How to test your smart ring against a chest strap in seven nights

Why test your own ring

If you bought a ring that charges no subscription, you probably bought one with little published accuracy evidence. The strongest ring study we found, Dial et al. in Physiological Reports (2025), tested Oura Gen 3 and Gen 4 against a Polar H10 chest strap. The no-fee rings have, at most, a preprint (RingConn Gen 2) or nothing peer-reviewed at all (Galaxy Ring, Ultrahuman Ring PRO, Circular Ring 2). Our smart ring accuracy guide lays out every study.

Even a good study only tells you about its participants on average. Liang et al. (Sensors, 2024) found Oura's HRV error was higher in older adults, with more than half of the older group above 10% (Liang et al.). Herberger et al. (Scientific Reports, 2025) found sleep results that were close on average but badly off for individuals (Herberger et al.). The only way to know how your ring does on your finger, at your age, is to check it. It costs a chest strap and a week of slightly less comfortable sleep, and the strap stays useful for training afterward.

What you need

Four things, and you probably own two of them already:

1. A chest strap heart rate monitor. The Polar H10 is the obvious choice because it's the device Dial et al. used as “the criterion measurement,” recording single-lead ECG (Dial et al.). Polar lists a CR 2025 coin cell rated about 400 hours, Bluetooth and ANT+, and memory for one session of up to 30 hours (Polar). Garmin's HRM 600 is rechargeable, rated up to 2 months at an hour a day, and tracks heart rate and HRV (Garmin). Wahoo's TRACKR is rechargeable over USB-C and sends HRV “with a compatible app,” but Wahoo's own app doesn't record HRV (Wahoo). Our best chest strap for HRV compares them in full.
2. An app that records the strap's beat-to-beat data. For HRV you need an app that logs beat-to-beat (RR) intervals and can show rMSSD for a time window you choose, or export the raw intervals. We haven't verified any specific app, so we don't name one; check the app's own description for RR logging and rMSSD before you rely on it. Heart rate alone is easier: almost any strap app records it.
3. Your ring's app, with the numbers visible. You need to see the ring's nightly resting heart rate and HRV, not just a score. That's a catch for Oura owners: without the membership, Oura says you'll only see three daily scores (Readiness, Activity and Sleep) plus battery, profile, settings and Explore content (Oura Support). New members get one included month, so run your test inside it rather than paying $5.99 a month just to check.
4. A spreadsheet or a notebook. One row per night, columns for the ring's numbers, the strap's numbers, the time window and anything unusual about the night.

Before you start: fit and setup

A bad fit makes a fair ring look bad. Sort these out before night one:

• Size the ring with its own brand's kit. No brand says its sizes match another's, and Oura says its sizes “differ slightly from standard US ring sizes.” Our smart ring sizing guide walks through it.
• Use the finger the brand recommends. Oura recommends the index finger for best accuracy (Oura Support). Pick one finger and keep it there for the whole test. See which finger to wear a smart ring on.
• Wear the ring for a few normal days first. Rings build a baseline over time, so start logging once the app is showing nightly numbers routinely.
• Charge everything. A ring that dies at 4 a.m. or a strap app that stops recording ruins a night. The Polar H10's coin cell avoids charging; the rechargeable straps need topping up.
• Put the strap on the way its maker describes. Follow the manual for placement and contact.

The protocol, step by step

1. Decide what you'll compare. Resting heart rate and HRV (as rMSSD, the measure the studies use). If your ring app shows a graph through the night, you can compare matching blocks; if it only gives one nightly number, you'll compare that number with the strap's figure for the same stretch of the night.
2. Put on the strap at bedtime and start recording. Write down the start time. With the Polar H10 you can also store the session on the strap itself, up to 30 hours, if the phone disconnects.
3. Sleep normally. Same finger, same bedtime routine. Don't change anything else during the test week if you can help it.
4. In the morning, stop the recording and note the end time.
5. Read the ring's numbers. Log its nightly resting heart rate and HRV, and the time window it covers if the app shows one.
6. Pull the strap's numbers for the same window. Use a block of at least 30 minutes, not a single reading. Liang et al. found that windows of 30 minutes or longer, with stricter data-quality filtering, improved accuracy when deriving HRV. If the ring gives one number for the whole night, use the strap's average over the same sleep period.
7. Log both, with notes. A late workout, a drink, a cold, a bad night: write it down. Unusual nights are still data, but you'll want to know why one stands out.
8. Repeat for at least seven nights. One night can mislead in either direction. Dial et al. ran 536 nights across 13 people precisely because single nights vary.
9. Optional: a daytime seated check. Sit still for 10 minutes with both on and compare heart rate. That mirrors the setup of the Tao et al. preprint on the RingConn Gen 2 (seated, 10-minute recordings). It's a quick sanity check, not a substitute for the nights.

