What is HRV (heart rate variability)

Published September 30, 2026 · Educational content reviewed against peer-reviewed literature

HRV (heart rate variability) is one of the data points that raises the most questions for people using a wearable: why it goes up, why it goes down, and why yesterday's number doesn't look like today's. This guide explains what it measures, what moves it, and how to read it without over-interpreting.

What HRV is

The heart doesn't beat at perfectly even intervals. The time between one beat and the next varies constantly, in milliseconds, depending on how signals from the sympathetic (activation) and parasympathetic (calm) nervous system alternate over the sinus node. HRV measures exactly that variation: how much the interval between beats changes.

One of the most commonly used ways to calculate it is RMSSD (root mean square of successive differences between beat-to-beat intervals). It's the metric Nurik uses. A higher RMSSD doesn't mean a "healthier heart" in an absolute sense; it means more beat-to-beat variation in that measurement window.

Why it varies day to day

HRV isn't a fixed number for a given person: it changes with the previous night's sleep, stress level, alcohol, illness, recent training, hydration, and even the phase of the menstrual cycle. It's sensitive because the autonomic nervous system reacts to all of that, and HRV is, in large part, a reflection of that system.

That has a practical consequence: a single day's value says almost nothing on its own. What matters is the trend against your own recent average, not the isolated number.

What it means when it goes up or down

As a general guide, not an absolute rule: HRV above your recent normal usually accompanies good recovery — you slept well, you're not coming off a hard effort, there's no illness or accumulated background stress. HRV below your recent normal is associated with higher nervous system activation — fatigue, stress, a demanding training session the day before, or the onset of illness.

It's an association, not a single cause or a diagnosis: HRV drops because there's background activation, and that activation can come from many sources at once. It doesn't identify which one.

Why comparing your HRV to someone else's doesn't work

The normal HRV range between different people varies enormously — by age, genetics, training level, body composition and the measurement method used. Two healthy people can have a nighttime RMSSD of 25 ms and 80 ms without one being "better" than the other. Comparing absolute values between people has no scientific basis.

What does make sense: comparing yourself to yourself

What's comparable is your HRV today against your own recent average. That's why Nurik builds your "normal" from your own history, not from a population table, and shows it as a trend against that personal baseline — never as a ranking between users.

How to measure it well

HRV can be measured at different times of day, but the overnight measurement, during sleep, gives the most stable signal: the body is still, without the noise of movement, active digestion or conversation that distort the reading during the day.

Nurik calculates nighttime HRV as the median of the 90-second RMSSD windows that fall within the stable stretch of sleep — from 30 minutes after falling asleep until 30 minutes before waking — discarding wake periods, arousals, and windows outside the plausible physiological range (5–250 ms; outside that range it's treated as an optical-sensor artifact, not real data). A minimum of 20 valid windows (about 30 minutes of data) is required for a night to count; if that minimum isn't reached, no number is shown for that night.

The median is used instead of the average or the last sample of the night because the average is sensitive to movement artifacts and isolated irregular beats — to which RMSSD is particularly sensitive — and the last sample of the night tends to fall right when the body starts waking up, which systematically biases the value downward.

What moves the baseline ("your normal")

Nurik compares each night's HRV against the mean of the last 7 nights (with a minimum of 5 nights of history to start showing a trend). That comparison uses the natural logarithm of RMSSD (lnRMSSD): HRV isn't evenly distributed on a linear scale, and transforming it to a logarithmic scale makes night-to-night comparisons statistically more reasonable.

The trend is flagged as meaningful when the change exceeds the smallest worthwhile change (SWC) of that baseline — a threshold designed to avoid reacting to normal night-to-night measurement noise. With fewer than the minimum number of nights, no trend is shown: Nurik avoids inventing a comparison without enough history.

The limit of the optical sensor

The band derives beat-to-beat intervals from an optical sensor, not an ECG. At rest and during sleep, the agreement with ECG is good, but RMSSD is very sensitive to poorly detected beats, and in the presence of frequent irregular beats (extrasystoles, atrial fibrillation) the value stops being interpretable. If your HRV looks erratic without explanation, that's a reason to see a doctor, not to interpret the trend.

What HRV is not

How Nurik uses it

In the app, nighttime HRV feeds Recharge, the 0–100 score that summarizes how much battery you have to train today, together with resting heart rate, sleep, and blood oxygen from the previous night, always compared against your own average. The HRV trend (whether it's above, within, or below your normal range) also factors into the day's movement goal adjustment. Your HRV is never compared against another user's, nor shown as a ranking.

This page is general educational information, not medical advice. Nurik is not a medical device, does not diagnose disease, and does not replace a health professional's evaluation. If you have symptoms such as chest pain, disproportionate shortness of breath, dizziness or palpitations, consult a health professional instead of interpreting this data on your own.

References

Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology, 1996 (HRV measurement standards) · Plews DJ et al., 2013, Sports Medicine (HRV in sport, use of lnRMSSD and smallest worthwhile change) · Hopkins WG, 2009 (smallest worthwhile change).

The formulas and measurement windows described correspond to Nurik's current implementation, documented in internal product specifications (readiness-meta-diaria.md, glosario.md).

Want to see your HRV and your Recharge in a real app?

▶ Try the live app Reserve for $30 USD

© 2026 Nurik Health · Santiago, Chile