Why Is My HRV Always Low?

Last updated: August 20, 2026

Before you look for a cause, check the question. If you are comparing an overnight average from a wrist or ring against a "normal HRV by age" chart you found online, you have made three errors before physiology even enters the picture: there is no agreed normal range to be below, the charts that circulate are built on a completely different measurement, and your device is not measuring HRV in the first place. It measures pulse rate variability, which reads systematically lower. Once those are out of the way, there are real reasons a reading sits persistently low, and they are different from the reasons it drops for a night.

Want to see which factors actually move HRV, ranked by how good the evidence is? The HRV factor explorer sorts all of them by strength of evidence, then come back for what "low" actually means.

I wrote this because the HRV content on the internet, ours included, answers "why did my HRV drop this week" far better than it answers "my HRV has always been like this." Those are different questions and the second one usually has a different answer.

Is your number actually low? Four checks first

Run these before you conclude anything.

One: compare to yourself, not to a chart or a friend. The only comparison with meaning is your own rolling baseline, on the same device, worn the same way, over the same measurement window. HRV is not a leaderboard metric; between-person differences are dominated by things you did not choose.

Two: check which metric you are reading. This one catches a lot of people, because the numbers are not on the same scale.

Device	Metric reported	Comparable to the others?
Oura	rMSSD	Yes, with other rMSSD devices
Garmin	rMSSD	Yes
Fitbit	rMSSD	Yes
WHOOP	rMSSD	Yes
Apple Watch	SDNN	No

A "low" Apple Watch number and a "normal" Oura number are not in conflict. They are different calculations on different scales.

Three: check the recording window. Overnight averages, morning readiness readings and manual spot checks are three different measurements. Comparing across them tells you nothing.

Four: check how long you have been looking. A single night is close to meaningless for an individual. One small study found nocturnal HRV is far more stable across nights than sleep architecture is, with intraclass correlations of .91 to .96, but that was 15 people across two nights, so treat it as directional rather than a rule.

One device-specific wrinkle almost nobody writes about: Oura's clean nightly HRV average is produced partly by throwing data away. An independent study of 114 people against polysomnography with concurrent ECG, no Oura authors involved, found 30 to 35% of five-minute segments rejected at an 80% validity threshold and 55 to 67% at 95%. That is not a scandal, it is how you get a usable nightly figure, but the tidy number you see is a survivor of heavy filtering.

There is no normal HRV to be below

This is the finding that should change how you read your own number.

In 2025, Brozat, Böckelmann and Sammito published a systematic review in the Journal of Cardiovascular Development and Disease covering 58 studies from 1989 to 2022, with samples ranging from 20 to 84,772 people and ages from one day to 99 years. Their conclusion, in their own words: "There are no generally accepted HRV normal values (yet)." Some large studies give figures usable "for orientation purposes," but heterogeneity in method, duration, equipment and confounders meant, again verbatim, that "it was not possible to merge the results of the studies in terms of a meta-analysis."

That is not academic hedging. It is the answer. The field looked at everything published across three decades and declined to issue a reference range.

The numbers that do exist are also not interchangeable with each other. Shaffer and Ginsberg's 2017 overview in Frontiers in Public Health is the standard reference here, and its own abstract cautions that "24 h, short-term, and ultra-short-term normative values are not interchangeable." Nearly every "normal HRV by age" chart circulating online is built on five-minute seated ECG recordings. Your wearable reports an overnight average across six to eight hours. Those are different quantities.

We have deliberately not published an age-normal table on this page, because doing so would contradict everything above and there is no source for one that survives scrutiny.

Your wearable is not measuring HRV

It measures pulse rate variability. HRV comes from the electrical signal of the heart, read by ECG. PRV comes from an optical sensor watching blood volume change in the skin, and the two are not the same quantity.

