Why do HRV and resting heart rate readings differ between wrist, bicep, and chest-strap trackers?

This is a different problem from why your HRV baseline suddenly shifts over weeks or why different devices' biological-age algorithms disagree. Those are about drift over time or different math applied to similar inputs. This one is about a single instant: three sensors reading the same heartbeat, right now, and reporting three different numbers because of where they sit and how they work.

PPG vs. ECG: two fundamentally different ways to "see" a heartbeat

ECG (electrocardiography) reads the heart's electrical activity directly through skin contact — it's precise enough that it's the reference standard hospitals and validation studies use to judge every other method against. Chest straps use ECG because the chest sits close to the heart and provides a strong, stable electrical signal.

PPG (photoplethysmography) is what's inside virtually every wrist watch, ring, and arm band. It shines light into the skin and measures how much bounces back, which changes slightly with each pulse of blood. It's a clever proxy for heartbeat timing, but it's still a proxy: it can be fooled by motion (the sensor shifting against skin), by blood flow changes (cold hands, tight straps), and by skin tone and hair density affecting how much light returns. This is also why nearly every validation study — including the ones behind the numbers below — uses a chest strap as the "true" reference and measures how far each PPG placement drifts from it.

How much placement actually matters: the identical-device test

The cleanest way to isolate placement from device quality is to put the same device in different spots on the same person — which is exactly what a December 2025 study in the journal Sensors did. Researchers had 28 adults wear three identical WHOOP 4.0 units simultaneously (left wrist, forearm, upper arm) alongside a Polar Verity Sense arm sensor, a Garmin, and a chest-strap ECG reference, through rest, warm-up, high-intensity burpees, and a graded treadmill test.

The pattern across conditions: placement barely matters when you're still, and matters progressively more as movement and intensity increase — because motion artifact, not the sensor's baseline capability, is what separates a wrist reading from a chest-strap reading. A separate wrist-monitor validation study found the non-dominant wrist runs a systematic bias of roughly 2.5 bpm and a mean absolute error of about 6.4 bpm versus a reference device — small at rest, more consequential once you're trying to judge training zones from it.

Why RMSSD is more placement-sensitive than SDNN

There's a second, less visible reason two devices can disagree even when both are optical: which HRV formula they calculate. HRV isn't one number — it's a family of statistics computed from the same beat-to-beat interval data, and the formulas don't tolerate sensor imprecision equally.

• RMSSD (root mean square of successive differences) is what WHOOP, Oura, Fitbit, and most consumer wearables report. It's built from the differences between consecutive beats, which means it's exquisitely sensitive to small timing errors — and PPG sensors are less precise at pinpointing the exact instant of each beat than ECG is.
• SDNN (standard deviation of NN intervals) is what Apple Watch reports. It's a broader measure of overall variability across a longer window, which tends to average out small per-beat timing noise rather than amplify it.

Studies comparing PPG-derived HRV against ECG under controlled resting conditions have found SDNN holds up well — relative errors in roughly the 2-9% range — while RMSSD shows meaningfully more error, commonly cited in the 30-34% range under the same conditions. That gap is a direct consequence of the math, not of one brand's engineering being worse than another's: RMSSD amplifies exactly the kind of small timing jitter that optical sensors are prone to, and SDNN mostly doesn't. It's also a second reason — beyond placement — that switching between an RMSSD-reporting device and Apple's SDNN can look like your HRV "changed" when nothing physiological did.

A practical protocol: matching device, placement, and formula before you compare

1. Identify what you're actually comparing. Before concluding two readings disagree, check three things separately: is it the same physical placement, the same sensor technology (PPG vs. ECG), and the same HRV formula (RMSSD vs. SDNN)? A mismatch on any one of these can fully explain a gap that looks alarming at first glance.
2. Pick one placement for trend-tracking and stay with it. If you wear a band on your bicep for workouts and your wrist overnight, expect the two contexts to produce different absolute numbers — that's expected, not a malfunction. Track each context's trend against its own history rather than comparing bicep-daytime numbers to wrist-overnight numbers.
3. Trust wrist and ring data most for sleep, arm or chest-strap data most for exercise. Overnight, you're still, so the biggest source of PPG error — motion — is mostly absent, which is why overnight wrist and ring HRV tends to be fairly reliable. During structured training, an upper-arm band or chest strap is the more trustworthy read.
4. After any placement or device switch, treat the next few weeks as a new baseline, not a broken one. Give it time to accumulate before judging whether a change in your number reflects your body or your setup — a genuine recalibration typically takes several weeks on major platforms, not days.
5. For a single alarming reading, don't over-read it. A one-off spike or dip during a high-intensity moment is far more likely to be motion artifact than a real physiological event, especially on wrist-worn devices — the burpee condition above dropped every device's accuracy, not just one brand's.

When a real discrepancy is worth checking with a doctor

Placement and sensor-technology differences explain the overwhelming majority of "my devices disagree" situations, and none of them are medical concerns on their own. Treat it differently if:

• Your resting heart rate is elevated across every device and placement you own, not just one
• The disagreement doesn't track with movement — it's just as large sitting still as it was during exercise
• You notice new symptoms alongside the discrepancy: chest discomfort, unusual breathlessness, or fainting
• A chest-strap ECG reading itself looks abnormal (irregular rhythm, unusually high or low rate at rest)

A device disagreeing with itself across placements is a measurement story. A device agreeing with itself but reporting something genuinely abnormal is a different conversation, and one worth having with a doctor rather than another wearable.

Why this matters more once you're combining sources

Placement and formula differences are exactly the kind of noise that gets baked in silently if you casually mix HRV readings from different devices, wrist positions, or straps into one trend line — a bicep-band workout reading and a wrist-band overnight reading were never measuring quite the same thing, even on the same device. This is part of why Vita's cross-source merge picks one consistent source for a given metric across a window rather than splicing readings from different placements or devices into a single average — mixing them wouldn't produce a baseline that measures anything coherent, even though every number involved would display as "HRV." If you're trying to figure out whether a reading you're seeing reflects your body or your hardware, Vita's recovery score and Body Age are built to keep that distinction honest rather than quietly averaging it away.