The Apple Watch sleep score is a 0 to 100 number built from three inputs, and Apple publishes exactly how much each one is worth. Sleep duration carries 50 points, bedtime consistency 30, and interruptions 20. That transparency is unusual in this category, where most brands describe their inputs and keep the weightings private, and it changes how you should read the number.
The Verdict
What the score is built from
Apple's own documentation sets out the three components and their point values, which makes the score reverse-engineerable in a way most competitors' are not.
| Component | Worth | What it actually counts |
|---|---|---|
| Sleep duration | 50 points | How long Apple estimates you actually slept, not how long you were in bed. Half the score rides on this one input. |
| Bedtime consistency | 30 points | When you fell asleep, judged against your own recent nights rather than a fixed clock time. |
| Interruptions | 20 points | How often you woke and how long you stayed awake once you did. |
Two details matter more than they look. Duration is estimated asleep time, not time in bed, so lying awake reading costs you twice by shortening sleep and adding an interruption. And consistency is judged against your own recent nights rather than a clock, which means the score is not biased against people who genuinely sleep late. It is biased against people whose bedtime moves.
The Apple sleep score bands
| Score | Apple classification | What usually produced it |
|---|---|---|
| 96–100 | Very High | Long sleep, a bedtime in line with your recent pattern, and few awakenings. |
| 81–95 | High | The realistic target band for most adults on most nights. |
| 61–80 | OK | Usually a short night or a bedtime well outside your usual window. |
| 41–60 | Low | Two of the three components are down, most often duration plus interruptions. |
| 0–40 | Very Low | A badly fragmented or very short night, or an incomplete recording. |
The bands are Apple's conventions, not clinical thresholds, and they are not comparable to a score from another brand. A 78 here and a 78 on a different platform describe different nights.
How Apple's score compares to other wearables
Platforms below are listed alphabetically. The point is not which score is best but that each brand is measuring a different thing under the same word.
| Platform | What feeds the score | How it differs |
|---|---|---|
| Apple | Duration, bedtime consistency, interruptions | Publishes the exact point weighting. No HRV or resting heart rate input. |
| Fitbit | Duration, sleep stages, restoration signals | Publishes the contributors but not point weightings; folds resting heart rate and restlessness into the score. |
| Garmin | Duration, stages, stress and heart-rate data | Blends recovery-oriented inputs into the sleep number. |
| Oura | Duration, efficiency, latency, timing, heart rate, HRV | Uses more physiological inputs than Apple and separates the sleep score from a readiness score. |
| Whoop | Sleep need versus sleep achieved | Expresses the idea as Sleep Performance, a percentage of calculated need rather than a 0 to 100 composite. |
Because Apple excludes physiological recovery signals and Oura includes them, the same night can score in the 90s on one and the 70s on the other without either being wrong. Our guide to what any sleep score measures covers the general problem, and Oura vs Whoop compares two of the alternatives in detail.
Why the score goes missing or reads "no data"
A missing score is almost always a recording problem rather than a bug. Work through these in order.
- Wear time. The watch has to be on your wrist for the bulk of the sleep period. Taking it off at 4 a.m. usually voids the night.
- Battery. A watch that dies at 3 a.m. produces a partial recording that cannot be scored. Charging during the evening wind-down rather than overnight fixes this permanently.
- Sleep schedule or Sleep Focus. Tracking needs a schedule set up in the Health app. Without one, sleep may be logged inconsistently.
- Software version. The score is a watchOS 26 feature, paired with an iPhone running iOS 26. watchOS 26 runs on Apple Watch Series 6 and later, Apple Watch SE (2nd generation) and later, and every Ultra model, so the feature is not restricted to the newest hardware. The classification bands were revised in a later 26.x update, which is why band labels can differ between two watches on the same major version. Apple's own sleep score documentation is the reference.
- Time zones. A night spanning a zone change can be split or logged against the wrong day. It usually resolves itself within a night or two.
- Very short sleep. A nap-length recording is stored as sleep but is not scored as a night.
How accurate is it, really
The accuracy question splits cleanly along the point weighting, which is the useful way to think about it. Total sleep time and awakenings, worth 70 of the 100 points, are the measurements wrist devices do best, because both are inferred largely from movement and heart rate patterns that a wrist sensor genuinely detects. Sleep stage classification, which every consumer device struggles with against laboratory polysomnography, is deliberately excluded from the score. Apple's decision to leave stages out therefore makes the composite more defensible than a score that folds them in, even though it makes the score less physiologically interesting.
The blind spot is what the sensors never see. A person with untreated obstructive sleep apnea can post consistent High scores by lying still, sleeping long enough, and waking only briefly at each event, because those arousals are too short for a wrist algorithm to catch reliably. A good score is not a clean bill of sleep health.