Ben-David and colleagues published the clearest demonstration of this in Sports Medicine Open in 2026, comparing beat-to-beat PPG against ECG in 103 male Division I collegiate American football players across three seasons. Heart rate agreement was near perfect, with a bias of 0.24 to 0.44 bpm. Variability agreement was not.

Metric	PPG pulse rate variability	ECG heart rate variability
rMSSD	80.9 ms (SD 23.1)	103.9 ms (SD 22.0)
SDNN	141.3 ms (SD 41.7)	167.9 ms (SD 40.0)

The authors' closing line is unusually blunt for a journal: the PRV calculated with PPG "should not be called HRV as it confuses scientists and consumers."

Three caveats. The sample is young male college athletes whose ECG rMSSD averaged around 104 ms, well above typical, so the direction generalises but those millisecond values do not describe you. Six of the eleven authors are affiliated with a company whose product competes with PPG, worth knowing even though the finding runs against the wearable market rather than for it. And the underestimation is not uniform, so there is no correction factor you can apply; it even flips by site, with upper-arm sensors under-reading and finger or toe sensors over-reading under vasoconstriction. For the longer version of why optical sensors behave this way, see how accurate is your heart rate monitor and the sensor comparison tool.

Use your number as a trend for yourself. Do not use it as an absolute value, and do not compare it to an ECG-derived chart.

The causes you cannot change

These explain most persistent differences between people, which is exactly what "always low" describes. Telling a 55-year-old their HRV is low against a chart built on 25-year-olds is the most common false alarm in this whole topic.

Factor	Direction	Evidence	What it means for you
Age	Lowers HRV	Strong	The strongest single predictor. Steady decline from loss of pacemaker cells and vagal fibres. A fit older person still beats a sedentary younger one
Sex	Varies	Strong	Women generally show higher HF power; estrogen enhances vagal tone; the gap narrows after menopause
Genetics	Varies	Emerging	Twin studies estimate 40 to 50% heritability, though a genome-wide study at n=6,740 found no significant individual genes. Lifestyle still dominates
Menstrual cycle	Varies	Moderate	Shifts across phases, so compare like phase to like phase rather than week to week
Circadian timing	Varies	Strong	HRV peaks overnight and dips in the morning, which is why measurement timing matters more than most people assume

None of these are fixable, and none of them mean anything is wrong. They are the reason a number can be genuinely, permanently lower than someone else's and still be entirely normal for you.

The chronic causes you can change

This is the list that separates "always low" from "dropped last night." Each of these is a level you are living at, not an event that happened.

Factor	Direction	Evidence	The chronic framing
Cardiorespiratory fitness	Raises HRV	Strong	The largest modifiable driver by some distance
Body fat and BMI	Lowers HRV	Strong	Adipose tissue produces inflammatory cytokines that suppress HRV. Weight loss reverses it
Chronic stress	Lowers HRV	Strong	Sympathetic dominance held on rather than switched on. One of the top reasons for a persistently low reading
Chronic sleep debt	Lowers HRV	Strong	Sleep quality is the top predictor of nocturnal HRV, and consistency beats duration
Regular alcohol	Lowers HRV	Strong	Dose-dependent, roughly 2 ms at one drink up to 13 ms at three or more. Fitness does not protect you. A daily drinker has a permanently depressed floor, not a bad night
Smoking	Lowers HRV	Strong	Active and passive both. Nicotine acts on sympathetic ganglia, particulates drive inflammation
Overtraining	Lowers HRV	Strong	A downward trend despite training is the classic early overreaching signal

On fitness, the best current evidence is a 2024 network meta-analysis by Yang and colleagues in Reviews in Cardiovascular Medicine, pooling 29 randomised trials across 1,317 people. High-intensity interval training ranked first for SDNN, with a SUCRA of 98.7%, first for rMSSD at 84.9%, and first for LF/HF at 99.75%. Resistance training ranked first for HF power, and combined training first for LF power.

The caveat most pages skip: those are group-level effect sizes from supervised interventions. They tell you a population mean moves. They do not promise you a specific millisecond gain on a specific timeline.

