Your wrist now claims it can read your blood pressure and guess how dehydrated you are. No cuff. No lab. Just a strap.
WHOOP backs part of that claim with real data, but only for one condition: lying still, asleep. WHOOP 5.0’s Blood Pressure Insights lands within 10 mmHg of a real cuff 88.68% of the time, based on a study of 10,746 overnight readings. During a threshold interval, or three hours into a hot century ride, the moments where a reading might actually change what you do, WHOOP has published exactly zero accuracy data. That gap matters more than the launch marketing lets on.
What Changed in WHOOP 5.0’s Hardware
WHOOP launched the 5.0 and the pricier WHOOP MG on May 8, 2025. Both use a new multi-wavelength optical sensor. That sensor is the hardware behind both new scores. Battery life jumped from 4-5 days on WHOOP 4.0 to 14 days on the new strap. The device itself shrank by 7%.
None of that is controversial. It’s the two scores riding on top of the new sensor that need a harder look.
WHOOP 5.0 Blood Pressure Insights: How It Works and What the Data Shows
Blood Pressure Insights only runs on WHOOP MG, which costs roughly $359 a year according to third-party pricing reviews. Before it estimates anything, you calibrate it with three readings from an actual cuff. WHOOP is direct about one thing: this is not a medical device. It can’t diagnose or manage high blood pressure. It’s still labeled beta.
WHOOP’s own validation study is genuinely large. Researchers compared 10,746 paired overnight measurements against an FDA-cleared cuff, using a model trained on more than 32,000 sleep sessions from over 11,000 members. Systolic pressure landed within 10 mmHg of the cuff 88.68% of the time, with an average error of 5.03 mmHg. Diastolic pressure did better: within 10 mmHg 96.57% of the time, average error 3.64 mmHg.
In short: while you sleep, the number is close most nights. What that same algorithm does mid-workout is a different question. It’s one WHOOP hasn’t answered yet.
The FDA Warning Letter Nobody’s Resolved
The FDA doesn’t share WHOOP’s confidence. On July 14, 2025, the agency sent WHOOP a formal warning letter. Its argument: Blood Pressure Insights meets the legal definition of a medical device, because it estimates a clinical vital sign used to diagnose and manage disease.
As of reporting in May 2026, the two sides are still in talks. WHOOP calls the discussions “productive.” The feature still runs for existing users.
One more detail worth sitting with: a WHOOP executive has said 40% of members who calibrate the feature discover a reading in the elevated or hypertensive range. That’s a serious number to surface from a device the FDA hasn’t cleared.
The problem runs deeper than one company. Independent research on cuffless blood pressure devices as a category finds no convincing evidence that any current cuffless technology hits clinical-grade accuracy. Major hypertension societies don’t recommend cuffless devices for diagnosing or managing high blood pressure. Even the standard cuff-validation protocol, ISO 81060-2, was never built for dynamic conditions like a bike ride or a hard track session.
One independent study did test a different, non-WHOOP cuffless device during real exercise. A 2023 paper found a mean systolic error of just 0.32 mmHg (plus or minus 7.76) on a cycling ergometer test, meeting the ISO accuracy standard. That proves exercise-condition accuracy is possible with the right model. It says nothing about WHOOP’s specific algorithm. Nobody has published that test.
| Source | Condition Tested | Result | Sample Size |
|---|---|---|---|
| WHOOP internal validation | Overnight, at rest | SBP within 10 mmHg 88.68% of readings (MAE 5.03 mmHg); DBP within 10 mmHg 96.57% (MAE 3.64 mmHg) | 10,746 paired readings |
| Hayashi et al. 2023 (independent, non-WHOOP prototype) | Cycling ergometer, moderate effort | SBP mean error 0.32 mmHg ± 7.76; met ISO 81060-2 standard | 17 subjects (14 completed) |
| ESH Working Group statement, 2022 | General cuffless-device category | Not recommended for diagnosing or managing hypertension | Position statement |
| WHOOP algorithm during exercise | Any exercise condition | No data published | 0 |
Nobody has validated WHOOP’s blood pressure math mid-workout.
The Hydration Score: What “Interstitial Fluid Shift” Actually Measures
The Hydration Score comes from the same sensor. It reads shifts in your interstitial fluid, the fluid sitting between your cells, and combines that with local humidity to suggest an electrolyte target for the day.
