Data Scientist Breaks Down why WHOOP doesn't Work.

Fitness wearable review. I dig into my WHOOP Strap data and show that the WHOOP strap's data and recommendations are just as bad as worthless.

[TheHeadincharge]

While I think this video is very good in concept, I think you make a lot of assumption errors about what these metrics are supposed to represent according to WHOOP themselves. The assumption that strain and recovery should be a negative correlation makes sense on the surface for example, but is most certainly an improper understanding of the metrics themselves. Recovery is based off of how well you actually recovered from the strain, more specifically it's really just a measure of how well you slept as it records its key metrics during sleep, not wake (a potential problem IMO, but not the one you address). You can heavily exceed your strain goal in a day and also have an extremely high recovery score if you get good, effective sleep; the two variables are not inherently correlated. WHOOP even clearly states in their strain section that it is ok and even good to exceed your strain goals as long as you have good recovery habits after. Every individual's recovery habits are going to be different, so making broad assumptions based on a sample size of 1 person is also already problematic in this context. Your body may simply recover very well because you have good sleep habits or sleep longer after a high strain day for example. I personally have an issue with WHOOP in that I think the names for some of the things they track can be confusing, but I think your key assumptions about these metrics are quite problematic. While I appreciate the effort of this video as I've been interested about WHOOP myself for a bit now and equally curious about their claims, I think this video definitely missed the mark in a lot of ways and I wish you would have done a bit more research. As anyone who deals with data knows, anyone can analyze some data and make some charts, it's the interpretation that is the most crucial and difficult aspect.

You also don't seem to state how long you wore the strap for and what range you're including data from? Or at least I can't find it in the video. Since most of the metrics are inherently based upon your own personal baselines, I would have pretty major concerns if you included data from within the first two weeks of wearing the device in these charts. Once again, I appreciate the effort here, but as a researcher, I found it pretty difficult and at times frustrating to watch this video and have it be presented as an objective dismantling of WHOOP when you are making some pretty large errors in basic experimental design and analysis through most of the video. I'm not trying to be rude here, but I think you can do a lot better if your goal is to try to teach people about how to properly analyze and interpret data. As it is now, this video is not much better than most of these styles of "analytic" YouTube videos that often underwhelm in the actual analysis and interpretation of the subject and lack much needed nuance. Once again, parts of the video were great and I loved the concept and effort you put into it, I just think there's some key things you overlooked that are worth considering in future content, especially when the average person watching these videos likely doesn't have a good grasp of how to properly interpret an r-coefficient or an effect size.

[WaileaRose]

Same here...I returned and got my money back! I was testing my heart rate with my chest strap through my Peloton and my apple watch and manual pulse check and these 3 were very close. The Whoop on the other hand was way out to lunch not to mention the crazy spikes up and down. Thanks for the YouTube Video...confirms what I experienced!

[tomastorasen]

First, I'll say what was my expectation and need. Im an elite powerlifter, who takes lower heart rate walks as cardio. I have an extremely stressful job that is possible to adapt in different ways due to my privelage of planning my days pretty freely.
My interest was maximizing recovery, sleep and keeping track of stress points that I wasnt aware of. How exactly correct it measures I cant say, but I can say it was extremely consistent in telling me the same things depending on doing many things differently, meaning many many small changes, showed me how they added up or down depending on what I did, which blew me away of how well it quickly taught me what was good and what was not.
Calories and so on, how many steps Ive done in a day, metabolism and so on, those things I already rely completely on my own knowledge.
But for recovery, absolutely amazing.

[bastikuhn]

About the accuracy of the Whoop, I have used Whoop for almost 3 years now. Also, I always used a Garmin Fenix Next to it, and have the HRM Pro from Garmin. Garmin on the Wrist, and Whoop on the wrist are both off. But when I put the Whoop into the Pants or on the upper Arm, the measurements of Whoop and the HRM Pro only differ by around 3bmp on average and are simular to each other. So when I don't feel like wanting to wear the HRM Pro, for example on MTB tours, I connect the Fenix to the Whoop which I have usually in the Pants and trust the results. To be honest, I have used whoop so long now, that I can predict when I gone have a green, yellow or red recovery. I only just use it, since I still have credits until end of 2026 and hoping for some cooler Updates until then. The Strength trainer is useless as it is, the breathing stuff hidden away in the app, and I actually like the new Garmin since the Update way more than the Whoop app.

