What Is the Most Accurate Wearable for Respiratory Rate? Why the Numbers Do Not Compare

Short answer: Three consumer devices publish a respiratory rate accuracy figure, and they cannot be ranked against each other, because none of them is measuring the same thing the same way. Oura reports an average error under 1 breath per minute against an ECG-derived reference, over one night of sleep. Whoop's independent validation reports a 1.8% bias and 6.7% precision error against polysomnography, also overnight. Alveos One reports landing within 2 breaths per minute of laboratory equipment in 94.6% of measurements at rest and 79.3% during graded exercise. Different references, different conditions, different statistics. Anyone who tells you which is most accurate is comparing a mean to a percentage to a hit rate.

Say you are choosing between two devices and both say "clinically accurate." One claims a 0.71 breaths-per-minute average error. The other claims 94.6%. The first number sounds smaller and therefore better, which is the wrong instinct: 0.71 is an average and 94.6% is a proportion, and they are not on the same scale, or even measuring the same property. This is the most common way accuracy claims mislead, and it is almost never deliberate.

What are these numbers actually measured against?

Every accuracy figure is a comparison, so the first question is always: compared to what?

The laboratory references for breathing are capnography, which measures carbon dioxide in exhaled air, and inductance plethysmography, which measures the chest and abdomen expanding through bands. Both watch the act of breathing.

An ECG-derived respiratory rate does something different. It infers your breathing from the rhythm your breath imposes on your heartbeat, the same respiratory sinus arrhythmia that powers the respiratory rate on your ring or watch. It is a good estimate. It is still an estimate, which means a device validated against it is being scored against another approximation, not against the breath itself.

So when Oura reports an average error of 0.71 breaths per minute in its Need For Sleep study of 43 adults, the comparison was against an ECG-derived rate on one night per person (Oura). That is a real result. It answers a narrower question than "how close is this to your actual breathing."

What does each device actually report?

Oura. Average error under 1 breath per minute, 0.71 in the reported study, against an ECG-derived reference, 43 healthy young adults, night-time sleep only.

Whoop. The often-cited independent validation is Berryhill and colleagues in the Journal of Clinical Sleep Medicine, 32 healthy participants wearing the device through one night of polysomnography. Respiratory rate came out at 15.7 breaths per minute against the lab's 15.6, with a bias error of 1.8% and a precision error of 6.7% (Berryhill et al., 2020). Note the units. Those are percentages, not breaths, and the widely repeated "within about one breath per minute" is a conversion of the precision error rather than a phrase the paper uses.

Alveos One. In a University of Kent study, published as a preprint in June 2026, respiratory rate landed within 2 breaths per minute of laboratory reference equipment in 94.6% of measurements at rest and 79.3% during graded exercise, across 20 healthy adults. That is a hit rate against a threshold, not an average error, and it is the third incompatible statistic in this article.

If you want the two market leaders set against each other feature by feature, we did that in Alveos One vs Oura vs Whoop.

Apple Watch, Fitbit, Garmin and the rest. They report a nightly respiratory rate and publish no validation figure for it that we could find. Absence of a number is not evidence of a bad one, but you cannot check what nobody publishes.

Why three good studies still do not settle it

Line them up and the problem is obvious. The references differ: an ECG estimate, polysomnography, laboratory equipment. The statistics differ: a mean absolute error, a bias-and-precision pair in percent, a proportion of readings inside a 2-breath window. The conditions differ, and this is the one that matters most for daily use, because the first two validations were run while people slept.

Sleep is the easy condition. The body is still, the breathing is regular, and the motion that wrecks these measurements is absent. It is entirely reasonable to validate a sleep metric during sleep. It just means the resulting number describes the device asleep, and says nothing about it on a commute, in a meeting, or on a hill.

How to read an accuracy claim

Four questions get you most of the way, and they work on any device including ours.

