What Sleep Trackers Can and Can't Tell You

Your watch says you got 41 minutes of deep sleep, your sleep score is 68 and your “readiness” is low. You feel fine. Should you believe it?

Sleep trackers have gone mainstream on wrists, fingers and nightstands, and they produce a lot of confident-looking numbers. Some of those numbers are reasonably solid. Others are educated guesses. This guide explains how trackers estimate sleep, which metrics you can lean on, which you should hold loosely, and when tracking itself becomes the problem.

How sleep trackers actually measure sleep

The gold standard for measuring sleep is polysomnography (PSG), the overnight study done in a sleep lab or, in simplified form, at home. It records brain waves (EEG), eye movements, muscle activity, breathing, oxygen levels and heart rhythm. Sleep stages are defined by brain activity, so only a test that measures the brain can score them directly.

Consumer trackers don’t measure your brain. Instead, they infer sleep from signals that tend to change when you sleep:

• Movement, from an accelerometer. Lying still suggests sleep.
• Heart rate and heart rate variability, from optical sensors (the green or red lights on the back of a watch or inside a ring).
• Skin temperature on many rings and some watches.
• Blood oxygen estimates on some devices.
• Breathing patterns, estimated from movement, heart signals or, in bedside devices, radar or sensors under the mattress.

An algorithm combines those signals and labels each 30-second chunk of the night as awake, light, deep or REM sleep. That’s an estimate based on patterns learned from people who wore the device during lab studies.

What trackers are good at

Research comparing consumer devices with PSG shows a consistent pattern. In a lab study published in the journal Sleep in 2021, researchers tested 7 consumer sleep trackers, including wearables from Fitbit and Garmin and bedside devices, against PSG. Every device correctly identified the vast majority of sleep periods; sensitivity for sleep was at least 93%. A 2025 study of 6 wrist-worn devices found the same thing: all caught more than 90% of sleep.

That makes trackers useful for:

• Total sleep time, give or take
• Bedtime and wake time
• Consistency: whether your schedule shifts on weekends
• Long-term trends: whether you’re sleeping more or less than last month
• Resting heart rate and heart rate variability trends, which can shift with illness, alcohol, stress or training load

Key takeaway: Trust your tracker for how long and how regularly you sleep. Be skeptical about exactly what happened inside each night.

What trackers struggle with

Time awake in bed

The same research shows trackers are much weaker at detecting wakefulness. In the 2021 Sleep study, specificity, the ability to correctly flag wake periods, ranged from roughly 18% to 54% depending on the device. The 2025 study found similar results. If you lie still but awake, many trackers will count that as sleep. Devices also did worse on nights when sleep was disrupted, which is exactly when you most want accurate data.

This matters most for people with insomnia, who often lie awake quietly. A tracker may report a decent night that didn’t feel like one.

Sleep stages

Light, deep and REM breakdowns are the most eye-catching part of most sleep apps, and the least reliable. The 2021 study described device sleep stage assessments as inconsistent. Without measuring brain waves, a device is estimating stages from heart rate, movement and temperature. Night-to-night swings in “deep sleep” often say more about the algorithm than about you.

Sleep scores and readiness scores

Scores combine several estimates into one number using each company’s own formula. They’re not standardized, and a 75 on one brand doesn’t equal a 75 on another. Use them as a nudge, not a grade.

Sleep disorders

The American Academy of Sleep Medicine (AASM) said in a 2018 position statement that consumer sleep technology can’t be used to diagnose or treat sleep disorders, though data can support a conversation with a clinician as part of a proper evaluation. That remains true for most features today.

There’s one notable exception: some watches now have FDA-authorized features that flag possible sleep apnea. In 2024, Samsung received FDA De Novo authorization for a sleep apnea feature on Galaxy Watch, and Apple introduced FDA-authorized sleep apnea notifications on recent Apple Watch models. Apple’s feature uses the accelerometer to track breathing disturbances over 30-day periods and requires at least 10 nights of wear. Apple states the watch isn’t intended to diagnose sleep apnea, and not everyone with sleep apnea will get a notification. These features screen; they don’t replace a sleep study.

Reliability at a glance

When sleep tracking backfires

Sleep researchers have a name for an unhealthy fixation on tracker data: orthosomnia. In a 2017 case series, researchers led by Kelly Glazer Baron described patients whose pursuit of perfect tracker numbers increased their anxiety about sleep and complicated insomnia treatment. In one case they described, a patient felt unrefreshed because of her device’s report, even though a lab study showed she had plenty of deep sleep.

Warning signs that tracking is hurting more than helping:

• You check your score first thing and let it set your mood for the day.
• You feel more tired after seeing a “bad” number, even if you felt fine before.
• You spend extra time in bed trying to raise your score.
• You worry about sleep stages before falling asleep.

If that sounds familiar, try hiding the score, checking only weekly averages, or taking a break from wearing the device at night. If poor sleep persists, cognitive behavioral therapy for insomnia (CBT-I) from a qualified clinician is a better next step than a new gadget.

How to use a sleep tracker well

1. Wear it consistently. Same device, same wrist or finger, every night, so trends are comparable.
2. Look at weekly and monthly averages, not single nights.
3. Use it to test one change at a time, such as an earlier bedtime, less late caffeine or no alcohol before bed, and watch the trend for 2 to 3 weeks.
4. Pair data with how you feel. A short note on energy and mood each morning keeps the numbers honest.
5. Focus on sleep duration and regularity. The CDC says adults aged 18 to 60 should get at least 7 hours of sleep a night, and that quality matters too.
6. Set a limit. Check the app once a day at most.
7. Bring trends, not screenshots, to your doctor. A month of sleep-time and heart rate data is more useful than one alarming night.

Shopping for a device? Our comparisons of smart rings vs smartwatches for health tracking and smartwatches vs fitness trackers cover comfort, battery life and subscriptions.

When to see a clinician instead

A tracker can’t rule out a sleep disorder. Talk with a doctor if you have:

• Loud snoring, choking or gasping, or pauses in breathing that someone has noticed
• Ongoing daytime sleepiness, especially while driving
• Trouble falling or staying asleep most nights for several weeks
• Unusual movements or behaviors during sleep
• A sleep apnea notification from an FDA-authorized feature

Your doctor may recommend a home sleep apnea test or an in-lab sleep study. Many sleep clinics now offer telehealth visits; see our overview of telehealth and remote patient monitoring. If you’re curious how other wearable health features hold up, read how accurate smartwatch ECG readings are.

The bottom line

Sleep trackers are useful mirrors for your habits: how long you sleep, how regularly and how your body’s overnight signals trend over time. They’re unreliable narrators of what happens inside each night, especially time awake and sleep stages, and most can’t diagnose anything. Use the trends, ignore the nightly drama and let how you feel during the day have the final word.

This is general information, not medical advice.

Sources

• American Academy of Sleep Medicine: Consumer sleep technology position statement — trackers can’t diagnose or treat sleep disorders; role in clinical conversations
• Chinoy et al., Performance of seven consumer sleep-tracking devices compared with polysomnography, Sleep (2021) — high sleep sensitivity, low wake specificity, inconsistent staging
• Performance validation of six commercial wrist-worn sleep-tracking devices compared to polysomnography (2025) — more than 90% sleep sensitivity, lower specificity
• Apple Support: Sleep apnea notifications on your Apple Watch — how the feature works, 10-night requirement, not a diagnostic device
• Samsung: Sleep apnea feature on Galaxy Watch authorized by US FDA — 2024 FDA De Novo authorization
• CDC: About sleep — at least 7 hours for adults 18 to 60, and the importance of sleep quality