For several model generations now (stretching back to the 3.0 strap) many of us have highlighted the same problem: the Whoop optical sensor on the wrist consistently under-reports heart-rate during strength training. Dozens of forum threads document it, but the issue remains unresolved while new analytics and paid add-ons continue to roll out. My lower-body workout this morning is the latest proof.
What I did today:
Polar H10 chest strap on the sternum (my reference).
Apple Watch Ultra 2 on the left wrist.
Whoop MG on the right wrist—exactly where Whoop recommends for Velocity-Based Training.
56-minute session: dumbbell lunges, Romanian deadlifts, back-squats, calf raises.
What happened:
Whenever the Polar pushed into the 130-to-150 bpm range, the Apple Watch followed within a handful of beats. The Whoop, however, sat 20 to 40 bpm lower on every lift. For example:
During lunges the Polar and Watch climbed to the mid-130s, yet the Whoop hovered near 100 bpm.
On Romanian deadlifts the Polar peaked at 152 bpm, the Watch read 155 bpm, but the Whoop reported just 103 bpm.
Later back-squat sets repeated the pattern—high-140s on the strap and Watch, high-110s on the Whoop.
Because the MG never registered the peaks, my Strain and Healthspan scores show almost no time in Zones 2-3, even though the raw effort was clearly there.
Why this has the community upset
The problem is well known. Threads about wrist under-reads date back years; they resurface after every firmware update. Users swap tips such as: “move it to the biceps,” “tighten the strap,” “warm up longer” yet the core optical algorithm still can’t follow dynamic, isometric lifts on the wrist.
New features depend on data that isn’t accurate. Velocity-Based Training, Strain Coach, Healthspan, and stress-tracking all rely on real-time heart-rate. When the underlying sensor misses 20-40 bpm, the fancy insights become noise.
Adding features without fixing hardware feels tone-deaf. People gladly pay a premium subscription for Whoop, but it’s hard to justify when basic heart-rate accuracy still lags behind every mainstream watch.
What we need now
A public roadmap (or beta) for a sensor fix. Tell us what is being done to close the gap for resistance training and when we can test it.
An option to merge chest-strap data into Strain and Healthspan so hard work isn’t lost while the optical issue is solved.
Transparent communication. Acknowledge the long-standing nature of the problem and keep users updated, rather than quietly releasing yet another “insight” that rests on shaky data.
I’m ready to send my Polar TCX file, Apple HealthFit export, and placement photos—whatever engineering needs. The community’s patience, though, is wearing thin. Please focus on fixing the core measurement before stacking more features on top of it.
Thank you for reading and eager to hear back from the Whoop Team.
Nicely stated, concise and some great points. Something is not right in my opinion. Ironically for cycling it reports too high a HR, move up the arm and maybe it gets close but am I moving to far and not catching the highs properly.
Allow external HR chest straps or import the work out from what the user trusts and use that data. I would say Whoop analytics are pretty good, perhaps stick to that, dump the device and import the data from other devices
I could not agree more. There is often a significant difference in HR readings during sports other than strength training. If the optical sensor on my Ultra 2 can do a much better job than the MG, it is clearly not just about the limitations of optical HR sensors.
All of this makes data coming out as insights incredibly difficult to take seriously.
I find it both interesting and concerning that Whoop still hasn’t responded to the repeated concerns being raised about device accuracy. Particularly during strength training sessions. This platform was supposed to be a space for real conversation between the Whoop team and its users, not just a user-only forum. But more and more, it’s starting to feel like a support board where feedback goes into a void.
The community has been great, and it’s clear that many of us are noticing similar patterns in data inconsistency, especially under movement-intensive workouts. At some point, we need acknowledgment or transparency from the Whoop team! Even just an update saying the issue is being reviewed or investigated would go a long way.
With new wearables entering the market and offering competitive or even superior tracking capabilities, Whoop can’t afford to keep ignoring these issues. If they don’t get their act together soon, they risk falling seriously behind.
Agreed. For people like me, who own a top Garmin watch and an AW Ultra 2, I am seriously start to doubt my commitment to Whoop. Especially given how much else there is in the market.
I like the idea of Strain but if heart rate measurements are inaccurate so is everything else. Unacceptable.
1 year-the cost difference vs my previous Whoop (initially went for 5.0);
Definitely cancelling then if there is no change in how this company treats the most important reading these devices need to reliably and accurately track.
Even on bicep its not getting the HR peaks during weight lifting. My Garmin Fenix 7 pro ss on the wrist is much better on reading that peaks, than 5.0 on the bicep, I see after almost every set 120 -125 bpms. I‘m already really tired of the 5.0.
It’s been nearly two weeks since I posted evidence, complete with device comparisons and workout data, about serious heart rate underreporting issues on the Whoop MG during strength training. I outlined how the sensor misses by 20 - 40 bpm compared to the Polar H10 and Apple Watch Ultra 2, and I made clear requests for a path forward: transparency, a sensor fix roadmap, and a chest strap integration option.
