HRV on Apple Watch: What the Number Means (and Doesn't)

The short answer: Apple Watch HRV is accurate enough to build a personal baseline from — validation studies show good agreement with chest-strap and lab ECG measurements, especially for readings taken at rest and overnight — but it isn’t a clinical instrument, and its number (SDNN) shouldn’t be compared with other devices or other people.

Apple Watch has quietly become one of the most common ways people encounter HRV at all, since it measures it passively, mostly overnight, without any deliberate effort from the wearer. That convenience raises an obvious question: how much can you actually trust the number it gives you? For the broader context on what HRV means and how to use it, see HRV, sleep, and training readiness for lifters — this piece focuses specifically on the Apple Watch measurement itself.

How Apple Watch actually measures HRV

Apple Watch uses an optical sensor on the back of the case, photoplethysmography, which shines light into the skin and measures tiny changes in blood volume with each heartbeat to infer timing between beats. That’s a genuinely different technology than the chest-strap ECG (electrocardiogram) most lab validation studies use as the reference standard, which reads the heart’s actual electrical signal directly.

The metric Apple reports is SDNN, the standard deviation of time between normal heartbeats, measured in milliseconds, most often captured during an overnight sleep window when the watch has a stable, low-motion signal to work with.

What the validation research actually shows

Multiple independent studies have compared Apple Watch HRV readings against lab-grade ECG. A study on the Apple Watch Series 6 found near-perfect agreement for heart rate and beat-to-beat interval measures at rest, with a mean absolute error around 1%, but noticeably more variability specifically in the HRV metric itself during resting conditions, with error in the range of 30% for some HRV-specific measures. A separate validation of Series 9 and Ultra 2 against a Polar H10 chest strap found good agreement under free-living, real-world conditions specifically, which matters since lab conditions and daily life aren’t the same test. An earlier study focused on relaxation and mild mental stress found very good reliability and agreement, above 0.9, and found the Apple Watch signal was able to reflect real stress-induced changes.

Read together, the pattern is consistent: Apple Watch performs well for heart rate and beat timing generally, and reasonably well for HRV specifically at rest, with accuracy that’s good but not identical to clinical ECG. That’s a meaningfully different claim than “as accurate as a hospital monitor,” and also a meaningfully different claim than “not accurate enough to be useful.”

Why the overnight reading is the one that matters

Apple Watch can technically read HRV at other points during the day, but the overnight measurement, taken during stable sleep with minimal motion, is both the most reliable technically and the most useful practically, since it reflects your recovery state heading into the next day rather than a moment mid-activity. Spot-checking HRV during the day introduces motion artifact and captures a snapshot influenced by whatever you were doing right before, a stressful call, a flight of stairs, caffeine, rather than your underlying recovery status.

If you’re building a personal baseline, the overnight reading, checked consistently over weeks, is the number worth anchoring to. That baseline, not any single night’s reading, is what Corpus’s Recovery score actually weighs.

What to actually do with the number

Nothing changes about the underlying advice just because the measurement comes from your wrist instead of a chest strap: don’t compare your absolute number to anyone else’s, watch your own trend relative to your own baseline, and read it alongside sleep, training load, and how you actually feel rather than in isolation. A validated-but-imperfect measurement is still a real signal. It’s just not a diagnosis.

The takeaway

Apple Watch HRV is good enough to build a genuine personal baseline from, validated against lab equipment with real, published agreement, particularly for the overnight reading. It’s not a clinical-grade instrument, and it was never trying to be one. Used the way it’s meant to be used, tracked consistently, read as a trend, combined with sleep and training context, it’s a legitimately useful recovery signal.