The math, with a worked example

For each night, work out the percentage difference between the ring and the strap:

Here is how that looks over a week. These numbers are made up to show the arithmetic. They are an illustration, not test data from any ring.

Reading the example: night 1's heart rate difference is (53 − 52) ÷ 52 × 100 = +1.9%. Add up the seven heart rate differences without their signs (1.9 + 0 + 2.0 + 1.8 + 1.9 + 2.0 + 0) and divide by seven: about 1.4%. Do the same for HRV and you get about 5.6%. In this made-up week, the ring's resting heart rate misses by 1.4% on a typical night and its HRV by 5.6%, both in line with what Dial et al. reported for Oura Gen 4 (1.94% and 5.96%). Notice the HRV misses go both ways; the bias of -2.1% is smaller than the typical miss, which is what you'd hope to see.

What differences are normal

There's no official pass mark for a consumer ring. The fairest benchmarks are the published results, so compare your numbers with these:

How to read your result against those:

• Near Dial's figures (about 2% heart rate, 6-7% HRV): your ring is performing on you about as well as the best-validated ring did in the best study.
• HRV off by 10% or more, heart rate fine: if you're older, or your HRV runs very high, that's consistent with the studies. Liang found larger HRV error in older adults, and the5krunner noted Oura's accuracy dropped above roughly 60 ms rMSSD (the5krunner). Read your HRV as a trend, not a precise number.
• Heart rate off by well over 2%: investigate. Heart rate is the easier measurement, and it's where rings did best in every study. Start with fit, below.
• A consistent bias with a small spread: a ring that always reads, say, a beat high is still useful for trends, because the trend moves together. The Tao preprint showed exactly this kind of small, consistent bias for heart rate.

For the studies behind those benchmarks, see smart ring HRV vs a chest strap.

If you test during exercise

You can run the same comparison during a workout, but set your expectations first: we found no peer-reviewed study of ring heart rate during exercise, and the reviewers who compared rings with a strap on the move saw problems. DC Rainmaker found Oura Ring 5 “largely pretty good” against a Polar H10 for indoor cycling and road running, until a trail descent where “the HR accuracy basically falls apart for the entire rest of the run” (DC Rainmaker). Gadgets & Wearables found RingConn Gen 2 “consistently reported higher values” than a chest strap on runs (Gadgets & Wearables).

If you want to try it anyway:

• Start with a steady indoor session (a bike or treadmill at an even effort), where reviewers saw the best results.
• Record both devices for the whole session and compare average heart rate over the steady middle part.
• Expect bigger gaps than at night, and treat any rough-terrain or interval result as a stress test, not a verdict.

Whatever you find, the evidence points the same way: use the strap for training and the ring for sleep and recovery. More in smart ring heart rate accuracy.

What to do if it's off

1. Check the windows first. Most big gaps come from comparing different stretches of the night. Make sure the strap numbers cover the same period as the ring's.
2. Check the fit. Re-size with the brand's own kit if the ring spins or slides. Move to the finger the brand recommends; for Oura, the index finger.
3. Run another seven nights. Change one thing at a time (finger, size) so you know what fixed it.
4. Contact the brand. Share your log. A week of side-by-side nights is far more useful to a support team than “it seems off.”
5. Know your return window. Test early. Circular, for example, offers 14-day returns (Circular), and rings bought on Amazon follow Amazon's return process. Our warranty and returns guide has each brand's terms.

Mistakes that spoil the test

• Comparing a single night. Seven is the minimum; more is better.
• Comparing against a watch. A watch reads pulse with light, like the ring. Checking one estimate against another tells you whether they agree, not whether either is right. Use a chest strap.
• Comparing minute-by-minute readings. Use 30-minute-plus windows or whole-night averages.
• Mixing HRV measures. Compare rMSSD with rMSSD. Cao et al. found Oura's rMSSD accurate but its frequency-based HRV measures (LF and LF:HF) had high error (Cao et al.).
• Changing fingers mid-test. Pick one and stick with it.

If you're still choosing a ring, pick one knowing you can check it: the best smart ring for sleep tracking and the best smart rings without a subscription rank the options, and our RingConn Gen 2 review covers the no-fee ring with the only chest-strap preprint. Everything else on accuracy is in the accuracy hub.