HRV (Heart-Rate Variability)

Why HRV is a useful signal

Your heart does not beat at a perfectly regular pace. Even at rest, the time between successive beats varies — and the pattern of that variation is the cleanest peripheral physiological signal you can measure that correlates with autonomic nervous system state. The autonomic nervous system, in turn, governs sleep stages, recovery, and stress response. Measuring HRV at the wrist or finger gives a consumer device a window into this state without requiring electrodes on the chest.

The metrics that matter

The two most-cited HRV time-domain metrics are RMSSD (root mean square of successive differences) and SDNN (standard deviation of normal-to-normal intervals). RMSSD reflects parasympathetic tone (the rest-and-digest system); higher RMSSD over time generally indicates better recovery. SDNN reflects overall variability across both sympathetic and parasympathetic systems. Both are reported in milliseconds.

Frequency-domain HRV metrics (LF, HF, LF/HF ratio) measure the same underlying signal in a different mathematical decomposition. Most consumer trackers use time-domain metrics because they’re more robust to short measurement windows.

How smart rings and smartwatches measure it

Optical sensors (PPG — photoplethysmography) at the wrist or finger illuminate skin with green LEDs and measure the light absorbed by capillary blood. The pulse waveform produced by each heartbeat shows up as a peak; the time between peaks is the RR interval. From a continuous stream of RR intervals, the device computes RMSSD and SDNN.

Finger-worn devices (smart rings) generally produce cleaner HRV measurements than wrist-worn devices because the finger has tighter blood-vessel coupling and less motion noise. This is why HRV-based sleep staging is more accurate on rings than on watches.

Why this matters for our verdicts

HRV is the dominant signal in our smart ring and shift-worker sleep-tracker verdicts. The Oura Ring 4 wins both because its HRV measurement (validated against medical-grade chest-strap references in published studies) is the most accurate consumer implementation we have access to.

For training-load and recovery purposes, HRV trends over time matter more than absolute values. Your baseline HRV is your baseline; what’s actionable is whether today’s reading is meaningfully below your 30-day rolling average. Garmin, Whoop, and Oura all surface this trend.

What HRV is not

HRV is not a direct measure of sleep stages, fitness level, or recovery readiness. It is a peripheral signal correlated with these things, with a meaningful but imperfect relationship. The marketing in this category tends to overclaim the directness of the relationship; the academic literature is more careful. If a wearable says you “had a bad recovery night” because your HRV was low, the best interpretation is “your autonomic system was tilted toward sympathetic activation last night, which is associated with — though not identical to — worse recovery outcomes.”

Related concepts

For the broader framing of how consumer wearables compare against medical references, see sleep stages. For the machine-learning models that consumer devices use to translate raw HRV into sleep-staging predictions, see machine learning.