Apple Watch readiness score: how it works

Apple announced a Readiness app for Apple Watch. This Readiness app turn overnight measurements into an estimate of how recovered you are. The basic concept is that you’d work out harder on days with a higher readiness score.

iOS 27’s new readiness screen shows a score of 7 (“Ready”), with fitness, overnight vitals, and sleep score listed under “Raising Your Score.”

Apps like Whoop, Athlytic, and Bevel made readiness scores popular. Most wearable readiness scores use the same inputs, like heart rate variability, resting heart rate, sleep, and other signals, compared to your personal baseline. Let z be how far last night’s reading was from your previous 30-night average, divided by its standard deviation (your usual fluctuation). Then an example readiness score would be:

$$\text{Readiness} = 50 + 10z_{\text{HRV}} - 5z_{\text{resting HR}} + 5z_{\text{sleep}}$$

(This is an example, not a validated formula or a reconstruction of Apple’s exact algorithm.)

Which Apple Watch models does the Readiness app work on?

The Readiness app is exclusive to Apple Watch Series 12 and Apple Watch Ultra 4, according to Apple’s model comparison. It isn’t available on older Apple Watch models or Apple Watch SE.

Are readiness scores actually validated?

Studies that evaluate the accuracy of readiness scores are unfortunately pretty thin. One study of 389 male professional golfers analyzed 35,140 nights of wearable data across 521 events. A 10-point higher average WHOOP Recovery score was associated with 0.48 fewer strokes per round between golfers; within the same golfer, a season with a 10-point higher average Recovery was associated with 0.24 fewer strokes. Both associations were statistically significant.

Trials of HRV-guided training (the basis of commercial readiness scores, but no one company’s proprietary score) provide some encouragement. In 40 runners, HRV guidance produced a small advantage with fewer hard sessions. But then meta-analysis of eight studies found no statistically significant pooled advantage for performance or peak oxygen uptake. So different studies tell us different things.

So take readiness scores with a grain of salt.

How is Apple’s readiness score different than Whoop, Bevel, and Athlytic?

Devices like Whoop and Apple Watch apps like Bevel and Athlytic already calculate readiness scores. Apple’s readiness score looks broadly similar to these. Apple’s presentation showed recent fitness activity alongside overnight vitals and sleep score. Apple hasn’t yet revealed the formula for their recovery score, so we’ll update this post as more information becomes available.

Appendix: known readiness formulas and calculation details

The documentation below doesn’t disclose a complete formula for any of these four commercial scores. It identifies inputs, time windows, and some intermediate calculations. Numerical weights and the final mapping to a score remain undisclosed in these sources. A 2025 review of 14 wearable scores also identified gaps in algorithm transparency.

Score	Published calculation details	What’s missing for reproduction
WHOOP Recovery	HRV and resting heart rate are the biggest contributors. HRV is compared with a 30-day baseline. Other inputs include sleep, breathing rate, temperature, blood oxygen, and cycle context where applicable.	Numerical weights, input transformations, and final scoring function.
Oura Readiness	Nine contributors: resting heart rate, HRV balance, temperature, recovery index, sleep, sleep balance, sleep regularity, previous-day activity, and activity balance. HRV balance compares a weighted 14-day average with a three-month average.	Daily averaging weights, contributor-to-points mappings, and final combination.
Garmin Training Readiness	Combines last night’s sleep score, recovery time, HRV status, acute load, three-night sleep history, and three-day stress history. Updates throughout the day.	Weights, interactions, and conversion to the 1–100 score. This isn’t Garmin’s separate Body Battery metric.
Fitbit / Pixel Watch readiness	Combines HRV, resting heart rate, and recent sleep against a personal baseline. Initial calibration requires seven nights. Resting heart rate replaced the older activity component.	Weights, normalization, and final scoring function.

Published formulas for the HRV input

These standard formulas calculate an input to readiness, not readiness itself. For N normal heartbeat intervals, let NNᵢ be interval i and mean(NN) their average:

$$\mathrm{RMSSD}=\sqrt{\frac{1}{N-1}\sum_{i=1}^{N-1}(NN_{i+1}-NN_i)^2}$$

RMSSD is used by WHOOP, Oura, Garmin’s overnight HRV, and Fitbit. Apple’s existing HealthKit HRV field uses SDNN, conventionally calculated as the sample standard deviation:

$$\mathrm{SDNN}=\sqrt{\frac{1}{N-1}\sum_{i=1}^{N}(NN_i-\overline{NN})^2}$$

Both are measured in milliseconds when the intervals are in milliseconds. Filtering, recording length, and overnight averaging can still differ between devices using the same formula. Neither equation reveals the proprietary conversion from HRV to readiness points.