What Does Your Wearable’s Recovery Score Actually Tell You?

02From a signal to a score

There are several steps between a sensor and the number on a screen: recording a signal, processing it into a metric, comparing it with a reference, and combining selected information. Each step adds a choice. An overnight HRV estimate and a final recovery score are therefore different objects.

For HRV itself, recording conditions, the measure chosen and treatment of artefacts matter.^[2] A composite score introduces further decisions: which contributors count, how much history is used, and whether a low value in one contributor can be offset by another.

A 2025 review examined 14 composite health scores across 10 wearable manufacturers. It identified differences in time windows, weighting and proprietary methods, alongside limited transparency and validation evidence.^[1] That review is a dated examination of documented systems, not a claim that every product or later algorithm is identical.

03Two examples from manufacturer documentation

Oura describes Readiness using nine contributors, including resting heart rate, HRV balance, temperature, sleep-related measures and activity. Its HRV balance compares a recent average with a longer personal reference. The resulting number combines information across different timescales.

WHOOP’s recovery documentation separates the composite score from reported physiological measures and distinguishes a scored result from one with insufficient data. A missing score and a low score therefore convey different information.

These official documents explain product behaviour; independent studies address how well that behaviour answers a scientific question. The links below were consulted on 28 October 2026. Check the documentation for the device and software you use, because contributors and algorithms can change.

• Oura: Readiness contributors — manufacturer documentation ↗
• WHOOP: Recovery — manufacturer documentation ↗

04What does a score of 80 mean?

Consider a hypothetical recovery score of 80. Without its definition, the number alone does not establish that you are 80% recovered, that you can safely perform 80% of a training load, or that another brand’s 80 means the same thing.

Ask what the scale represents. Is it a comparison with your own recent history, a population reference, or a proprietary combination of contributors? Then ask which outcome it has been tested against: reported fatigue, performance on a particular task, or something else.

Personal history is useful context, but it is not automatically an ideal target. If your usual routine changes, the reference can change too. A reading close to your recent baseline answers a different question from whether your current routine is supporting the function you care about.

05Three different things a validation study can test

Measurement agreement: does the device estimate a signal accurately against an appropriate reference under relevant conditions? This is the first question, not the final one.

Meaning of the score: does the combined output relate to a clearly defined recovery outcome in the people and situations where it will be used? Accurate component measurements do not by themselves validate the combination.^[1]

Benefit of acting on it: do decisions guided by the score improve a meaningful outcome compared with another reasonable approach? That requires evidence about a decision strategy, rather than only a correlation between a score and how people feel.

This sequence is TAP’s way of organizing the questions to ask. It is not a brand ranking or a new grading system. Sport-recovery research and physical-resilience research also make the underlying issue clear: define the demand, the outcome and the observation period before deciding what recovery means.^[3][4]

06When the score and your experience disagree

Start by looking at what differs. You may feel mentally tired while a score is driven mainly by overnight physiological measures. Or a score may change when your subjective energy does not. Neither pattern establishes that one source is wrong: they may be sampling different aspects of the situation.

Inspect the contributors and recording conditions. Is the change concentrated in sleep, heart rate or HRV? Was the recording complete? Was the observation made in the usual way? Interpretation of HRV is particularly dependent on context.^[2]

Then consider what happened around the measurement and what you can actually do now. A diary of events and a relevant functional observation can add context. A score cannot identify the cause of a change from the number alone. Persistent symptoms or a marked loss of function deserve attention regardless of a reassuring device score.

07Make the score useful

Read the headline number once, then open its contributors. Identify what changed, how much usable data supports it, and whether the pattern persists across comparable observations. Pair that with one question you care about, such as how you feel after similar workdays or how a familiar training session goes.

The value of a score is the understanding it helps you build. For recovery, that means connecting a demand with the course of a particular response, while keeping the measurement, interpretation and decision distinct.

• Can HRV Tell Us How Well We Have Recovered? ↗
• Is Feeling Recovered the Same as Recovering Function? ↗
• Explore Measurement & Interpretation ↗