Readiness score apps for balancing rest and exercise

Short answer

A readiness score compresses last night into one number meant to tell you how hard to push today. WHOOP gives 0–100% each morning from four overnight signals. Apple, since watchOS 27, gives 0–10 that moves through the day and includes training load. They look like the same product and are not.

If you have compared two of these scores and found they disagreed, you were not imagining it — that is expected. Neither has published validation against any physiological criterion, and they are not built from the same inputs.

Note: this is a wellness comparison, not medical advice. None of these products replaces clinical assessment. Statements about other companies come from their own documentation, support pages and regulatory filings, read on 22 September 2026, and from named peer-reviewed papers.

Do you need a device to get a readiness score?

No. A readiness score is arithmetic on heart-rate variability, resting heart rate, respiratory rate and sleep — which is why the question of the best HRV monitor matters less than people expect. The hard part is capturing beat-to-beat intervals accurately, and a phone camera does that from a fingertip using the same physical principle a wrist sensor uses — photoplethysmography, reading the pulse from changes in light absorption.

Welltory's camera measurement was validated against simultaneous ECG in Computer Methods and Programs in Biomedicine in 2022: 26 elite cyclists, correlations of r = 0.77–0.94, no significant differences from ECG, and no difference between one-minute and five-minute recordings. HRV4Training published a comparable smartphone validation in the International Journal of Sports Physiology and Performance in 2017.

What a wearable adds to a recovery tracking app is the overnight window and continuous background sampling — not accuracy. If you already own one, that is a reason to use it. If you do not, a minute with your phone in the morning produces the same class of measurement, and you keep the money.

Are WHOOP Recovery and Apple Readiness the same thing?

This is the part most comparisons get wrong, and it got worse in September 2026 when Apple shipped a Readiness app and a lot of published comparisons went out of date overnight.

WHOOP Recovery is 0–100%, banded green at 67% and above, yellow 34–66%, red 33% and below. It is computed while you sleep and delivered when you wake. Four inputs go in: heart-rate variability, resting heart rate, respiratory rate and hours of sleep. HRV and resting heart rate are both taken from your last period of slow-wave sleep. The weighting of the four is not published.

Apple Readiness is 0–10, banded Recover, Pace Yourself, Ready and Go For It. It takes three groups of inputs: Activity, which includes active calories, your workout effort ratings and a training readiness component built from training load; Vitals, which includes overnight heart rate, respiratory rate, wrist temperature, HRV and blood oxygen, each compared against your own recent baseline; and Sleep, meaning your recent sleep score. It updates through the day as new data arrives.

So WHOOP tells you what your body did overnight. Apple tells you what your body did overnight and what you have been asking of it lately. A 70% WHOOP Recovery and a 7 on Apple are not the same statement about a body, and moving between the two will not give you a continuous series.

Two practical consequences. Apple Readiness requires Apple Watch Series 12 or Ultra 4 — on older hardware it does not exist, so "WHOOP vs Apple Watch readiness" is only a real question if you have the newest watch. And WHOOP's score is available from day one on any current strap.

The HRV numbers are in different units of meaning

Both companies report something called HRV. They do not report the same statistic.

WHOOP reports RMSSD — the field in its developer API is literally named `hrv_rmssd_milli`. RMSSD is a short-window measure weighted towards parasympathetic activity, which is why it is the usual choice for overnight recovery.

Apple's long-standing Health app metric is SDNN — the HealthKit identifier is `heartRateVariabilitySDNN`, and Apple defines it as the standard deviation of intervals between normal heartbeats. SDNN mixes short- and long-term variability and is conventionally a five-minute or 24-hour clinical measure.

These are not interchangeable. A number from one will not sit on a chart with numbers from the other.

Apple has also added a separate identifier, `heartRateVariabilityRMSSD`, new in iOS and watchOS 27. Its documentation page is a bare stub with no description and no units, and Apple has published nothing connecting it to the overnight metric it now calls recovery HRV. Apple's own framing of that metric is that recovery HRV is for daily stress and recovery signals while overall HRV is for broader health — but how recovery HRV is computed is not public.

The measurement windows differ too, and this matters more than the statistic. WHOOP produces one value per night from a defined physiological state. Apple samples opportunistically — its own support page says HRV is sampled while you are still, so the number of measurements varies with your activity level. That means an Apple HRV series is partly a record of how much you happened to sit still, which is why it looks noisier night to night.

One asymmetry in WHOOP's favour and one against. In its favour: a defined nightly window is a cleaner comparison across days. Against it: WHOOP publishes two mutually inconsistent descriptions of when the value is taken — one support article says the last slow-wave period, another says a dynamic average across sleep weighted towards slow-wave and towards waking. It publishes no window length and no artefact-filtering method.

Sleep: same four boxes, different signals

WHOOP reports slow-wave, REM, light and awake, estimated from heart rate, HRV, respiratory rate and motion together.

Apple reports Awake, REM, Core and Deep, estimated from the accelerometer alone. Apple says so plainly in its own methodology document: the updated algorithm is based on accelerometer sensor data. Heart rate is not used for staging.

