What a recovery score actually tells you

The scores are not comparable

Oura's readiness, WHOOP's recovery and Garmin's body battery are different calculations over different inputs with different weightings. A 70 on one is not a 70 on another, and there is no conversion.

They are also relative to the individual by design. Most of these scores already compare an athlete against their own recent history, which is why an elite athlete and a beginner can both sit at 85. That is a sensible choice by the manufacturers and it is also why ranking a squad by recovery score is meaningless.

So the score is genuinely useful for the person wearing the device, tracking themselves over time. It travels badly the moment you try to use it across people.

What a score hides

Which input moved

A drop from 85 to 60 could be short sleep, suppressed HRV, elevated resting heart rate, or all three. The coaching response differs in each case, and the composite number does not tell you which.

Whether the change is large for this athlete

Some athletes are naturally stable and some swing widely night to night. A fifteen-point drop is a real signal for the first and ordinary noise for the second.

Why it happened

Nothing in the sensor data knows about the argument at work, the flight, or the fact that yesterday's session was much harder than planned. The athlete knows all three.

Whether the data is even there

Some devices will still produce a score from a partial night. A number generated from thin data looks exactly like a number generated from good data.

How to read one as a coach

Treat it as a prompt, not a verdict

A low score is a reason to look at the components and ask a question. It is not an instruction to change the session, and athletes who treat it as one tend to train worse.

Look underneath it

Sleep duration, HRV and resting heart rate individually tell you more than the composite, because you can see which one moved.

Weight the trend

Three days drifting downward is far more informative than one bad morning, which is very often just a late meal or a disturbed night.

Ask the athlete

Subjective context resolves most flags in under a minute and carries information no sensor has.

A note on what these numbers are not

Frequently asked

Should I compare recovery scores between athletes?

No. Most of these scores are already normalised to the individual, so a cross-athlete comparison mostly measures physiology and device differences rather than who needs attention. Compare each athlete to their own history instead.

Which wearable has the most accurate recovery score?

The question does not have a good answer, because there is no ground truth to be accurate against — recovery is not a directly measurable quantity. What you can assess is whether a device measures consistently enough to make its own trend meaningful.

My athlete's score is low but they feel fine. Who is right?

Both are information and the disagreement is itself worth noting. Physiology sometimes moves before an athlete notices, and equally a score can be depressed for reasons that have nothing to do with training. Treat it as a prompt for a conversation.