Recovery Scores Explained

What a recovery score is

A recovery score is a single number or color that a wearable's software builds from overnight body signals. It is a proprietary score: the company decides which inputs to use, how to weigh them and how to compare them with your own past nights. There is no shared standard, so one brand's 70 is not another brand's 70, and neither is a medical result.

Measured, estimated, calculated

WHOOP's community help text says its Recovery is "calculated based on your heart rate variability (HRV), resting heart rate (RHR), and sleep performance", and describes it as "how well your body has recovered from the previous day's strain". A higher (green) score suggests you are well rested; a lower (yellow or red) score "may indicate fatigue, stress, or poor sleep". [1] Weights and the exact formula are not given on that page. A WHOOP-authored research paper describes the recovery score as a composite that integrates sleep and biometric data, reported as a percentage. [2]

Most such scores lean on HRV. A qualitative study of regular exercisers noted that consumer "recovery" or "readiness" scores are computed primarily from HRV. [3] For HRV itself, see Heart-Rate Variability Explained. For another brand's version, see Readiness Scores Explained.

Units and conditions

Scores have no physical unit and are brand-specific (a percentage, a number or a color band). The inputs are collected mainly during sleep, so a night without the device on gives the algorithm little or no data. Wear instructions differ by brand and model; read the brand's own guide.

How accurate are the inputs?

One study compared five wearables with a reference ECG over 536 nights in 13 healthy adults. Accuracy for overnight resting heart rate and HRV "varied significantly" by device. For HRV, WHOOP 4.0 showed moderate accuracy (concordance 0.94, mean absolute percentage error 8.17%) and Oura Gen 4 higher (0.99, 5.96%). [4] That study tested the inputs, not the final scores, and it was small and short-term.

A separate study of the Oura ring against ECG found HRV accuracy improved when poor-quality segments were filtered and results were averaged over longer windows, and more than half of older participants still exceeded 10% error. [5] Do not assume a device with accurate inputs produces a meaningful score.

Does the score predict anything?

• Performance. In 35,140 nights from 389 professional golfers, higher recovery scores were associated with better play, but the paper came from WHOOP's own performance science team and shows an association, not that acting on the score improves results. [2]
• Training guided by HRV. A meta-analysis of 8 studies (198 participants) found small, non-significant effects on performance and VO2 peak when endurance training was guided by HRV compared with a fixed plan. [6] See HRV Trends for Recovery.
• Individual patterns. In 20 elite endurance athletes monitored for a year, personalized analyses were more accurate than group averages, and individual variability was large. [7]
• Independent validation of the score itself. Proper evidence not found (PubMed search 2026-10-22): no independent study read here shows that a brand's recovery score predicts injury, illness or better outcomes for ordinary users.

Label: insufficient evidence for the score as a decision tool; the HRV inputs are mixed and device-specific.

Normal variability

The athlete study above found large differences between individuals, so there is no single normal range. [7] Compare yourself only with your own trend on the same device, and treat one night as one night.

Appropriate uses

• A prompt to notice sleep, stress, alcohol, illness and training load, then compare with how you feel. A study of 17 regular exercisers found they valued self-awareness and personal judgment over the number. [3]
• Optional, low-stakes personal logging.

What it cannot do

It cannot diagnose illness, overtraining or heart conditions, and it does not tell you when to see a doctor. Some users change non-exercise behavior to raise the score itself. [3] If checking the number raises worry, see Orthosomnia and Sleep-Tracking Anxiety. Whether the score changes long-term adaptation is covered in Recovery Claims vs Long-Term Adaptation.

When professional assessment may be appropriate

Persistent unusual tiredness, new breathlessness, chest discomfort, fainting or an irregular heartbeat need a clinician's assessment whatever the app says. If severe or sudden, seek urgent care; in the US call 911. People with eating-disorder history should be careful with any tracker; see Eating Disorders and Disordered Eating.