Apple's "Core" is not a synonym for WHOOP's "light" — Core maps to the clinical N1 plus N2. Percentages from the two apps cannot be compared without saying that.

The two companies publish very different amounts about their own sleep validation.

Apple publishes a detailed methodology document with a held-out validation set of 166 participants over 299 nights, scored by certified technologists against in-lab polysomnography. The updated watchOS 26 algorithm reports sensitivity to sleep of 96.8%, specificity for wake of 78.9%, and four-stage agreement of κ = 0.68. It also reports degraded performance in a clinical population, κ = 0.66.

WHOOP describes training its algorithm against sleep-lab polysomnography in a blog post but publishes no numbers, no sample size and no study of its own. Every WHOOP sleep accuracy figure in the literature comes from third parties, and the largest WHOOP-specific one was funded by a WHOOP grant to the University of Arizona, disclosed in the paper.

The cleanest citation for both is independent: Miller and colleagues, Sensors 2022, 53 adults wearing six devices simultaneously with polysomnography and ECG, funded by the Australian Institute of Sport. Sleep versus wake: Apple Watch 88% agreement, κ = 0.30; WHOOP 86%, κ = 0.44. Stage level: Apple Watch 53%, κ = 0.20; WHOOP 60%, κ = 0.44. Both agreement figures are low in absolute terms.

Accuracy of the HRV itself

Here the honest answer requires two numbers for Apple, and quoting only one of them is how most articles mislead.

In a controlled laboratory session — 78 participants, ages 20 to 75, against a three-lead ECG — Apple Watch Series 6 showed near-perfect concordance for R-R intervals with a mean absolute percentage error of 1.15%. That is the number you will see quoted.

Over 14 days of real-world repeated measurement — 39 participants, 316 measurements, against a Polar H10 chest strap — Apple Watch Series 9 and Ultra 2 underestimated HRV by 8.31 ms with a mean absolute percentage error of 28.88%. Resting heart rate was fine at 5.91%. The authors concluded that resting heart rate is usable and HRV does not meet clinical accuracy standards for serial measurement.

Both studies are real and peer-reviewed. The difference is single controlled session versus repeat measurement in life, and it is the second one that describes how you would actually use a readiness score.

For WHOOP, the largest nocturnal comparison against ECG is Dial and colleagues in Physiological Reports, 2025: 13 participants, 536 nights. WHOOP 4.0 resting heart rate concordance 0.91 with 3.00% error; HRV concordance 0.94 with 8.17% error. WHOOP's own most-quoted accuracy paper, Bellenger and colleagues in Sensors 2021, had six participants — a fact often misreported as twelve. A formal correction was published in 2022; cite the corrected version.

Neither WHOOP Recovery nor Apple Readiness has any peer-reviewed validation. Not against training outcomes, not against injury, not against anything. That is symmetric and it is the single most important sentence in this comparison.

One warning if you go looking. Two confident-sounding citations for a WHOOP Recovery validation circulate widely — a 2023 paper in the Journal of Science and Medicine in Sport reporting r = 0.68, and a 2024 Frontiers in Physiology study with 42 participants. Neither exists in PubMed. Both trace only to affiliate pages.

Regulatory status: eight against one

This is a factual gap, not an opinion.

Apple holds eight distinct FDA authorisations across five heart and sleep features: the ECG app, cleared by De Novo in 2018 and updated by 510(k) in 2020; irregular rhythm notifications, De Novo 2018 and twice re-cleared; AFib History in 2022; sleep apnea notifications in 2024; and hypertension notifications in 2025. Two of those were De Novos, meaning they created new device classifications that other companies now use as predicates.

WHOOP holds one. Its ECG was cleared by 510(k) in April 2025 — and the predicate it used was Apple's ECG App 2.0. The clearance covers WHOOP MG only, not WHOOP 5.0 and not 4.0, which the 510(k) summary states in the compatibility row. Any claim that WHOOP 5.0 has FDA-cleared ECG is incorrect. WHOOP's irregular heart rhythm notifications have no US clearance found, and the United States is absent from WHOOP's own availability list for that feature.

Both companies now ship something blood-pressure-adjacent, and they got there differently. Apple's hypertension notifications went through 510(k) before launch, with published performance that is not flattering: sensitivity 41.2%, specificity 92.3%, from 2,229 subjects. WHOOP's Blood Pressure Insights launched without authorisation, and in July 2025 the FDA issued a warning letter finding the feature to be an unapproved device and rejecting the general-wellness argument. WHOOP subsequently modified the product and labelling, and trade press reported in June 2026 that the FDA closed the letter out for the modified version. WHOOP's current labelling states that Blood Pressure Insights is not a medical device, requires cuff calibration, and is in beta.

We could not confirm CE certificate or Notified Body numbers for either company in public sources.

Your own data: one lets you check the work

This is the sharpest difference in the whole comparison, and it is not about accuracy at all.

Apple exposes per-beat data. `HKHeartbeatSeriesSample` returns a handler called once per heartbeat with the time of that beat since the series start and a flag for whether beats may be missing. From that you can compute RMSSD, SDNN, pNN50 or frequency-domain HRV yourself, in any app you choose. You cannot get the raw optical waveform, but you can get the intervals that everything else is built from.

