A finger cannot see the stage of sleep your brain is in. It can still tell you a useful story about how long you slept, how your heart behaved overnight, and whether you moved. Those two facts are not a contradiction. Treating them as the same thing is the mistake.
On August 20, 2026, a proposed class action was filed in San Francisco accusing Oura of advertising sleep-staging accuracy it cannot deliver. The complaint, Surber v. Oura, targets marketing that claimed 79% agreement with a clinical sleep lab and, more recently, 95% sleep staging accuracy. Oura disputes the allegations. This piece is not a verdict. It is a practical rule for anyone who has ever skipped a workout because REM looked low.
Keep the ring. Stop letting a derived score write the day.
What The Lawsuit Is Actually About
The filing does not claim heart rate, temperature, or movement tracking are fake. It argues that a ring without brain or eye sensors cannot measure sleep stages the way a hospital polysomnography lab does, and that advertising those stages as near-clinical accuracy misleads buyers who then make real decisions from the number.
That last part is the part that matters even if a court never agrees with the rest. People cancel lifting, panic about recovery, and rearrange their week around a pie chart. The hardware is not the problem. The authority we give the score is.
What A Finger Measures Well, And What It Infers
Sleep happens in the brain. Staging it in a lab means electrodes on the scalp and sensors on the eyes. A ring infers stages from pulse, motion, and temperature. Inference can still be useful. It is not the same as measurement.
A 2025 Scientific Reports comparison in a university sleep lab put that split in numbers. Against polysomnography, Oura's sleep-versus-wake distinction landed around 85% accuracy, and total sleep time was close at the group level, within about 12 minutes. Four-stage classification (wake, light, deep, REM) was 53.18%. The researchers also warned that decent averages can hide large errors on a given night, especially in a clinical patient cohort rather than a healthy consumer sample. That caveat belongs next to the headline, not after it.
In plain English: duration and “were you asleep” are in a different league from “how many minutes of REM.” Marketing that blurs those leagues is what the lawsuit is fighting over. You do not need a lawsuit to stop treating the weaker layer as orders.
| Signal | Treat it as | What it is good for |
|---|---|---|
| Sleep duration and timing | A sensor | Bedtime rhythm, sleep debt, consistency |
| Resting heart rate, HRV trend | A sensor | Load, illness onset, recovery direction over a week |
| Steps and movement | A sensor | Whether the day was actually sedentary |
| Light / deep / REM minutes | Beta | Curiosity. Not a reason to cancel strength day |
| Sleep Score / Readiness as a veto | Beta | One noisy input. Never the coach |
The Decision Test
If the score and how you feel disagree, the score does not automatically win. A green 92 after a wrecked night is a known failure mode. A red 58 when you slept, ate, and feel ready to train is the other one. Derived staging is one of the reasons those mismatches happen.
Use overnight heart rate and HRV as a weekly trend. Use duration to protect bedtime. Use the staging pie the way you use a weather app on a 10-day forecast: interesting, not binding.
This matters more if you are on a GLP-1. A low sleep score should not drop your protein floor or skip the lift that keeps the weight loss from eating muscle. The plan has higher-signal inputs than last night's inferred REM.
How Mallet Uses The Ring
Mallet does not claim to stage sleep better than Oura. That would be the same overreach in a different logo. Wearable sleep still feeds our readiness number. The difference is what that number is allowed to do.
Sleep is one factor next to HRV, resting heart rate, training you actually did, and a check-in for how you feel. The weekly plan still lives in protein, lifting, labs, and medication context. Readiness is a gauge. It is not a coach that gets to cancel the week because a finger guessed your sleep architecture.
Keep the wearable. Give the score a demotion. Mallet pulls sleep, heart, and movement as sensors, then sits them next to how you feel, training, and labs so a pie chart cannot run the day. Get early access →
If the score and your body already argue, read who gets the vote when they disagree. If you want the hardware to stay and the brain to upgrade, the Oura alternative is usually not a new ring.
Selected References
- Herberger S, et al. Performance of wearable finger ring trackers for diagnostic sleep measurement in the clinical context. Scientific Reports. 2025.
- Surber v. Oura Inc., No. 3:26-cv-08686 (N.D. Cal. filed Aug. 20, 2026). Proposed class action. Oura disputes the allegations.
This article is for education, not legal or medical advice. The lawsuit is pending. Wearable scores are not a diagnosis and should not replace clinical sleep testing when it is indicated.