[niles_stone]

Same experience with the inconsistent feedback and inconsistent hr data. Thanks for the video.

[dawnshea]

When WHOOP was first described to me, I had the exact same conclusion as you and your friend. Sorry you were out all that $, but thanks for giving us the data. Great break down!

[RIP]

Hey, recently Whoop updated their software that improved their heartrate precision significantly, hence it may also improved sleep precision by the same margin.

Will be nice to see that remark somewhere in the pinned comment and see a retest from your side :)

[sammydolezal]

You should do a video on Apple Watch! Just curious if it also has the same issues with its health and fitness features

[liluliana]

I love the idea of WHOOP. Being a data nerd myself I love the idea of being able to see what's going on in my body and how that changes over time and what's affecting it. But these sorts of insights mean nothing if the data being collected isn't even accurate or consistent.

I think I'll hold out for 5.0 and wait to see what the accuracy is like on that and I'll just stick to my pixel watch and fitbit insights for now

[dennisstepanek]

i've had Whoop about six months and I just couldn't believe how it measured my heartbeat during any activity. I run, strength training and bike ride and couldn't believe the garbage heart data i was getting. Whoop's strain is totally useless and forget about using chest strap. Sorry Whoop is a total garbage. The only nice thing about Whoop is that it has awesome app, but totally useless optical sensor to but in quality hr data. And pay for this every month is totally throwing the money out of the window.

[SirRasor]

Gold standard in HR measurement is Polar H9/10 but I'm just nit-picking.
Also thumbs up for the good video :)

[michaelkeaton]

I just got the Whoop, and it's almost dead on with my Garmin FR935. It's interesting you said the Garmin chest strap is the good standard, I've never heard that. It's usually the Polar H10, and i can 100% agree the polar is more accurate; the Garmin was the worst I've ever used. It constantly stopped or would say my hr was over 200 when i was walking. I think all this shows that these devices seem to work better for some than others.

[AndrewG]

My whoop tracks my heart rate really, really well. I do wear it fairly tight, though, so maybe that is a difference. I wear a chest strap when I am on the bike, and it tracks within 1-2 bpm of the chest strap. Same with a soccer workout.

[pinfilm]

absolutely underrated video, thank you so much for this video !!

[loopba]

Think the real value of Whoop is in the community aspects. Guess could be worth it for groups, local clubs, etc if helps improve socialization. I cancelled also, as maybe its worth $5/month for me. Using Apple Watch and Athyletic instead. Also didn’t auto track my naps, and workouts

[gregmattson]

I don't know.. you've got the data guy as far as heart rate goes, my suggestion would be to analyze it rather than give a subjective view? It looks like they correlated quite well just based on the data you gave.

As far as the correlation of perceived effort vs actual, yes that is going to differ quite a bit between individual and individual, at a macro level there is no way that that measure is going to be completely precise. Personally, I feel that it tracks quite well, with the extremes being a better than the middle.

[GutoAmorim]

Thanks for that! I find all reviews are so LOVE MY WHOOP but don't show proper data. It's an expensive thing to be giving the same data most watch can.

[mike]

Polar H10 is the gold standard for heart rate measurement

[tryptamigo]

5:22 what city is this? kinda envious of this nice wide sidewalk and view.

[johanesburg]

great review and really interesting to hear your experience. As a garmin user, I really was looking forward to the deeper insights. Initially it seemed really impressive but then I noticed that some activites strain score simply didn't make sense. A slow jog was recorded as a near max-heart rate effort. this was clearly wrong and my garmin epix was reporting a way lower heart rate. The final straw was when walks with my dog were repeatedly registering as a high strain activity. I simply couldn't trust the data and thats where it fell down - all the sophisticated modelling is based on questionable data, which then skews the recommended recovery. I've gone back to just using the garmin epix and taking more notice of the overnight HRV, which seems to be the main factor the whoop uses in determining recovery. Garmin have just added a new sleep coach feature as well which is more realistic in my opinion. It never recommends more than 9 hours sleep. Having tried to get a recommended near 10 hours the whoop recommended once, unless you are a sloth it is never going to happen.