Wearables and Sleep Tracking: Accuracy and Limits

What this page covers

Smartwatches, rings and bands estimate sleep, heart rate and "readiness" scores. People use them hoping to protect long-term health. This page covers what published reviews say about accuracy, and the limits. It owns the sleep-tracking and longevity angle; the broader device guide is Wearables and Health Trackers. Evidence labels: promising but limited, mixed, insufficient evidence, marketing claim. Education only; this page does not recommend any device.

Sleep tracking

The reference method for measuring sleep is polysomnography (PSG), an overnight lab study that records brain waves, breathing and more. Consumer devices estimate sleep from movement, heart rate and similar signals.

• Fitbit-type bands. A 2019 review found 22 studies, 8 with data for meta-analysis. Models without sleep staging tended to overestimate total sleep time (by roughly 7 to 67 minutes) and underestimate time awake after falling asleep compared with PSG. Newer models using heart rate variability and movement performed better than early ones. [1]
• Several brands. A 2024 review of eight studies found the Fitbit Charge 4 and Garmin Vivosmart 4 showed moderate accuracy for sleep stages and total sleep time. WHOOP disagreed most on REM sleep (21 minutes). [2]
• Smart rings. A 2025 meta-analysis of six studies (388 people) found no statistically significant average difference between the Oura Ring and PSG or actigraphy for total sleep time, efficiency, wake time or sleep stages. [3] Average agreement does not guarantee accuracy for each person or night, and the pooled sample was small.

Label: mixed. Total sleep time is usually the most dependable number; stage breakdowns (deep, REM) are less certain. No source we read showed that improving a tracker's sleep score extends life.

Heart rate variability (HRV) and heart rate

HRV is the small variation in time between heartbeats. A meta-analysis of 23 studies found portable devices differed from ECG by a small average amount, but results were highly heterogeneous and depended on the HRV metric, body position and sex. [4] For the Apple Watch, a 2025 meta-analysis of 56 studies found heart rate and step counts generally accurate, but energy expenditure (calories burned) was not: all subgroups exceeded the 10% error threshold. [5]

Label: promising but limited for following your own trend; insufficient evidence that HRV or "readiness" scores guide long-term health decisions.

When tracking backfires

Sleep clinicians have described "orthosomnia": people seeking treatment for self-diagnosed sleep problems because of light or restless sleep on a tracker. They noted that tracker data often feels more consistent with a person's experience of sleep than validated methods. [6] If numbers make you anxious or you sleep worse after checking, taking a break is reasonable.

How to evaluate a wearable claim

1. Ask which study compared the device with a reference method (PSG or ECG), in which people.
2. Treat stage and "recovery" scores as estimates; trends over weeks matter more than one night.
3. Be cautious about "biological age" or lifespan claims from a wrist device. Even lab-based aging biomarkers have barriers to clinical use, including a need for validation at the individual level. [7]
4. Remember that sleep basics have more support than tracker scores: see sleep hygiene and how sleep works.

More on tracking and recovery is in Sleep Optimization and Recovery, and a general testing overview is in Biomarkers and Biological Age Tests.