Whoop Accuracy: Everything Rests on One Validated Number
Whoop leads with one number, the Recovery Score, but Peak and Life members can also check HRV, resting heart rate, respiratory rate, blood oxygen, and skin temperature as standalone readings in Health Monitor.
Whoop leads with synthesis over standalone numbers: the Recovery Score is the figure it wants a user to check each morning, even though Health Monitor also surfaces HRV, resting heart rate, respiratory rate, blood oxygen, and skin temperature individually for Peak and Life members. Below is what published research and Whoop's own methodology documentation say about how that score is built, plus what Whoop tracks separately and what it's still working toward.
How Is Whoop's Recovery Score Actually Calculated?
Whoop's Recovery Score is a 0-100 composite built from HRV, resting heart rate measured during the night's deepest sleep, respiratory rate, sleep, skin temperature, and blood oxygen. Whoop describes HRV as the single most influential input to the score. It does not publish the weighting, and no independent analysis quantifying how much of the score HRV accounts for could be located, so the relative contribution of the other inputs is not established here.
What Whoop calls it
"Recovery Score" is the number Whoop leads with each morning, but HRV and resting heart rate are also surfaced as their own standalone figures in the app's Health Monitor, in milliseconds for HRV, available on the Home Screen and the Health tab for Peak and Life members.
Correction, 2026-09-18
This section previously stated that Whoop doesn't display a standalone HRV or resting heart rate number prominently in its main interface. That was wrong: Whoop's Health Monitor feature, reachable from the Home Screen or Health tab, reports both as their own figures alongside respiratory rate, blood oxygen, and skin temperature. The description above has been corrected.
Worth knowing
A real peer-reviewed evaluation found that composite recovery scores including Whoop's, Oura's Readiness Score, and Garmin's Body Battery are not methodologically equivalent to each other, despite similar-sounding names and overlapping inputs. The same person's Whoop Recovery Score and Oura Readiness Score on the same day are not directly comparable, even though both claim to measure the same underlying question.
What published research found
Emerging, weaker in transitionsThe two inputs Whoop weighs most heavily into Recovery, resting heart rate and HRV, both depend on the same optical sensor accurately tracking heart rate in the first place. A 2025 study built specifically to test that under changing conditions, rather than steady state, put a Whoop 4.0 against a 12-lead ECG reference over a 20-minute protocol mixing rest and varied-intensity walking. The Whoop 4.0 performed acceptably during steady-state conditions but was less accurate during transient states, periods of rapid heart-rate change, a pattern the same study found across every wrist-worn device it tested, though a Garmin device in the same lineup showed comparatively more stability during those transitions.
That has a direct bearing on how Recovery should be read: the nightly resting-heart-rate input it depends on is measured during a relatively steady state (sleep), which is close to this device's best-case accuracy condition, while daytime HRV or strain-adjacent readings taken during more variable activity fall into the harder case this study describes.
Van Oost et al., Sensors, 2025 (doi:10.3390/s25206319)Head-to-head against an ECG. In the one study that has tested several consumer wearables against the same ECG reference at the same time, 13 healthy adults wore five devices overnight for a combined 536 nights. Whoop 4.0 was moderate for overnight HRV, with a concordance of 0.94 and an average error of 8.17 percent, but that error had a standard deviation of 10.49 percent, larger than the average itself, so some nights were much further off than others. For overnight resting heart rate Whoop erred by 3.00 percent (Physiological Reports, 2025). Whoop's Recovery score is built partly on that overnight HRV, so the night-to-night spread carries into it.
KnowYourPrime Evidence Registry: Finding KYP-ER-0005: Readiness and Recovery Score Validity.
How Does Whoop's Stress Monitor Work?
Stress Monitor gives a real-time score from 0 to 3, built from heart rate and HRV, using a log-transformed RMSSD calculation, compared continuously against a person's own 14-day baseline. Whoop factors in motion to separate physical exertion from psychological stress, and notifies a user when the score climbs, with breathwork exercises offered directly in the app.
What Whoop calls it
"Stress Monitor," a real-time complement to the once-daily Recovery Score. Both draw on the same underlying HRV and resting heart rate signals but operate on different timescales, one continuous through the day, one calculated fresh each morning.
What published research found
Insufficient evidenceA dedicated search for independent validation of Whoop's Stress Monitor score itself, the 0-to-3 composite, returned nothing. No study tests whether that specific score tracks a reference measure of psychological stress.
What exists is evidence for the two raw inputs it's built from: heart rate and RMSSD-derived HRV are each independently well studied, including on Whoop hardware in the resting-heart-rate research described above. Whether Whoop's log-transformed combination of them, calibrated against a personal 14-day baseline, tracks stress specifically rather than just general autonomic arousal is untested on the current published record.