What was the reference, and did it measure breathing or infer it? What does the number actually mean, given that a mean error, a percentage bias and a hit rate answer different questions? Under what conditions was it collected, since stillness and motion are different problems entirely? And who ran the study, and who paid for it. Manufacturer-funded work is not automatically wrong, but it earns the scrutiny you would give anyone marking their own homework.

Where Alveos One fits: We are not going to tell you Alveos One is the most accurate breathing wearable, because this article has just explained why nobody can support that sentence. What we can say is what was tested. The University of Kent's School of Sport and Exercise Sciences ran the study, it is published as a preprint and not yet peer reviewed, we funded it, and it covered rest and graded treadmill exercise rather than sleep alone. Respiratory rate landed within 2 breaths per minute of the laboratory reference in 94.6% of measurements at rest and 79.3% during exercise, and the reusable magnet mount matched a medical-style adhesive patch, beating it during exercise. The exercise figure is the lower one and we publish it anyway, because a number that only holds when you are lying still is the thing this article is warning you about. Full methodology, limitations included, is on our science page. For the device-by-device picture, see our comparison of breathing wearables, and if your constraint is where a device sits rather than how it scores, we covered the trackers that are not watches, rings or bracelets.

When to see a professional

None of these devices is a diagnostic instrument, and a consumer accuracy figure is not a clinical one. Breathlessness that arrives suddenly, chest pain, or breathing that feels obstructed belong with a clinician, not with a wearable reading. Alveos One is a wellness device for tracking your breathing patterns. It does not diagnose or treat any medical condition.

FAQ

What is the most accurate wearable for respiratory rate?

There is no defensible answer, because the published figures are not comparable. Oura reports an average error under 1 breath per minute against an ECG-derived reference during sleep, Whoop's independent validation reports 1.8% bias and 6.7% precision against polysomnography during sleep, and Alveos One reports 94.6% of measurements within 2 breaths per minute of laboratory equipment at rest and 79.3% during exercise. A mean, a percentage pair and a hit rate cannot be ranked against each other.

How accurate is Oura's respiratory rate?

Oura reports an average error of 0.71 breaths per minute, described as accurate within 1 breath per minute across the night, from a study of 43 healthy young adults compared against an ECG-derived respiratory rate. It is a night-time measurement, and the reference it was scored against was itself an estimate.

How accurate is Whoop's respiratory rate?

An independent study of 32 healthy participants found 15.7 breaths per minute against polysomnography's 15.6, with a bias error of 1.8% and a precision error of 6.7%. The paper reports percentages; converting the precision figure gives roughly one breath per minute, which is where the commonly quoted phrasing comes from.

Do wearables measure breathing during the day?

Mostly not. Devices that infer breathing from the pulse need you still enough for the underlying signal to read cleanly, which is why the number appears on your sleep summary and nowhere else. Devices that sense breathing directly are not bound by that limitation, though they carry their own.

What is the gold standard for measuring respiratory rate?

Capnography, which measures carbon dioxide in exhaled breath, and inductance plethysmography, which measures chest and abdominal movement. Both observe breathing itself, which is why validation studies use them as the yardstick and never another wearable.

Does a manufacturer-funded study count?

It counts, and it should be read with that in mind. What matters is whether the methods, the conditions and the limitations are published so someone else can check them. Ours was funded by us and run by the University of Kent, and the preprint states as much.

Sources

• Oura, "How Accurate Is Oura's Respiratory Rate?" (Need For Sleep Study, 43 adults, ECG-derived reference): ouraring.com

• Berryhill S, Morton CJ, Dean A, et al. "Effect of wearables on sleep in healthy individuals: a randomized crossover trial and validation study." J Clin Sleep Med. 2020;16(5):775-783: pmc.ncbi.nlm.nih.gov

• Oura, "How Oura Measures Respiratory Rate From Your Finger": ouraring.com

• Alveos science page (University of Kent validation study, respiratory rate within 2 breaths/min of reference in 94.6% of measurements at rest and 79.3% during exercise): alveoslabs.com