Sleep Tracking
Sleep trackers are good at one thing and much weaker at another. Research comparing consumer devices with clinical sleep studies finds they detect sleep versus wake reasonably well, but their sleep stage estimates agree with laboratory measurement far less closely than the apps suggest. Treat the totals as useful and the stage breakdown as a […]
Sleep trackers are good at one thing and much weaker at another. Research comparing consumer devices with clinical sleep studies finds they detect sleep versus wake reasonably well, but their sleep stage estimates agree with laboratory measurement far less closely than the apps suggest. Treat the totals as useful and the stage breakdown as a rough guess.
Your watch or ring cannot see your brain. A clinical sleep study can, and that difference explains almost everything about tracker accuracy.
The clinical standard is polysomnography. It records brain activity, eye movement, and muscle tone, and a trained scorer reads those signals to label each 30-second window of the night. Sleep stages are defined by brain waves, so measuring brain waves is how you identify them.
A consumer tracker works from what it can reach at the wrist or finger. That usually means movement, heart rate, and heart rate variability. Some devices add blood oxygen or skin temperature.
From those signals, the device runs an algorithm that estimates what your brain was probably doing. It is an informed guess built on patterns, not a measurement of sleep itself.
This matters because the guess is much easier for some questions than others.
Telling sleep from wake is the easier problem, and devices handle it reasonably well.
When you are asleep you move less and your heart rate settles into a different pattern. Those changes are visible from the wrist. Validation research generally finds consumer devices identify sleep periods with useful accuracy.
The measures worth paying attention to are the simple ones:
There is a known bias worth understanding. Because trackers lean on stillness, lying awake without moving can be scored as sleep. Devices have historically tended to overestimate how much you slept, particularly in people who sleep poorly.
That bias runs in an unhelpful direction. The people most likely to be checking a tracker are the ones whose sleep problems it is least likely to capture.
Sleep stages are the weak point. The deep sleep and REM numbers in your app are the least trustworthy figures on the screen.
Stages are defined by brain activity, and the device is inferring them from heart rate and movement. Agreement with laboratory scoring is noticeably lower for staging than for simple sleep and wake detection, and it varies by device, by night, and by person.
Two people wearing different brands on the same night can get different stage breakdowns. Neither is lying. They are running different algorithms on different signals.
Manufacturers also update their algorithms. A software release can change your deep sleep numbers without anything changing in your body, which is a useful reminder of what the figure actually represents.
A sleep score is not a measurement. It is a number the company builds by weighting several inputs, including estimates that are themselves uncertain.
Each brand weights differently, and none of them publish the full formula. A score of 82 means something specific to that company and nothing standardised across the industry.
It is reasonable as a rough daily summary. It is not a health metric, and it is not comparable between devices.
The pattern across validation studies is consistent enough to summarise plainly.
Researchers test consumer devices by having people wear them during a proper polysomnography study, then comparing the two records. A widely cited 2021 study in the journal SLEEP by Chinoy and colleagues tested seven consumer sleep-tracking devices against polysomnography in healthy adults. Its general finding matched earlier work: the devices performed acceptably at detecting sleep versus wake, and noticeably worse at classifying sleep stages.
Reviews of wearable sleep technology report the same shape of result across many devices and studies.
Three limitations are worth holding onto when you read any of this:
The professional position is clearer than the marketing. The American Academy of Sleep Medicine has published a position statement on consumer sleep technology stating that these devices are not substitutes for medical evaluation and should not be used to diagnose or treat sleep disorders. They can support a conversation with a clinician. They do not replace one.
There is a documented pattern where the effort to improve a sleep score makes sleep worse.
Clinicians named it orthosomnia after seeing patients whose sleep anxiety was driven by their tracker data. A 2017 case report in the Journal of Clinical Sleep Medicine by Baron and colleagues described people who became preoccupied with achieving perfect sleep data, and whose worry about the numbers appeared to be perpetuating the problem.
The mechanism is not mysterious. Sleep does not respond well to effort. Anxiety about sleeping is one of the things that keeps people awake, and a nightly score gives that anxiety something concrete to attach to.
Some signs the tracker is not helping you:
If several of those apply, taking the device off for a few weeks is a reasonable thing to try. How you feel during the day is the more meaningful measure anyway.
The device is more useful as a pattern detector than as a nightly verdict.
A paper sleep diary is worth mentioning here. Clinicians still ask for them because your own record of when you went to bed, how long it felt like it took, and how you felt the next day captures things no wrist sensor can.
A tracker can prompt a conversation. It cannot conclude one. Contact a healthcare professional if any of the following apply.
Bring a two-week sleep diary rather than months of app screenshots. It is more useful to a clinician, and it is what many of them will ask for.
There is no settled answer, and it changes as firmware updates. Studies that compare several devices tend to find they cluster together: broadly acceptable at sleep versus wake, weaker at stages. Picking based on stage accuracy is choosing on the least reliable measure.
Not automatically. A finger can give a cleaner heart rate signal than a wrist, but accuracy depends on the algorithm as much as the sensor. Both face the same core limit, which is that neither measures brain activity.
Because stillness looks like sleep to a movement-based algorithm. This is a known bias, and it is more pronounced in people who sleep badly. If your device regularly disagrees with your experience, your experience is the better guide.
No. Some devices flag possible breathing disturbances, and that is worth acting on, but a flag is a prompt to see a doctor rather than a diagnosis. Sleep apnea is diagnosed through clinical testing.
Chasing a deep sleep target from a consumer device is not a useful project, because that figure is the least accurate number your tracker produces. If you feel persistently unrested, that is worth discussing with a doctor regardless of what the app reports.
Only you can judge that. If it helps you notice that your bedtime has drifted an hour later over a month, it is doing something useful. If it is a source of nightly anxiety, that cost is real and worth taking seriously.
This article sits in our section on sleep tracking and measurement. For the wider picture, start with the main sleep guide, which explains the two systems behind sleepiness and where to go next. If tracking anxiety is the issue, our section on falling asleep and staying asleep covers the behavioural approaches with the strongest evidence behind them. How we research and correct our work is set out in the editorial policy.
Medical disclaimer
This article is for general information only. It is not medical advice, and it is not a substitute for diagnosis or treatment from a qualified healthcare professional. Do not start, stop, or change any medication, supplement, or treatment based on what you read here. If you have symptoms that concern you, contact a doctor. If your symptoms are severe or sudden, seek emergency care.
Every claim in this article links to its source in the text above. We use peer-reviewed research and high-authority health bodies, and we explain our sourcing standards in the editorial policy.
Medical disclaimer
This article is for general information only. It is not medical advice, and it is not a substitute for diagnosis or treatment from a qualified healthcare professional. Do not start, stop, or change any medication, supplement, or treatment based on what you read here. If you have symptoms that concern you, contact a doctor. If your symptoms are severe or sudden, seek emergency care.
Sleep Tracking
A breathing or blood oxygen alert from a watch or ring is not a diagnosis. It is a signal that a pattern in your data resembled something worth checking. Repeated fl...
July 27, 2026
Sleep Tracking
A sleep diary is a short daily record of when you slept and how you felt. Two weeks is the usual length, and it takes about a minute each morning. Sleep clinicians s...
July 27, 2026
Sleep Tracking
Orthosomnia is the term clinicians use when the pursuit of perfect sleep data starts interfering with sleep itself. It was named in a 2017 case report describing pat...
July 27, 2026