Buying guide · Benchmark / measure / buy
Every wellness blog runs the same age-by-sex HRV chart, usually without saying which metric it used or how it was measured. The useful question is not whether your number beats a stranger’s chart. It is whether your ring, your watch and your friend’s strap are even measuring the same thing, because two of the three biggest brands publish their own average and it is not the same number.
A resting RMSSD in the high-40s to low-60s ms is typical in your 20s, falling to the high-teens to low-20s ms by your 70s in a large clinical reference population, but Oura’s own 2024 member average is 41 ms and WHOOP’s own member average is 62-65 ms for the same-named metric, and Apple Watch reports a different statistic (SDNN) entirely, so a “good” HRV depends on which device produced the number before it depends on your age.
Track the trend on whichever device you already wear, and stop there. Do not compare your number to a friend’s different brand, to Apple’s SDNN figure, or to a generic internet chart that never says which metric or measurement window it used.
Reports RMSSD averaged overnight. Oura’s own 2024 member data puts the average member at 41 ms, well below WHOOP’s published average for the same-named metric, a gap driven by measurement window and algorithm, not by whose members are healthier. Judge your own trend against your own history, not this number.
Also reports RMSSD, but calculated specifically during your last slow-wave sleep period rather than averaged across the night, which is part of why WHOOP’s own published member average (62-65 ms) runs roughly 50% above Oura’s. A different sampling window, not a different nervous system.
Garmin never shows a raw millisecond figure. It compares your 7-day average to your own 3-week personal baseline and reports only Balanced, Unbalanced, Low or Poor, with “Poor” defined against age-normed data internally. Arguably the most honest interface of the three, precisely because it refuses to hand you a number to compare against a stranger’s. Also Agewell’s existing VO2 max pick.
A chest strap reading electrical R-R intervals is the closest home replica of the controlled resting-ECG protocol behind the clinical reference tables in this guide. Take a 2-3 minute morning reading before getting out of bed, same time and position daily, and you get a number that is actually comparable week to week, unlike an all-night average that shifts with sleep architecture.
The cleanest reference table comes from a 2018 study of 13,943 resting 10-second ECGs pulled from five Dutch population studies, restricted to cardiologically healthy adults (no heart attack, heart failure, pacemaker, high blood pressure or diabetes). Using RMSSD, the metric Oura and WHOOP both report, the median for men runs about 52 ms in the 20s, 38 in the 30s, 30 in the 40s, 24 in the 50s, 21 in the 60s and 19 in the 70s. For women the medians run higher at every age: about 64, 48, 36, 27, 23 and 20 ms across the same decades.
The footnote that almost every fitness blog skips: this table comes from a single 10-second resting ECG in a clinic, not an overnight average from a ring or strap. Two other commonly cited numbers, a 2010 systematic review’s “normal RMSSD is 19-75 ms” and various brand charts, are not age-stratified at all, or are built from a single company’s own users. Read this table to find your rough band by age and sex. Do not expect your ring to land on it.
Here is the comparison none of the top search results run. Oura published its own 2024 member data in February 2025: the average member’s HRV is 41 ms (40.3 ms for men, 41.5 ms for women). WHOOP publishes its own member averages too, on its own site: 65 ms for men, 62 ms for women. Both companies call this number the same thing, RMSSD in milliseconds, measured overnight, in a broadly similar population of health-conscious adults who bought a $300-400 wearable.
The roughly 50% gap is not a health difference between Oura and WHOOP users. It comes from when and how each algorithm samples the night: WHOOP calculates RMSSD specifically during your last bout of slow-wave (deep) sleep, when parasympathetic activity runs highest, while Oura averages across a broader nightly window. Different sampling windows produce different numbers from the same nervous system, and neither company’s own page tells you that before showing you the figure.
Apple Watch does not report RMSSD at all. Apple’s own developer documentation states that Apple Watch computes SDNN, the standard deviation of normal beat-to-beat intervals, a related but mathematically distinct statistic that emphasizes different frequency components of your heart-rhythm signal than RMSSD does. A 35 ms reading on an Apple Watch and a 35 ms reading on an Oura Ring are not the same fact about your body; they are two different equations applied to two different recording windows.
