HRV Explained: What Your Wearable Is Actually Measuring, and Why the Number Changes Daily

Heart rate variability, or HRV, is the metric that quietly took over the biohacking wearable category. Whoop built its whole business model around it. Oura scores your morning readiness on it. Withings, Fitbit, Garmin, and Apple all report it. And most people wearing these devices do not really know what HRV is, why it swings so much from night to night, or what a "good" number even looks like.

This piece is a working explanation of what HRV actually measures, the specific metric your wearable is showing you, why absolute numbers are close to meaningless across people, and how to read the trend well enough to use it.

What HRV actually is

Your heart does not beat at a metronomic rhythm. Even at rest, the interval between one beat and the next varies by tens of milliseconds — a little longer when you exhale, a little shorter when you inhale. That variation is HRV.

The variation is not noise. It is your autonomic nervous system doing its job. Your heart rate is under continuous, competing control from two branches of the autonomic system: the sympathetic (fight-or-flight, accelerator) and the parasympathetic (rest-and-digest, brake, via the vagus nerve). At rest, in a healthy adult, the parasympathetic side dominates, and the beat-to-beat interval flexes as vagal outflow modulates the sinoatrial node.

More vagal tone means more variation. Less vagal tone — because of stress, illness, poor sleep, alcohol, or intense training the day before — means the intervals become more uniform and HRV drops.

That is the whole story in one paragraph. HRV is a proxy for parasympathetic activity, which is a proxy for your body's readiness to recover. It goes down when your system is under load and back up when it is recovered. This is a mechanistic story that has been well-established in the clinical literature since at least the 1990s, when the Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology published the standards paper that basically defined the field (Circulation, 1996).

RMSSD, SDNN, and LF/HF — what your wearable is actually showing

HRV is not one number. There are dozens of ways to summarize the beat-to-beat variation. The three most important ones are:

• RMSSD (Root Mean Square of Successive Differences). Calculated in milliseconds. Reflects primarily parasympathetic (vagal) activity. Robust to short recording windows. This is what nearly every consumer wearable actually reports when it says "HRV."
• SDNN (Standard Deviation of Normal-to-Normal intervals). Reflects total variability from both sympathetic and parasympathetic influence. Requires longer recording windows (24 hours is standard) to be meaningful. Used more in clinical research than consumer wearables.
• LF/HF ratio (Low Frequency to High Frequency). A frequency-domain metric that was for years marketed as showing sympathetic-parasympathetic balance. Modern interpretation is much more cautious — LF is contaminated with parasympathetic activity too, and the ratio does not cleanly represent what it was originally sold as. Some clinical papers still use it. Most consumer wearables do not.

When Withings, Fitbit, Oura, Whoop, or Apple show you an HRV number in milliseconds, they are almost always showing RMSSD. That is the standard for consumer wearables because it can be reliably calculated from a short overnight window (5 to 10 minutes) and it maps cleanly to vagal tone. The Task Force paper endorses RMSSD specifically as the recommended time-domain metric for short recordings.

Knowing the metric matters because comparing across apps only works when you know they are measuring the same thing. Whoop and Oura both report RMSSD in ms. So do Withings and Fitbit. Their algorithms for choosing the measurement window differ, which is the main reason two devices worn the same night can show different HRV numbers. Neither is wrong; they are sampling slightly different windows.

Why absolute numbers vary hugely between people

If you Google "what is a good HRV," you will find dozens of tables offering age-and-sex-adjusted ranges. Most of them are close to useless for individual interpretation. Here is why.

Age. HRV declines with age, and the decline is steep. Median RMSSD in healthy 20-year-olds is roughly 60 to 80 ms. In healthy 60-year-olds it is often 20 to 30 ms. Neither person is unhealthy relative to their peers.

Sex. Females tend to have modestly higher HRV than males of the same age at rest, though the differences are small compared to age effects.

Breathing rate. Slow breathing (roughly 6 breaths per minute) dramatically raises HRV because it phase-locks to the natural cardiac-respiratory oscillation. This is one of the reasons breathwork and meditation apps that emphasize slow breathing can move HRV scores.

Fitness level. Aerobic training raises resting HRV over months and years. Elite endurance athletes can have RMSSD values in the 100+ ms range. This is a training adaptation, not a health warning about anyone else.

Genetic baseline. Individual baseline HRV has a substantial heritable component. Two equally healthy 40-year-old men can have resting RMSSDs of 25 ms and 65 ms respectively, and the difference is largely genetic.

Put all this together and the takeaway is: your HRV number is not comparable to anyone else's. Do not compare with your training partner. Do not benchmark against a table on the internet. The only useful comparison is with your own trend.