Not only have we received no response from the Whoop team, but it’s becoming increasingly clear that legitimate concerns brought up by dozens of users across generations of devices are being ignored here in Whoop’s own forum. This is not a random Reddit thread. This is your official community platform, and we deserve acknowledgment.
The issue is not anecdotal anymore. It’s a pattern: underreporting during dynamic or isometric lifts, inflated recovery metrics, and faulty strain/Healthspan scores. All of these affect the core value proposition of Whoop as a data-driven recovery tool.
The silence is seriously concerning. With new competitors entering the market, many of whom are getting accuracy right, ignoring your base of loyal users sends the message that accurate performance metrics matter less than flashy feature rollouts.
Please respond. Not just to me, but to the dozens of members who have posted similar issues over the years. Transparency, even if the fix is months away, would go a long way.
I have reported issues with HRV and no response. Burying their heads in the sand at the moment. Whoop MG seems very unreliable and I suspect they know it from their own devices
The only responses I have received advise me not to repot again as doing so will put my report/complaint to the end of the list. Or to request I provide screen shoots. Quite clearly simply to again put me off reporting problems Whoops must know exist.
It’s been 18 days since I posted about years of wrist HR under-reporting on WHOOP MG, and I still haven’t received any response from WHOOP support or the team.
I’ve been a WHOOP member for years, and I’ve repeatedly raised the issue that the wrist sensor under-reports HR, especially during activities with arm movement. This is a core accuracy problem that should be addressed before layering on more features.
Silence for this long, especially on a long-standing and well-documented accuracy issue, is frustrating. I’d appreciate at least an acknowledgment that the problem is recognized and being worked on.
Read your thread on HRV spikes had switched form Oura in July. I also have a low HRV by nature and have had a few crazy spikes. Seeing their slow response form me is their norm I WANT OUT
Seems to me, from the few months I have been on Whoop, the two key factors are Steps and HRV. Steps has a huge bearing on Whoop age & Pace of aging. HRV is the main metric for Recovery.
The spikes are unusual but have seen a few people now experience this. On those days my Garmin and Oura did register a higher than normal HRV but not into the 170s like my two spikes. Garmin and Oura doesn’t carry these results into the future to give poor recovery
But the issue, and they say the do understand, is these outlier results are still included in any decisions on recovery. Because they use averages these spikes create a very high average so a great day followed by the next day dropping into the red and will stay there until the spike drops out the average period which looks to be a month.
I simply do not trust what whoop is telling which is a very sad state of affairs.
Yes. The funny thing is: Whoop startet without measuring steps (which I was in line with and I’m still not interested in). There was even an article “Why Whoop doesn’t count steps”. Now all of a sudden steps became a major focus and play a big role.
This alone shows in what direction Whoop is heading to. I also agree with HRV. I too get the feeling that recovery heavily relies on HRV. And the rest on heart rate which is erroneous.
Yeah, I read Whoop only just started to include steps after years of not. I can only assume as you appear to also seem to, that Whoop needs more members so is migrating to a more health focused device than a sports orientated one. If I wanted a pedometer and have my steps be important I would have bought a FitBit.
I would love see Ronaldo jump on this forum and ask what is going on. I bet he as paid millions for the brief appearence in the whoop commercials. Perhaps he joined Whoop because his sports career is coming to an end and he wants a health orientated device
It’s now been a full month since I raised this issue and still no meaningful response from WHOOP support. I first bought into WHOOP back in July of 2017 when the 2.0 strap was being sold for around $500 outright. Since then I’ve stayed on every generation of the hardware, hoping accuracy would improve. Yet here we are years later and the same flaw persists: the wrist sensor consistently under-reports heart rate by 20–40 bpm during weight training and other high-intensity activity.
In that time WHOOP has raised billions in funding, scaled to a valuation north of $3 billion, and pushed new features like Strain Coach, Healthspan, and velocity-based training. But the fundamental promise of the strap (accurate heart rate) still isn’t being delivered. When the underlying heart-rate trace is wrong, those “insights” are built on sand. Healthspan in particular highlights the issue: you can’t give members a longevity score based on time in zones if the zones themselves are being missed entirely.
The frustrating part is the silence. Long-time members keep posting data, and the reply is usually to “try a different placement” or “send screenshots,” even though the problem is well documented across years of posts. It feels like blaming the user instead of addressing the flaw. Meanwhile, competitors are rolling out cheaper devices that appear to be tracking heart rate more reliably in the exact same conditions.
I’ve been a believer in WHOOP since 2017, but at this point it feels like the company doesn’t care to fix what’s broken. Until WHOOP publicly acknowledges the problem and puts a roadmap in place to correct it, members are going to keep losing patience. Ignoring this core issue while stacking features on top is shooting yourselves in the foot, and if it continues, it will sink the brand.
The steps recorded by my Garmen Fenix 5 are up to 100 % more than whoop. The claim is whoop eliminate what they believe are not relevant steps during daily activities. Yet my recovery, estimated age and rate of ageing are affected by the incorrect number of steps I take. As for recovery I often feel good and ready to go but according the whoop I am only 50% or so recovered, Making whoop less than useful in monitoring my use of exercise to avoid overtraining.
