Are smart rings accurate for sleep?

How a ring estimates your sleep

Inside a smart ring, a motion sensor notices when you stop moving, an optical sensor reads your pulse, and a thermometer follows your skin temperature. Software trained on sleep-lab recordings turns these signals into sleep and wake, and into light, deep and REM sleep. The lab itself defines those stages from brain waves, eye movements and muscle activity, which a ring cannot measure, so every stage on a ring is an estimate.

The test it is measured against

Polysomnography records brain waves, eye and muscle activity, heart rhythm and breathing through the night, and a trained scorer marks every 30-second period as wake or one of the sleep stages. Validation studies compare a ring with that record, night by night and period by period. A cheaper reference at home is actigraphy: a medical-grade movement sensor worn on the wrist.

What the studies found

• 2019, one night in a lab. De Zambotti and colleagues compared the Oura ring with polysomnography in 41 healthy adolescents and young adults. Total sleep time, time to fall asleep and time awake after falling asleep were not different from the lab’s. The ring missed about 20 minutes of the deepest sleep and added about 17 minutes of REM sleep. It detected sleep 96% of the time, but wake only 48% of the time, and agreed with the lab on stages 51% to 65% of the time.
• 2020, a week at home. Asgari Mehrabadi and colleagues tracked 45 healthy adults for seven days with an Oura ring, a Samsung smartwatch and a medical-grade actigraph. The ring’s total sleep time correlated strongly with the actigraph’s, and its differences stayed within satisfactory ranges; the watch’s differences were considerably larger.
• 2024, several nights with a newer ring. Svensson and colleagues tested the third-generation Oura ring with an updated algorithm against up to three nights of polysomnography at home in 96 healthy adults aged 20 to 70 in Japan, over 421,045 thirty-second periods. Total sleep time, time to fall asleep, time awake after falling asleep and time in light and deep sleep did not differ significantly from the lab’s. The ring underestimated REM sleep by four to six minutes. It classed periods as sleep or wake with about 92% accuracy, detecting about 94% of sleep and about 73% to 75% of wake.

What that means for you

• Good at: how long you slept, when you went to sleep, and how both change over weeks.
• Weaker at: brief awakenings, which look like quiet sleep to a motion sensor, and the exact split between stages on a given night.
• Healthy people only: these studies did not test rings as a way to find sleep disorders. If you snore loudly, stop breathing at night or are always tired, a doctor and a sleep study are the way to find out.
• Results belong to a version: the 2024 study tested one ring generation with one algorithm. Updates can change the numbers, and in our search of the research literature most independent studies concern one brand, so other rings are less tested.

Why our first ring skips sleep

The first Lychnos Ring does not track sleep: its battery goes to conversation. Sleep and other wellness trends are planned for a second model, as wellness insights, not medical measurements. It is in development: why we chose that order.

Sources, read 4 November 2026: de Zambotti and colleagues, the sleep of the ring: comparison of the ŌURA sleep tracker against polysomnography, Behavioral Sleep Medicine, 2019; Asgari Mehrabadi and colleagues, sleep tracking of a commercially available smart ring and smartwatch against medical-grade actigraphy in everyday settings, JMIR mHealth and uHealth, 2020; Svensson and colleagues, validity and reliability of the Oura Ring Generation 3 with Oura sleep staging algorithm 2.0, Sleep Medicine, 2024. General information, not medical advice.