coming on, psychological stress, dehydration and a late heavy meal all lower it. Alcohol is the most striking and consistent of these — the overnight suppression after a few drinks is large enough that most people notice it immediately. A drop of a day or two after hard training is expected and is not a problem in itself.
- Should I skip training if my HRV is low?
- Not on one reading. Day-to-day HRV is noisy enough that acting on a single low morning usually makes training decisions worse. A sustained dip below your own rolling baseline, especially alongside feeling run down, poor sleep or unusual soreness, is a reasonable prompt to take an easier day. Treat it as one input among several, with how you actually feel carrying at least as much weight.
- Is higher HRV always better?
- No. Within your own baseline a higher reading generally reflects better recovery, but it is not a linear scale of health, and unusually high values can accompany deep fatigue in heavily trained athletes, where they indicate accumulated strain rather than freshness. HRV is also not a fitness measure — VO₂max and resting heart rate say much more about capability.
- What is the difference between RMSSD and SDNN?
- RMSSD measures beat-to-beat differences and mainly reflects vagal, parasympathetic activity; it can be derived from short recordings, which is why most wearables use it. SDNN captures overall variability including slower influences, and is properly interpreted only from long recordings, ideally 24 hours. They are different quantities and their values should not be compared with each other.