Which wearable is most accurate for HRV?
In a 2025 overnight test against a chest strap, Oura came closest, then WHOOP, Garmin and Polar. Apple Watch wasn’t in that test. For tracking your own trend, wearing one device every night matters more than which one you pick.
Accuracy matters most when you compare devices or switch between them. For your own trend, a device that’s a few percent off in the same way every night still shows your changes. What breaks a trend is missing nights and mixing devices.
How close each one came
The 2025 study put several devices on 13 healthy adults at once, at home, for 536 nights, with a chest strap ECG as the reference.
| Device | Agreement with the strap (1.00 is perfect) | Average error |
|---|---|---|
| Oura Gen 4 | 0.99 | 6 percent |
| Oura Gen 3 | 0.97 | 7 percent |
| WHOOP 4.0 | 0.94 | 8 percent |
| Garmin Fenix 6 | 0.87 | 11 percent |
| Polar Grit X Pro | 0.82 | 16 percent |
Two WHOOP scientists replied that the test compared WHOOP’s number, which leans on late deep sleep, with the strap’s whole-night average, so part of WHOOP’s gap came from the window rather than the sensor. The study was also small, and only in healthy adults.
What about Apple Watch?
In a separate 2024 test of 39 adults against a Polar H10 strap, Apple Watch Series 9 and Ultra 2 came in about 8 ms low on average. Apple Watch also reports SDNN, a different measurement from the RMSSD above.
The limits of optical sensors
Rings and watches time your pulse with light, and a pulse at the skin isn’t quite a heartbeat. In 103 athletes, an optical sensor gave an average RMSSD of 81 ms where an ECG gave 104. It also caught only 17 to 57 percent of the changes the ECG flagged, 1.8 to 5.5 days late. In a sleep lab study of Oura, accuracy was good once at least 80 percent of beats in each five-minute window were clean and windows were averaged over 30 minutes or more. More than half of the people over 45 still had errors above 10 percent.
What we’d do: pick the device you’ll wear every night. If you want the most exact number, a chest strap with an app, used the same way every morning, is the reference the studies use.
5 sources
- Dial MB et al., Validation of nocturnal resting heart rate and heart rate variability in consumer wearables, Physiological Reports, 2025pubmed.ncbi.nlm.nih.gov
- Grosicki GJ, Presby DM, Accurate comparison of wearables requires contextual equivalence, Physiological Reports, 2025pubmed.ncbi.nlm.nih.gov
- O'Grady B et al., The validity of Apple Watch Series 9 and Ultra 2 for serial measurements of heart rate variability and resting heart rate, Sensors, 2024pmc.ncbi.nlm.nih.gov
- Liang T, Yilmaz G, Soon CS, Deriving accurate nocturnal heart rate, rMSSD and frequency HRV from the Oura Ring, Sensors, 2024pubmed.ncbi.nlm.nih.gov
- Ben-David K et al., Pulse rate variability is not the same as heart rate variability, Sports Medicine Open, 2026pubmed.ncbi.nlm.nih.gov