Your watch or smart ring probably shows you a heart rate variability score every morning, and you have most likely watched it move without knowing what to do about it. HRV is one of the few numbers on a wearable with a serious research base behind it. It is also one of the most misread. Here is what the evidence says the number reflects, and what genuinely shifts it.
What heart rate variability actually measures
A resting heart rate of 60 beats per minute does not mean one beat every second. The gaps between consecutive beats vary constantly: 940 milliseconds, then 1,010, then 970. Heart rate variability (HRV) is the measurement of that beat-to-beat variation.
Those fluctuations are a window onto your autonomic nervous system, and specifically the balance between its two branches. The sympathetic branch accelerates your heart; the parasympathetic, or vagal, branch slows it. Greater variation generally indicates stronger vagal influence, meaning a heart that adjusts quickly to changing demands. Less variation suggests that vagal brake is working less effectively.
Two figures dominate the literature. RMSSD, the root mean square of successive differences between beats, is what most wearables report, because it tracks vagal activity closely and can be calculated from short recordings. SDNN, the standard deviation of normal-to-normal intervals, captures overall variability from both branches combined.
Why your HRV is worth paying attention to
The prognostic evidence is substantial. A systematic review and meta-analysis pooling 32 studies and 38,008 participants found that lower HRV predicted higher all-cause and cardiac mortality consistently across different ages, sexes, continents, populations and recording lengths. People in the lowest quartile of five-minute RMSSD carried a 56% higher risk of death than the remaining quartiles combined, with a hazard ratio of 1.56 (95% CI 1.32–1.85) (Jarczok et al., 2022).
That is an association, not a demonstration of cause. Nudging your morning score upward is not the same as extending your life. But HRV is tracking something real about autonomic health, which is why it deserves more than a passing glance.
What counts as a good HRV?
There is no universal healthy number. There is only your number, and its direction of travel.
HRV varies enormously between individuals, and it declines with age. In 260 healthy people aged 10 to 99, RMSSD fell to roughly 47% of second-decade values by the sixth decade, and pNN50 to around 24%, before broadly levelling off (Umetani et al., 1998). Two people of the same age and similar fitness can differ several-fold.
So comparing your RMSSD with a friend’s tells you close to nothing. Comparing this week’s average with your own three-month baseline tells you a great deal. Treat HRV as a trend, not a score.
Does exercise raise HRV?
Yes, and the vagally mediated measures respond most. A meta-analysis of 16 randomised controlled trials in 623 healthy adults found exercise training improved RMSSD (standardised mean difference 0.84, 95% CI 0.36–1.31), SDNN (0.58, 95% CI 0.16–1.00) and high-frequency power (0.89, 95% CI 0.27–1.51) compared with controls (Amekran and El Hangouche, 2024).
The effect does not disappear with age. A meta-analysis and meta-regression restricted to healthy people over 60 confirmed that endurance training still increases HRV past that threshold, and identified training frequency as the factor that determined the size of the gains (Raffin et al., 2019). How often you train mattered more than how long each session lasted.
Does intensity matter, or just consistency?
The comparative evidence is thinner, and worth reading with appropriate caution. In a small study of 13 physically inactive adults, eight sessions completed across two weeks, interval cycling (10 seconds hard, 50 seconds recovery, at or above 90% of peak heart rate) was compared with 40 minutes of continuous cycling at 60–75% of peak heart rate. Both improved inter-beat interval; only the interval group improved the LF/HF ratio, despite training for half the time (Alansare et al., 2018). With that sample size, treat the finding as suggestive rather than settled.
REHIT, or reduced-exertion high-intensity interval training, takes the time-efficiency principle further. CAROL’s signature REHIT ride lasts around five minutes and is built around two 20-second all-out sprints. Its documented physiological signature is a pronounced disturbance of homeostasis: sharp muscle glycogen depletion, phosphorylation of ACC and activation of PGC-1α, the pathways underpinning aerobic adaptation (Metcalfe et al., 2016).
We should be straightforward here: HRV has not been a primary outcome in REHIT trials, so we would not claim a direct effect on your morning reading. What the wider literature does support is that frequency is what accumulates autonomic benefit, and a five-minute session is considerably easier to repeat three times a week, for years, than a 45-minute one.
The two habits that move HRV fastest
Alcohol is the most reliable overnight suppressor. In a dose-response laboratory study, 26 healthy adults spent three nights under polysomnography with low, high or placebo doses before bed. Even the low dose, producing breath alcohol of around 0.02%, raised nocturnal heart rate, suppressed total and high-frequency (vagal) HRV and reduced baroreflex sensitivity, with effects more pronounced at around 0.05% (de Zambotti et al., 2021). If your score dropped this morning, last night’s two glasses are the likeliest explanation.
Sleep matters too, though the pooled picture is less tidy than you might expect. A meta-analysis of 11 randomised trials in 549 participants examined sleep deprivation and autonomic function; the time-domain reduction in SDNN did not reach statistical significance, and the authors attributed much of the inconsistency to differing measurement methods, deprivation durations and uncontrolled confounders (Zhang et al., 2025). Protecting your sleep remains sound advice; expecting a single short night to explain every dip is not.
How to measure HRV so the number means something
Wearable HRV is derived from optical pulse signals rather than an ECG trace, and accuracy depends heavily on how the data are filtered. Validating a popular smart ring against ECG in 114 participants across two in-lab sleep studies, researchers found accurate RMSSD required a stringent quality threshold of roughly 80% valid interbeat intervals within each five-minute segment, and averaging across windows of at least 30 minutes. Errors were larger in participants aged 45 and over, more than half of whom exceeded 10% median absolute percentage error (Liang et al., 2024).
In practice: take readings overnight or immediately on waking, under the same conditions and with the same device, and read seven-day rolling averages rather than reacting to any single morning.
Should you let HRV dictate your training?
Partly. A methodological systematic review with meta-analysis found HRV-guided training superior to predefined programmes for improving vagal-related HRV indices (SMD+ 0.50, 95% CI 0.09–0.91), but the advantage for maximal aerobic capacity was small and not statistically significant (SMD+ 0.20, 95% CI −0.07 to 0.47) (Manresa-Rocamora et al., 2021).
A second meta-analysis of eight studies and 198 participants reached a similar conclusion: a medium effect on submaximal physiological parameters (Hedges’ g 0.296, 95% CI 0.031–0.562), small and non-significant effects on performance and VO₂peak, but notably fewer non-responders. Most HRV-guided groups completed fewer moderate- and high-intensity sessions than the predefined groups, and still matched them (Düking et al., 2021).
Read that carefully. HRV is useful for deciding when to push and when to ease off, and it appears to reduce the odds of being a non-responder. It is not a reason to skip training whenever the number dips.
The bottom line
- HRV reflects vagal control of your heart, and lower values predict higher mortality risk in large pooled datasets — a 56% higher risk in the lowest RMSSD quartile.
- There is no good universal number. HRV declines with age and varies widely between people, so only your own trend is informative.
- Regular aerobic training raises vagally mediated HRV in healthy adults, and frequency matters more than session length — including after 60.
- Alcohol reliably suppresses vagal HRV overnight, even at low doses. Protect your sleep, and expect noise.
- Measure overnight or on waking, same conditions each time, and read seven-day averages rather than single mornings.
- Use HRV to modulate intensity across a week, not to justify doing nothing.
If you want the number to move, the most reliable levers are unglamorous: train often, drink less, sleep properly, and give it three months.