WHOOP exposes its own conclusions. The developer API returns a recovery score, resting heart rate, one `hrv_rmssd_milli` value per night, blood oxygen and skin temperature; sleep stage totals; strain, average and maximum heart rate. There is no beat-to-beat endpoint, no heart-rate time series and no raw signal. The CSV export is the same aggregate level. The OAuth scope descriptions confirm the ceiling: workout access is described as strain and average heart rate.

So with Apple you can audit, recompute and reinterpret. With WHOOP you can accept or ignore. For anyone who wants to check whether a score is doing something sensible on their own body, that is decisive.

How long until it knows you?

Readiness only means something relative to your own baseline, so the calibration period is a real part of the product.

WHOOP declares itself calibrated after four consecutive recoveries. Before that it compares you against typical values for your age, sex and fitness level — meaning your first scores are partly demographic rather than personal. WHOOP states this openly, and its API exposes a `user_calibrating` flag so downstream apps can tell. Confusingly, WHOOP's marketing describes a 30-day calibration with a day 1 / 14 / 21 / 30 timeline, so its own two answers do not agree.

Apple uses a different window for each feature: seven nights for Vitals, at least seven of the past 49 nights for Readiness, the last 13 nights for sleep consistency, seven days against the previous 28 for training load, and ten nights in 30 for sleep apnea. Apple withholds the metric until the window is met rather than showing you a partly demographic estimate.

Which one costs more in the long run?

WHOOP has no hardware price. The membership is $199 a year for WHOOP One, $239 for Peak, $359 for Life — and the device only functions while the membership is active. ECG requires the $359 Life tier specifically.

Apple Watch is $249 for SE 3, $399 for Series 12, $799 for Ultra 4, once. Apple does not publish an explicit statement that health features need no subscription, but every health feature on the Series 12 specification page carries hardware, age and region footnotes and no subscription footnote, while Fitness+ carries one.

Over one year, WHOOP One is cheaper than any Apple Watch. Over three years, WHOOP One is roughly $597 against $399 for a Series 12. The crossover is somewhere in year two. Which is cheaper depends entirely on how long you keep it.

One more difference that looks like convenience and is actually about data quality: WHOOP runs 14 days on a charge and charges on the wrist, so it is never off. Apple Watch runs about 24 hours and Apple's own guidance is to charge to at least 30% before bed. Missing nights are structurally more likely on Apple, which affects how many nights a baseline is built from.

What neither of them can do

Neither tells you why. A low score is consistent with a hard session, a late meal, alcohol, a coming infection, poor sleep, a stressful week, or nothing identifiable. The score does not distinguish these, and no readiness product claims to.

Neither predicts anything. There is no published evidence that either score anticipates illness, injury or a bad day, and neither company claims a validated predictive function.

Neither is a diagnostic. A persistent unexplained change in resting heart rate or HRV is a reason to talk to a clinician, not a finding in itself.

How to bring this up with your doctor

If you want to use any of this in a medical conversation, bring trends rather than single days, and bring the raw numbers rather than the score. Resting heart rate over weeks is the most interpretable thing a consumer device produces and the one a clinician is most likely to find useful. A score of 34% means nothing outside the app that produced it.

Say which device and which statistic. "My RMSSD has dropped from around 45 to around 30 over six weeks, measured overnight" is a sentence a doctor can work with. "My recovery is red" is not.

How Welltory fits

Welltory computes the same class of overnight and daily measures, reads Apple Watch, Samsung Watch and Bluetooth heart-rate monitors directly, and pulls data through Apple Health, Health Connect, Samsung Health, Oura and Garmin — so if you already wear something, it is an additional reading of the same data rather than a replacement.

Like every product in this comparison, Welltory holds no regulatory clearance, and its readiness-type outputs have no more outcome validation than WHOOP's or Apple's — nobody in this category has that. Its published validation, 26 participants against simultaneous ECG, covers the measurement rather than the score.

What that leaves is the part of this article you can act on. The reason two readiness scores disagree is that the companies use different statistics over different windows and publish neither. Welltory names what it computes — RMSSD, SDNN, pNN50 — so you can see which number you are looking at, and it reads the device you already own rather than starting your history over.

How we made it

Facts about other companies were taken from their own developer documentation, support pages, methodology papers and FDA filings, retrieved on 22 September 2026, and from named peer-reviewed studies. Where a company publishes two inconsistent descriptions of its own method, both are reported. Where a figure could not be confirmed in a primary source, it was left out.

Made with AI tools, then edited and fact-checked by people.

Data analysis by Jane Smorodnikova — the founder and CEO of Welltory. A recognized tech leader with two Master's degrees and experience at MIT, she has scaled Welltory to over 17 million users.

Written by Tatsiana Yashyna — Deputy COO at Welltory. With a background in medicine and years of working with health data, she translates research and real physiological signals — sleep, stress, heart rate, and hormones — into clear, evidence-based explanations that help people understand what their bodies are telling them.

Discounts for blog readers: up to 36% off

See what affects your energy, stress, sleep, and daily state with Welltory

Explore your data