Overtraining and alcohol are the two items here that also appear on the acute list, because they work on both timescales. If your question is really about a specific bad week rather than a standing level, the twelve acute triggers are in why did my HRV drop.

The chronic causes are also where a food and wearable log earns its keep, because a standing level is invisible night to night and only shows up across weeks. Kygo lines your HRV up against what you actually ate and drank and finds which patterns are yours rather than population averages. Free on iOS and Android.

When a persistently low reading is worth raising with a doctor

Reduced HRV is documented in several conditions. That is not the same as a low reading meaning you have one, and nothing below is a diagnosis. What it means is that a persistently low reading alongside symptoms is worth mentioning to a doctor, and a persistently low reading with no symptoms usually is not.

Condition	What the evidence shows
Type 2 diabetes	Benichou et al. 2018 pooled 25 case-control studies, 2,932 people. HRV reduced across every metric: standardised effect sizes of -0.65 for SDNN, -0.92 for rMSSD, -0.79 for HF, -1.08 for LF, all p below 0.001. These are effect sizes, not millisecond differences, and they do not convert
Obstructive sleep apnea	Wang et al. 2023 pooled 22 studies, 2,565 patients and 1,089 controls. Severity matters and the headline is usually written wrong: severe OSA showed lower HF, rMSSD and SDNN plus higher LF/HF, while moderate OSA differed from controls on LF/HF only
Cardiovascular disease	A measurement warning rather than a cause. A 2023 comparison of a Garmin vivoactive 4 against 1000 Hz ECG in 263 people, 104 after STEMI and 129 after stroke, found heart rate concordance of 0.9998 and SDANN 0.9617, but rMSSD only 0.6617. In the population where HRV matters most clinically, the consumer number is least trustworthy

On mental health, the popular claim runs well ahead of the evidence. The best current synthesis is a 2025 umbrella review by Zou and colleagues in Translational Psychiatry covering 21 systematic reviews, 53 meta-analyses, 19 mental disorders and 34,625 participants. Only 7 of the 53 patient-versus-control comparisons, 13.2%, reached even "suggestive" evidence, and none reached "highly suggestive" or "convincing." The suggestive four were dementia and neurocognitive disorders, PTSD, somatic symptom and functional somatic syndromes, and schizophrenia. The association is real and best supported there, and even there it grades as suggestive. Nothing in that literature supports reading a wearable number as a mental health indicator.

On medication, several drug classes do shift HRV and this is a genuine reason a reading sits persistently low. We are not naming classes or directions here because we have not verified that literature to primary sources, and getting it half right would be worse than not saying it. If you take a regular prescription and your HRV is persistently low, that is a question for your prescriber.

What actually raises it, and how long it takes

Briefly, because this belongs elsewhere.

The highest-evidence levers are cardiorespiratory training, with HIIT ranking first on three of the five HRV metrics in the Yang analysis, slow breathing at around six breaths per minute, sleep consistency, and removing alcohol. Slow breathing is the fastest reliable lever, with SDNN improving after four weeks in a randomised trial. Everything else is slower: the exercise trials that moved HRV ran eight weeks or more.

Nothing here moves in a week, which is worth internalising before you start checking daily. The full ranked list, 44 factors deep with the evidence grade on each, is in how to improve HRV.

Common questions

Why is my HRV always low?

Usually one of three things: you are comparing against a number that does not apply to you, your device under-reads by design, or there is a standing lifestyle or medical reason rather than a one-off trigger. Work through them in that order.

What is a normal HRV?

There is no agreed answer. A 2025 systematic review of 58 studies concluded there are no generally accepted HRV normal values, and that the published figures could not even be pooled into a meta-analysis.

Is my HRV low for my age?

Age is the strongest single predictor of HRV and it declines steadily. But the age charts circulating online are mostly built on short seated ECG recordings, which are not comparable to an overnight wearable average.

