Smart ring heart rate accuracy: close at rest, shaky on the run

Rest vs workout, in one table

“Is a smart ring's heart rate accurate?” has two different answers, depending on whether you're lying still or moving. The published evidence sits almost entirely on the still side. Here is the split:

So the honest answer: trust a ring's resting heart rate, and don't trust its workout heart rate until you've checked it against a strap yourself. The rest of this page shows why, and our smart ring accuracy pillar puts it next to the sleep evidence.

Resting heart rate: the evidence

Resting heart rate is the thing rings do best, and it's the thing the best study measured. Dial et al. (Physiological Reports, 2025) had 13 adults wear Oura Gen 3 and Gen 4 rings alongside a Garmin Fenix 6, a Polar Grit X Pro and a WHOOP 4.0 for 536 nights, with a Polar H10 chest strap as the reference. For Oura Gen 4, the ring sold as Oura Ring 4, nightly resting heart rate had a concordance of 0.98 and a mean absolute percentage error of 1.94% (Dial et al.). If the strap said 55 beats per minute, the ring typically said something within about one beat of it.

Two ECG studies agree. Cao et al. (JMIR, 2022) compared Oura with a Shimmer3 chest ECG in 35 healthy adults sleeping at home and found it “could accurately measure nocturnal HR and RMSSD,” with no conflicts declared (Cao et al.). Liang et al. (Sensors, 2024) compared Oura with lab ECG in 92 younger and 22 older adults and found strong agreement overall; the weakness they flagged was HRV error in older adults (Liang et al.).

Beyond Oura, the file thins out fast. The Tao et al. preprint (Research Square, 2026) had 64 athletes sit for 10-minute recordings after exercise wearing a RingConn Gen 2 and a Polar H10. Heart rate bias was -0.42 bpm, with limits of agreement from -0.74 to -0.09 bpm: the ring read a fraction of a beat low, consistently (Tao et al.). That's a tight result, but it's seated, low-motion, and not peer-reviewed.

Ultrahuman publishes an in-house pilot of the Ring AIR's sleep heart rate: 6 volunteers, 36 sleeps, a mean absolute error of 2.4-2.6 bpm and a bias of -1.2 to -1.8 bpm against a SleepImage ring and an Apple Watch (Ultrahuman). Three reasons to discount it: all the authors work for Ultrahuman, it isn't peer-reviewed, and it compares the ring with another ring and a watch rather than with a chest strap or ECG. The Ring AIR also isn't sold in the US; the US model is the Ring PRO, for which we found no independent data.

Workout heart rate: what reviewers saw

We found no peer-reviewed study of smart ring heart rate during exercise. What exists are careful reviewers comparing a ring with a chest strap, and their results are mixed at best.

DC Rainmaker on Oura Ring 5. Against a Polar H10, Ring 5's workout heart rate was “largely pretty good” for indoor cycling and road running. Then, on a trail descent, “the HR accuracy basically falls apart for the entire rest of the run” ( DC Rainmaker). Note the shape of that failure: not a brief spike, but a ring that didn't recover for the rest of the session.

Gadgets & Wearables on RingConn Gen 2. On runs, compared with a chest strap, “the ring consistently reported higher values,” overestimating both average and maximum heart rate (Gadgets & Wearables).

Tech Advisor on Circular Ring 2. The reviewer went further than one ring: “I've not tested a smart ring that I really think does a good job of tracking heart rate during exercise” ( Tech Advisor).

Why would a ring that nails resting heart rate struggle on a run? A ring reads blood flow in the finger with light, and exercise changes everything around that reading at once: the hand swings or grips, the ring can shift, and the signal has to be separated from all that movement. The evidence doesn't tell us which of those mattered in each reviewer's case. It tells us the result: steady efforts can be fine, rough ones can go badly, and you won't know which you got unless something else is recording too.

Smart ring vs smartwatch accuracy

The one study that measured rings and watches against the same chest strap is Dial et al., and at night the Oura rings came out ahead: they had the highest agreement with the Polar H10 of the five devices, beating a Garmin Fenix 6, a Polar Grit X Pro and a WHOOP 4.0. That's a real result, with two limits. The5krunner noted the Garmin and Polar watches were older devices (the5krunner), and the study measured only nighttime rest.

For exercise, we found no study comparing rings with watches. The reviewer evidence above is ring-versus-strap, not ring-versus-watch. So the fair summary of smart ring vs smartwatch accuracy is: at night, a good ring can match or beat a watch; during workouts, nobody has shown a ring holds up. Our smart ring vs smartwatch comparison covers the rest of that choice, including battery and cost, and Oura vs WHOOP covers the band that charges a membership too.

Ring by ring

• Oura Ring 4 (Gen 4): directly validated overnight against a Polar H10. The strongest heart rate evidence of any ring. See our Oura Ring 4 review.
• Oura Ring 5: no published study yet. DC Rainmaker's review is the best data: good on steady efforts, not on a trail descent. More in our Oura Ring 5 review.
• RingConn Gen 2: a seated preprint (bias -0.42 bpm) and a reviewer who saw it read high on runs. Our RingConn Gen 2 review weighs both.
• RingConn Gen 3 and Gen 2 Air: no published validation found.
• Samsung Galaxy Ring: no peer-reviewed validation found in our September 2026 searches.
• Ultrahuman Ring PRO: no independent data; Ultrahuman's pilot covered the Ring AIR, which isn't sold in the US.
• Circular Ring 2: no published validation found.

What to look for

If heart rate is why you're buying, sort your needs before you sort the rings:

1. Resting heart rate trends and recovery? Any ring with published overnight data will do the job; Oura has the most, RingConn Gen 2 has a preprint.
2. Workout heart rate for zones or training load? Buy a chest strap. No ring has published evidence here.
3. No monthly fee? Then the Oura evidence comes with a catch: Oura Ring 5 needs the Oura Membership for its full insights, $5.99 a month or $209.97 over three years on the annual plan. RingConn Gen 2 charges nothing. Our best smart rings without a subscription ranks the no-fee field, and the scoring weights are on our methodology page.

How to use ring heart rate well

• Read resting heart rate as a trend. A 2% typical error is small next to the changes that matter over weeks, like a resting rate that creeps up during a hard training block or a bad stretch of sleep.
• Don't set training zones from a ring. Use a strap, or a device you've confirmed against one.
• Check your own ring. A week of nights with a strap tells you whether your ring, on your finger, lands within the range the studies found. Our test your ring against a chest strap guide has the steps and a worked example, and the best chest strap for HRV helps you pick the strap.
• Fix the fit first. Oura recommends the index finger for best accuracy (Oura Support), and size the ring with that brand's own kit, since no brand says its sizes match another's. See which finger to wear a smart ring on.

If HRV matters to you as much as heart rate, read smart ring HRV vs a chest strap next; HRV has more error than heart rate, even in the best study. Everything else on accuracy lives in the accuracy hub.