Why the number changes so much night to night

Once people internalize that only their own trend matters, the next confusion is why the trend itself is so jumpy. HRV commonly swings 15 to 30 percent night to night in the same person. Reasons include:

• Alcohol. The single most reliable HRV-suppressor in most people. Two drinks in the evening will typically drop overnight RMSSD by 20 to 40 percent. The effect lingers into the second night.
• Training load. A hard endurance or resistance session pushes HRV down for 24 to 72 hours as the body clears metabolic waste and repairs tissue. This is not bad — it is exactly what recovery looks like.
• Illness. HRV often drops one to two days before you feel symptoms of a cold or flu. Some people use a persistent multi-day dip as an early warning to add sleep before the illness lands.
• Sleep quality. Restricted sleep, disrupted sleep, and late bedtimes all suppress next-night HRV.
• Late meals. Eating within two hours of bed measurably lowers overnight HRV in most people. Digestion is sympathetic work.
• Psychological stress. Acute stress registers as reliably in HRV as alcohol does. A difficult conversation before bed, a work deadline, or a rumination-heavy evening will typically drop the number.
• Menstrual cycle. In menstruating individuals, HRV varies systematically across the cycle — typically higher in the follicular phase and lower in the luteal phase. This is normal and predictable.
• Room temperature and hydration. Both matter more than most people expect. A hot room and mild dehydration both push HRV down.

Any single night's number is a noisy sample of a system with many inputs. Reading a single-night HRV as a verdict on your health is like reading a single stock price as a verdict on the economy.

Why nightly measurement is the honest window

Consumer wearables that measure HRV during the day struggle. Motion artifacts, changing breathing rate, transient sympathetic activation from a phone notification, walking, talking — all of these confound the reading. Some devices attempt "stress HRV" during the day; the readings are often more noise than signal.

Overnight measurement is much cleaner. You are still. You are breathing at a stable rate. Sympathetic tone is at its daily minimum, so parasympathetic activity is the dominant signal. Most modern devices sample HRV during a specific window — typically the last portion of deep sleep or the first portion of slow-wave sleep after sleep onset — because that window gives the most consistent, comparable reading night over night.

This is why an overnight-wear HRV number, sampled the same way each night, is the metric worth watching. Daytime spot-checks with a chest strap or finger sensor can also be useful for guided breathwork, but for trend-tracking, the same time, same posture, same recording window — every night — is the setup that produces a signal you can actually interpret.

Devices in the catalog that measure HRV well

Two we would specifically recommend for HRV tracking.

Both report nightly RMSSD, both have adequate PPG accuracy for trend-tracking (though not as clean as under-mattress signal), and both survive with reasonable battery life. Neither is a magic pill. Both will give you a curve worth watching if you wear them consistently.

How to read your own HRV trend

A practical protocol.

1. First 30 days: baseline. Wear the device every night. Do not react to individual readings. At the end of the month, note your 30-day average RMSSD and the day-to-day standard deviation. That is your baseline.
2. Watch the 7-day rolling average, not any single night. Whoop popularized this framing for good reason. A one-night dip below your average is noise. A five-night trend below is signal. A two-week sustained shift is a change you should try to explain.
3. Log the obvious inputs. Alcohol, hard training, late meals, sleep duration, whether you were sick. You do not need an app for this — a note in your phone works. The point is to correlate the trend with real-world causes.
4. Do not chase the number. Trying to "raise HRV" as an end in itself is the wrong game. HRV is downstream of sleep, training load, alcohol, and stress. Fix the upstream inputs and the number moves.

What HRV cannot tell you

Being honest about the limits.

• HRV is not a diagnostic instrument. It will not diagnose an arrhythmia (an ECG does that). It will not diagnose diabetes, hypertension, or sleep apnea.
• HRV in athletes is not a clean readiness score by itself. Elite endurance training can produce "low" HRV that is not a warning sign — it is a training adaptation. Interpretation requires knowing the training context.
• Interindividual comparison is not meaningful. Your HRV is not better or worse than anyone else's absolute number.
• Wearable HRV is not clinical-grade HRV. Photoplethysmography-derived HRV is noisier than ECG-derived HRV. It is fine for trend-tracking. It is not the number that goes into a research paper.

The bottom line

HRV is one of the few metrics your wearable measures that has clear mechanistic meaning, clinical validation going back thirty years, and a clean interpretation as a proxy for parasympathetic tone. It is also one of the most misused. The absolute number is close to meaningless across people. The trend within one person, over weeks, is genuinely informative.

Wear the same device every night. Watch the seven-day rolling average. Log alcohol, training, and sleep. Fix the upstream inputs, and the number takes care of itself. That is the honest version of HRV tracking, and it is a useful practice for anyone serious about recovery.