Why is my HRV lower than my friend's on the same device?

Between-person differences are dominated by age, sex, fitness, body composition and genetics. HRV is not a leaderboard metric, and a gap between two people says very little.

Does a low HRV mean something is wrong with me?

Not on its own. Persistently reduced HRV is documented in type 2 diabetes, severe obstructive sleep apnea and cardiovascular disease, so a persistently low reading alongside symptoms is worth raising with a doctor. A low reading with no symptoms usually is not.

Does low HRV mean I am depressed or stressed?

The evidence is weaker than the internet suggests. An umbrella review across 19 mental disorders found only 13.2% of patient-versus-control comparisons reached even "suggestive" evidence, and none reached "convincing."

Why does my Apple Watch show a much lower HRV than my Oura?

Apple reports SDNN and most other devices report rMSSD. Different calculations, different scales, not comparable. The measurement window differs too.

How long does it take to raise HRV?

The exercise trials that moved it ran eight weeks or more. Slow breathing is the fastest reliable lever and showed SDNN improvement after four weeks.

Is my wearable's HRV accurate?

It is measuring pulse rate variability from an optical sensor, not true HRV from an ECG, and it reads systematically lower. The gap is not fixed, so there is no correction you can apply. Use it as a trend for yourself, not as an absolute value.

The bottom line

Most people asking why their HRV is always low are comparing an overnight optical average against a five-minute ECG chart, from a field that has publicly declined to publish norms. Fix the comparison first and the question often dissolves.

If it survives that, the honest answer is that persistent differences between people are mostly age, sex, genetics and fitness, and that the levers worth pulling are the slow ones: aerobic training, sleep consistency, less alcohol, less standing stress. If the reading is persistently low and you also have symptoms, that is a conversation with a doctor rather than a supplement.

What your number is genuinely good for is watching yourself over time. Kygo connects your HRV to what you actually ate and drank so the standing patterns become visible across weeks instead of guesswork. Free on iOS and Android.

---

Key sources: Brozat, Böckelmann and Sammito 2025 (J Cardiovasc Dev Dis 12(6):214, 58 studies) for the absence of accepted normal values; Shaffer and Ginsberg 2017 (Front Public Health 5:258) for non-interchangeable recording lengths; Ben-David et al. 2026 (Sports Med Open 12(1):68, n=103) for pulse rate variability versus ECG HRV, noting six of eleven authors are affiliated with a company competing with PPG; an independent Oura Gen3 validation in Sensors 2024 (24(23):7475, n=114) for the segment rejection rates; Yang et al. 2024 (Rev Cardiovasc Med 25(1):9, 29 trials, n=1,317) for the exercise ranking; Benichou et al. 2018 (PLOS ONE 13(4):e0195166) for type 2 diabetes; Wang et al. 2023 (J Sleep Res 32(1):e13708) for sleep apnea by severity; Zou et al. 2025 (Transl Psychiatry 15(1):104) for the mental health umbrella review; and a 2023 Garmin versus ECG comparison in Eur Heart J Digit Health 4(3):155.

---

A note on what we left out. We did not publish a normative HRV table by age, because the evidence in this post says no defensible one exists. We did not convert the diabetes effect sizes into milliseconds, because standardised effect sizes do not translate. We did not write that sleep apnea lowers HRV without the severity qualifier, because moderate OSA differed from controls on one metric only. We did not name medications or their direction of effect, because that literature was not verified for this piece. And we did not treat the Ben-David millisecond values as general population figures, because the sample was college American football players with unusually high baseline HRV.

Written for general information, not medical advice. HRV is not a diagnostic test, and nothing here should be used to rule a condition in or out. If a persistently low reading comes with symptoms, talk to a doctor.

If you have tracked HRV for a year or more, I would like to know whether your baseline actually moved, and what you changed, because the honest answer for most people seems to be that it moves slowly or not at all.