Absence determined by a dedicated Whoop-specific search of the stress and HRV literature, 2026-09-17How Does Whoop Score Sleep?
Whoop's Sleep Performance score is built from four inputs: sleep sufficiency (hours slept against hours needed), sleep consistency compared against the prior four nights, sleep efficiency, and a sleep stress measure.
What Whoop calls it
"Sleep Performance," a distinct framing from Oura's stage-architecture-heavy Sleep Score or Garmin's stress-augmented version. Whoop's formula weighs consistency across recent nights more heavily than either competitor.
What published research found
Detection solid, staging weakerA 2025 polysomnography study tested a Whoop 4.0 alongside five other wrist-worn devices in a single overnight lab stay. Every device in that study, Whoop included, detected more than 90 percent of sleep epochs, but specificity for detecting wake ranged only 29.39 to 52.15 percent across the six devices, and stage-agreement (Cohen's kappa) ranged from 0.21 to 0.53, fair to moderate agreement with polysomnography. The study names the Fitbit Sense (κ=0.42), Fitbit Charge 5 (κ=0.41), and Apple Watch Series 8 (κ=0.53) specifically as its higher-agreement performers, which places the Whoop 4.0 outside that top tier without the study giving its individual figure.
That is the same detect-versus-stage split found across most of this device category: telling sleep from wake is close to solved, dividing sleep into stages is harder, and this particular study's own numbers put Whoop among the devices still working on the harder half.
Two earlier results point the same way. In a single sleep-lab night with six devices, the Whoop 3.0 agreed with polysomnography on sleep versus wake 86 percent of the time (kappa 0.44) and on specific stages 60 percent of the time (kappa 0.44). That lab receives research support and equipment from WHOOP, and WHOOP staff supplied the Whoop sleep data without seeing the lab results; the authors state WHOOP was not involved in designing, running or reporting the study (Sensors, 2022 [conflict statement: the authors' lab receives funding and equipment from WHOOP]). A systematic review of eight studies compared Whoop with the Fitbit Charge 4 and Garmin Vivosmart 4 against polysomnography or a portable EEG reference. Whoop was closest of the three on total sleep time (1.4 minutes short), light sleep and deep sleep, and furthest off on REM sleep, by 21.0 minutes (JMIR mHealth and uHealth, 2024).
Schyvens et al., Sleep Advances, 2025 (doi:10.1093/sleepadvances/zpaf021) · Miller et al., Sensors, 2022 (doi:10.3390/s22166317 [conflict statement: the authors' lab receives funding and equipment from WHOOP]) · Schyvens et al., JMIR mHealth and uHealth, 2024 (doi:10.2196/52192)KnowYourPrime Evidence Registry: Finding KYP-ER-0007: Wearable Sleep Tracking Accuracy Against a Sleep Lab.
How Does Whoop Calculate Strain and Calories?
Strain and calories are two separate calculations that share only their heart-rate input. Strain combines time spent in each heart-rate zone with an estimate of muscular load from strength activity, then converts that combined load into a 0-21 score through a nonlinear formula Whoop describes as inspired by the Borg Rating of Perceived Exertion scale. Calories are estimated separately, built on a real, named formula from 2005 predicting energy expenditure from heart rate, body weight, and age, with proprietary adjustments layered on top.
What Whoop calls it
"Strain," Whoop's replacement for the step-count-and-active-calories framing every other brand uses. There is no standalone step counter presented as a primary metric the way Apple's Activity Rings or Fitbit's step count are. Whoop's own materials are explicit that Strain is not the same thing as calories burned.
Correction, 2026-09-19
This section previously described Strain itself as built on the 2005 Keytel heart-rate-to-calories formula. That was wrong: Whoop's own documentation describes Strain as a separate calculation from heart-rate-zone time and estimated muscular load, converted through a nonlinear Borg-scale-inspired formula. The Keytel formula is specifically the basis for Whoop's calorie estimate, a distinct number. The description above now separates the two.
What published research found
Insufficient evidenceNo independent validation of Whoop's Strain score, or of its calorie estimate specifically, turned up in this search. The underlying Keytel formula the calorie estimate is built on has its own validation from 2005, but that tested the formula in general, not Whoop's implementation of it with Whoop's proprietary adjustments and its own heart-rate sensor. Strain's own nonlinear weighting and muscular-load estimation are entirely proprietary and unpublished, so there is nothing published to validate against at all.
Given how much energy-expenditure error the rest of this device category shows even on well-validated hardware, often 30 percent or more, a formula's general validity is not a substitute for testing the specific product built on it.