How to raise the number, in order of effect
- Add sleep opportunity. With 50 points on duration, 30 extra minutes in bed moves the score more than any other single change.
- Fix the bedtime, not the wake time. Consistency is scored on when you fell asleep. A variable bedtime is the most common reason an otherwise good sleeper sits in the 70s.
- Attack interruptions with the obvious causes first. Alcohol within three hours of bed, a bedroom above 70°F, and late fluid intake account for most avoidable awakenings.
- Charge the watch in the evening. A dead battery costs the whole night, which no amount of sleep hygiene recovers.
- Ignore the number for 60 minutes after waking. Reading a poor score first thing measurably worsens how people rate their own day, a pattern documented as orthosomnia.
When a low score is a medical question
Take these to a clinician rather than to the app:
- Persistent unrefreshing sleep despite scores in the High band, especially with snoring, witnessed pauses in breathing, or morning headaches.
- Interruption counts that climb steadily over weeks without a change in habits.
- Regular sleep onset taking more than 30 minutes for over a month.
- Daytime sleepiness severe enough to affect driving, at any score.
Frequently Asked Questions
How is the Apple Watch sleep score calculated?
Apple builds the score from three inputs with published weightings: sleep duration is worth 50 points, bedtime consistency 30 points, and interruptions 20 points. Duration is the estimated time you were actually asleep rather than time in bed. Consistency compares when you fell asleep against your own recent nights, so a steady 1 a.m. sleeper is not penalised for the hour. Interruptions counts both how often you woke and how long each awakening lasted. Sleep stages are recorded and shown separately, but they do not feed the score.
What is a good Apple Watch sleep score?
Apple classifies 81 to 95 as High and 96 to 100 as Very High, so anything in the 80s or above is a good night by Apple's own definition. Below that, 61 to 80 is OK, 41 to 60 is Low, and 0 to 40 is Very Low. Most adults with a settled schedule and adequate time in bed land in the 80s regularly and touch the 90s occasionally. Chasing 100 is not a useful goal, because the last few points usually require a bedtime precision that ordinary life does not allow.
Why is my Apple Watch sleep score not showing?
The four common causes, in the order they occur: the watch was not worn for enough of the night, the battery ran out before morning, Sleep Focus or a sleep schedule was never set up, or the software is too old. On that last one, the score requires watchOS 26 on the watch and iOS 26 on the paired iPhone; watchOS 26 supports Apple Watch Series 6 and later, SE (2nd generation) and later, and all Ultra models, so older hardware is rarely the blocker. A fifth cause catches travellers, because a night that straddles a time-zone change can be recorded oddly or split. If the score is missing but stage data is present, the recording was too short or too fragmented for Apple to score rather than absent.
Is the Apple Watch sleep score accurate?
It is accurate for what it actually measures and weaker at what it infers. Total sleep duration and the number of awakenings, which together carry 70 of the 100 points, are the two things wrist wearables measure most reliably against laboratory polysomnography. Sleep stage estimates from any wrist device are much less reliable, which is a reason to prefer Apple's score over stage percentages rather than a criticism of it. The score cannot detect what the sensors cannot see, and untreated sleep apnea is the important example.
Can I get a sleep score on iPhone without an Apple Watch?
Not from the phone alone. The score is calculated in the Health app from overnight sleep data that some device recorded, so something has to be worn while you sleep. An Apple Watch is the usual source, and a third-party tracker that writes detailed sleep data to Health can also supply it. What an iPhone cannot do on its own is measure the inputs: a sleep schedule set on the phone tells the system when to expect sleep, but it does not record how long you were actually asleep or how often you woke.
Does the Apple Watch sleep score include naps?
Naps are recorded as sleep but the daily score is built around the main overnight sleep period. A long afternoon nap therefore improves how you feel without improving the number, and it can lower the following night's score indirectly by pushing your bedtime later and costing consistency points. Shift workers and new parents see this pattern most.
Why did my sleep score drop after a time-zone change?
Because consistency is scored against your own recent bedtimes, and flying moves your clock without moving your history. A five-hour shift makes tonight's bedtime look several hours out of pattern, so up to 30 points come off for a reason that has nothing to do with sleep quality. Expect two to four nights of artificially low scores after long-haul travel, and read duration and interruptions instead during that window.
Should I act on a low sleep score?
Act on a run of them, not on one. A single low number after a late dinner, a drink, or a hard evening session is explained by the obvious cause and needs no response. A week of scores sitting well below your own average, especially when interruptions rather than duration are driving it, is worth investigating. If scores look fine and you still feel unrefreshed for several weeks, believe how you feel and raise it with a clinician rather than with the watch.