Garmin sidesteps the comparison problem by design. Its HRV Status feature never shows you a raw number: it tracks your own 7-day rolling average against your own 3-week personal baseline and reports one of four categories, Balanced, Unbalanced, Low or Poor, with “Poor” specifically defined internally as running well below the normal range for your age. It is the one major platform that makes cross-user comparison structurally impossible, which, given the other two, is arguably the safer default.
The reason this metric earns a place in a baseline stack, and not just a curiosity stack, is a 2000 analysis from the Atherosclerosis Risk in Communities (ARIC) study: 14,672 healthy adults aged 45-65 with no coronary disease at baseline, followed for several years. People in the lowest third for SDNN (below about 30 ms on a 2-minute resting strip) had roughly double the all-cause mortality risk of people in the middle third.
Two honest caveats sit next to that number. First, it used SDNN on a clinical resting strip, not an overnight RMSSD average from a ring, so it is directional evidence that low HRV matters, not a cutoff you can apply to your Oura app. Second, this is an association in a healthy-adult cohort, not proof that a lower number causes earlier death or that raising your wearable’s number extends your life; HRV likely reflects broader autonomic and cardiovascular health rather than driving it directly.
Ignore the absolute figure and watch the direction. A meaningful drop from your own 60-90 day average, sustained for several days, is a more useful signal than any chart position: it commonly shows up before you notice a cold, after a heavy alcohol night, after a training block that outran recovery, or during a stretch of poor sleep. A single bad night rarely means anything; a week-long slide is worth paying attention to.
If you want a number that is actually comparable across time rather than an all-night average dragged around by sleep stage mix, take a short seated or supine chest-strap reading first thing in the morning, before coffee, same days each week. That is closer to the protocol behind the clinical reference table above, and it is the only way to get a figure you can honestly compare to your own history, let alone a study.
Skip comparing your HRV to a partner, friend or online chart across different brands; the gap between Oura’s and WHOOP’s own published averages shows why that comparison is close to meaningless on its own. Skip chasing a specific millisecond target: HRV is highly individual, and the trend against your own baseline is the only version of this number worth acting on.
If you have a diagnosed arrhythmia, including atrial fibrillation, consumer HRV algorithms are validated on regular sinus rhythm and the number becomes unreliable, so do not use it to self-monitor a heart condition. And this is a wellness signal, not a diagnosis: the FDA treats HRV tracking on consumer wearables as a general-wellness feature specifically because it makes no diagnostic claim; a real symptom, chest pain, fainting or a racing irregular heartbeat, needs a clinician, not a trend line.
In a large clinical reference table (RMSSD, resting ECG), the median for men runs about 52 ms in the 20s down to about 19 ms in the 70s, and for women about 64 ms down to about 20 ms across the same decades. Treat this as a rough band: your wearable’s overnight average will read differently because it uses a different measurement window.
Almost certainly the devices, not your health. Oura’s own 2024 member average is 41 ms; WHOOP’s own published member average is 62-65 ms, for the same-named RMSSD metric, because WHOOP samples specifically during your last deep-sleep period while Oura averages across a broader window. The two numbers are not built to be compared.
No. Apple Watch reports SDNN; Oura and WHOOP report RMSSD. They are related but different statistics calculated from the same beat-to-beat data, so a number from one is not directly comparable to a number from another.
There is real evidence at the population level: in the ARIC study, 14,672 healthy 45-to-65-year-olds in the lowest third for SDNN on a resting strip had roughly double the all-cause mortality risk of the middle third over several years of follow-up. That is an association in a specific clinical measurement, not proof that your ring’s nightly number causes or predicts your personal risk.
A chest strap like the Polar H10 paired with a free app such as Elite HRV, read for two to three minutes each morning before getting up, most closely replicates the controlled resting protocol used in the reference studies. It is the closest thing to a comparable, study-grade number you can get without a clinic visit.
No. Buy or keep whichever device fits how you actually live, and judge it only against its own history. Switching brands to chase a higher number just resets your baseline against a new measurement method, it does not change your physiology.
By David Persson · Updated 2026-08-10 · Corrections: hi@agewell.guide
The Agewell Briefing
A sourced weekly read on longevity products, clinics and research, with a clear answer on what is worth trying.
One email a week. Unsubscribe with one click.