Absence determined by a dedicated Whoop-specific search of the activity and energy-expenditure literature, 2026-09-17Does Whoop Track Blood Oxygen and Skin Temperature?
Yes, since the 4.0 hardware generation. A pulse oximeter estimates blood oxygen saturation using red and infrared light absorption, useful context during sleep or altitude training, and a paired skin temperature sensor tracks nightly changes against a personal baseline. Both feed into the Recovery calculation alongside HRV and resting heart rate.
What Whoop calls it
"Blood Oxygen" and "Skin Temperature," both surfaced in the Health Monitor alongside heart rate and respiratory rate. Whoop doesn't report a standalone daytime SpO2 spot-check the way Apple's or Garmin's on-demand modes do; it's an overnight-only reading here as well.
What published research found
Insufficient evidenceA dedicated search for independent validation of Whoop's blood oxygen or skin temperature sensing against reference equipment returned nothing device-specific for either feature.
A 2020 commentary on racial disparities in consumer wearables names Whoop directly as one of several brands using green-light PPG technology, the same sensing approach behind both blood-oxygen and heart-rate readings, and documents that this technology is known to be less accurate, and can fail to read at all, in people with darker skin tones. That is a category-wide concern raised in the general literature rather than a Whoop-specific test result, but it applies to this feature by the same mechanism it applies to every PPG-based sensor on this page.
Colvonen et al., SLEEP, 2020 (doi:10.1093/sleep/zsaa159) · absence of device-specific validation determined by a dedicated Whoop-specific search, 2026-09-17KnowYourPrime Evidence Registry: Finding KYP-ER-0001: Smartwatch Blood Oxygen (SpO2) Accuracy Against Arterial Blood (SaO2); Finding KYP-ER-0002: Pulse Oximeter Accuracy and Skin Tone.
Does Whoop Measure Blood Pressure?
It estimates a range, and that range depends on a cuff. Blood Pressure Insights, available with WHOOP Life, gives a morning systolic and diastolic range built from overnight heart rate, HRV and pulse wave patterns after you log three readings from a traditional cuff, with a new cuff reading every 30 days. In Whoop's own data, 10,746 paired measurements against an FDA-cleared cuff, the average error was 5.03 systolic and 3.64 diastolic, and 88.68 and 96.57 percent of estimates landed within 10 points. Whoop does not say how many people took part, how widely the errors spread, or how the estimates hold up when pressure moves away from the calibration readings, which is where independent studies of other calibrated wearables find the largest errors.
What Whoop calls it
"Blood Pressure Insights," which Whoop describes as for general wellness purposes only and not a medical device.
What Is WHOOP Age, and How Is It Calculated?
WHOOP Age estimates physiological age using six months of sleep consistency and duration, daily steps, time spent in each heart-rate zone, strength-training time, VO2 max, resting heart rate, and lean body mass. It's framed as an "effective age," the age at which someone else would carry the same all-cause mortality risk, anchored to published mortality research. A separate "Pace of Aging" score, updated weekly and ranging from -1x to 3x, shows how quickly that number is currently moving.
What Whoop calls it
"WHOOP Age" and "Pace of Aging," part of the broader "Healthspan" feature set. It's a proprietary composite built from several individually-studied inputs (VO2 max, resting heart rate, sleep), but no independent peer-reviewed validation of the composite WHOOP Age estimate itself was found during research for this page.
What published research found
Insufficient evidenceNothing. No independent study of the composite WHOOP Age or Pace of Aging score, as opposed to its individual inputs, turned up in this search, and by its nature it may be difficult to validate at all: there is no external reference standard for "effective age" the way polysomnography is a reference standard for sleep or a chest-strap ECG is for heart rate.
Its inputs carry their own separate evidence, several of them reviewed elsewhere on this page and site, but a composite is not automatically as accurate as the sum of validated parts. Whether Whoop's specific combination and weighting of those inputs actually tracks mortality risk the way it's framed to is untested on the current published record.
Absence determined by a dedicated Whoop-specific search of the healthspan and biological-age literature, 2026-09-17Can Whoop Support a Corporate Wellness or Burnout-Prevention Program?
Nothing published tests that. Workplace studies have used WHOOP as a heart-rate recorder, but none has tested its Strain, Recovery or Stress Monitor scores against burnout, work performance or long-term health, and Whoop does not publish how the scores are weighted.
What the research covers
Used as a recorder, scores untested
Not testedIn hospitals. A pilot fitted 20 emergency-medicine residents and nurses with a WHOOP band for six weeks to see whether clinicians would wear it and use the platform. It measured feasibility, not whether the scores predicted anything. A review of 61 studies measuring surgeons' stress at work found WHOOP in 6 of them, used to record heart rate and HRV.