Here’s the catch. Interstitial fluid makes up about 75% of your extracellular fluid; plasma is the other 25%. It shifts for reasons that have nothing to do with sweating: posture, sodium balance, inflammation, heat exposure, or just yesterday’s training load.
Picture judging how much gas is left in your tank by watching a needle that also swings every time the road tilts. Some of that movement is real fuel loss. Some of it is just the hill. WHOOP’s sensor is reading the needle. It hasn’t published data showing it can tell the difference during exercise, and no outside lab has checked its work either.
The 2% Rule: What Real Dehydration Costs You, and How Sweat Testing Compares
Sports scientists use a working threshold for when dehydration starts hurting performance: a body mass loss of 2% or more. A blinded 2017 trial tested that number directly. Cyclists who lost 2.4% of body mass, versus 0.1% in a fully hydrated control, completed about 8.1% less work in a 15-minute all-out effort. The heat was 34C.
Translation: real dehydration costs real watts. But knowing you’re dehydrated requires knowing your own sweat rate, and that number swings wildly between athletes. Average sweat rate sits around 1.1 liters an hour. The individual range runs from 0.2 to 5.5 liters an hour. Average sweat sodium is 36 mmol/L, ranging from 10 to 90 mmol/L across athletes.
A single sensor on one wrist can’t capture that spread without your own numbers first. Here’s how the main hydration-tracking methods stack up:
| Method | What It Directly Measures | Exercise Validation Published? | Output | Approx. Cost |
|---|---|---|---|---|
| WHOOP 5.0 Hydration Score | Interstitial fluid shift (optical sensor) | No independent exercise validation | Unitless daily score plus electrolyte suggestion | Included with WHOOP MG (~$359/year) |
| Gatorade Gx Sweat Patch | Sweat chloride concentration and volume | Validated against lab sweat testing | Sodium loss and fluid loss estimate | ~$25-35 per test |
| hDrop Sensor | Real-time sweat conductivity | Published validation vs. gravimetric testing | Live sweat rate and sodium concentration | ~$150 device, plus patches |
| Manual pre/post weigh-in | Whole-body mass loss | Decades of field use; the reference method | Body mass loss per hour | Free (a scale) |
The Overdrinking Risk Nobody Mentions: Exercise-Associated Hyponatremia
Here’s the direction almost no WHOOP review covers. A low hydration score pushing you to drink more isn’t automatically safe advice.
Exercise-associated hyponatremia, a dangerous drop in blood sodium, is caused mainly by drinking faster than you’re losing fluid. It is not caused by under-drinking. Plasma sodium below 135 mmol/L counts as hyponatremia; below 125 mmol/L, it usually turns symptomatic. More than 100 documented cases have occurred since 1985, mostly in events lasting 8 hours or longer.
Take an athlete I’ll call Derek, 44, training for his first full Ironman. Five hours into a hot brick session, his WHOOP 5.0 flagged a low hydration score. He didn’t know his real sweat rate, so he doubled his intake to about 900 mL an hour to compensate. A sweat patch test two weeks later put his actual loss closer to 650 mL an hour, with sweat sodium around 42 mmol/L. He’d been over-drinking on a guess. Once he had real numbers, he set his race target at 600 to 700 mL an hour and stopped reacting to a single unitless score on his wrist.
A low score isn’t a license to drink without limits.
From a Score to a Decision You Can Actually Use
None of this makes WHOOP’s new sensors worthless. It means a raw score needs context before it becomes a decision, and that’s the step most wearables skip.
AthleteOS already reads your WHOOP HRV and recovery trend and uses it to adjust the day’s training plan, not just display a percentage. That same pipeline extends naturally to these newer signals. A low hydration score gets checked against your planned session load, the heat and humidity forecast, and your own logged sweat-rate and sodium history, whether from a sweat test or your own tracking, before it becomes an actual fluid and sodium target for that session. Not a vague nudge to “drink more.” For Blood Pressure Insights, AthleteOS tracks the trend across weeks of training load alongside your fitness score and fatigue score, since WHOOP itself only validates the single-day number overnight. One isolated reading with no training context isn’t much use on its own.
If you’re building a race-day fueling plan around wearable signals, it’s worth reading how drink-to-thirst compares with programmed hydration schedules in ultra-endurance racing, and how combining HRV with resting heart rate and wellbeing beats HRV alone for daily readiness calls. If you’re weighing wearables more broadly, see how WHOOP’s readiness signal compares against HRV4Training’s. Start free with AthleteOS to see how these signals turn into an actual number for your next session, not just another score to glance at.