Why the scores are hard to read. A 2025 review of 14 composite scores from 10 manufacturers, WHOOP's Strain, Recovery and Stress Monitor among them, found that none of the companies disclosed their exact formulas and few offered validation. When the weighting is not published, a user cannot work out why a score moved. The overnight HRV underneath is reasonably accurate on average: a WHOOP 4.0 agreed with a simultaneous ECG at a concordance of 0.94, with an average error of 8.17 percent, but the error varied more from night to night (standard deviation 10.49 percent) than its own average.
Two further limits apply to every vendor in this category. KnowYourPrime's HRV desk rates whether higher occupational stress goes with lower heart rate variability as Emerging, on six studies: HRV was lower when groups of workers under more job stress were compared (Industrial Health, 2019), and lower in doctors' own stress periods than in recovery (Occupational Medicine, 2026), but when 101 teachers rated their job demands through the school day, the demands did not lower state HRV (Anxiety, Stress, and Coping, 2021). And the recovery desk rates whether consumer readiness and recovery scores measure what they claim as Uncertain, on five studies. A workplace program acts on individual people, which is the level the evidence reaches least.
Correction, 2026-10-25
This section previously said the link between work stress and HRV did not hold within one person's day. Doctors' HRV was lower in their own stress periods than in recovery periods; the null result is narrower, job demands rated through a teacher's school day.
What randomized trials of workplace wellness programs found, including the trials that gave employees a tracker, is set out in whether workplace wellness programs and wearables work.
KnowYourPrime Evidence Registry: Finding KYP-ER-0006: Work Stress and HRV; Finding KYP-ER-0005: Readiness and Recovery Score Validity.
Whoop Review: Is It Worth It, and How Much Is the Membership?
Whoop is sold only as a yearly membership that includes the band: $199 for One, $239 for Peak and $359 for Life in the US. On published evidence, its overnight heart rate and HRV are moderately accurate with wide night-to-night swings, it detects sleep well but misses much of the time awake, and its headline scores (Recovery, Strain, Stress Monitor, WHOOP Age) have no independent validation of their own.
What the evidence and the price say
Heart. Against a simultaneous ECG over 536 nights, the Whoop 4.0's overnight HRV had a concordance of 0.94 and an average error of 8.17 percent, with a standard deviation of 10.49 percent, larger than the average itself. The Oura Gen 4 in the same study erred by 5.96 percent (Dial et al., Physiological Reports, 2025).
Sleep. Over five nights of home polysomnography in 18 adults, the Whoop 4 scored sleep stages correctly 68.68 percent of the time (kappa 0.54), behind the Oura Ring 3 (72.5 percent) and Apple Watch Series 9 (72.3 percent) and ahead of the Fitbit Charge 6 and two Garmin models. Most wrist devices overestimated sleep and underestimated time awake, more so on disrupted nights. This is a preprint, not yet peer reviewed (Topalidis et al., Research Square, 2026). In a separate one-night lab study of six wrist devices, every one detected more than 90 percent of sleep but caught only 29 to 52 percent of time awake (Schyvens et al., Sleep Advances, 2025).
Which Whoop. Every study above tested the Whoop 4.0. The current WHOOP 5.0 and WHOOP MG bands had no published independent validation in a search of Europe PMC in October 2026.
Price. Whoop's US membership page lists three yearly tiers, each including the band: One at $199 (Sleep, Strain and Recovery, VO2 max and heart rate zones), Peak at $239 (adds Healthspan with WHOOP Age, Health Monitor and the real-time Stress Monitor) and Life at $359 (the WHOOP MG band, adding ECG readings, irregular heart rhythm notifications and Blood Pressure Insights). Whoop describes the membership as HSA/FSA eligible and offers a one-month free trial of Peak on a certified pre-owned WHOOP 5.0. Prices differ outside the US.
What Whoop Doesn't Do (Yet)
Two real limitations worth knowing, one permanent, one in progress.
No ECG on Whoop 4.0 or 5.0
The standard Whoop band has no electrocardiogram sensor. Whoop MG, a separate medical-focused device requiring a WHOOP Life membership, added an on-demand ECG feature called Heart Screener via a conductive clasp, not available on 4.0 or 5.0 and gated by region.
No shipped CGM integration
Unlike Oura, which launched a live continuous glucose monitoring integration with Dexcom's Stelo sensor in 2025, Whoop has publicly discussed exploring glucose tracking but had not shipped an equivalent integration as of this writing.
If you want one clear daily number to check first, Whoop's synthesis-first design leads with that, while still letting Peak and Life members drill into the individual Health Monitor readings behind it. Apple's more granular approach puts each metric on its own screen from the start rather than behind a composite score.