Detailed Review : WHOOP – an athlete recovery monitor/advisor and HRV-enabled oHRM for the wrist.

WHOOP Strap 2.0

WHOOP is a potentially super-clever piece of performance/training optimising kit for serious-to-pro athletes covering most sports, including triathlon & team sports.

In simple hardware/software terms it’s ‘just’ a wrist-based heart rate monitor and an online/smartphone app.


It offers a holistic HRV-based view of your ‘recovery‘ coupled with a view of your daily HR ‘strain’. Specifically, strain & recovery feed into your readiness to train and into your sleep needs.

What is it?

It’s a wrist-based optical heart rate monitor made by WHOOP. It has the ability to be worn during swimming and is designed to be worn 24×7.

It takes continuous/high rates of readings and analyses nightly RR/HRV data.

It has an internal rechargeable battery and a variety of straps can also be purchased.

It has no display and is a ‘closed system’, with no standard means to export data and furthermore it does not broadcast HR over standard Bluetooth/ANT+ protocols. Data can only be viewed on the WHOOP app, to which there are no public import/export mechanisms.

For monitoring an exercise, the WHOOP app only has fairly rudimentary information on display but can use your phone’s GPS for speed and location data. Other than GPS, there is no other external sensor support.

The app does also allow numerous, subjective inputs (not used in calculations).

It’s focus is 24×7 strain, recovery and readiness based on HR data and HRV data. The resulting knowledge and insight into your training regime is likely to be currently missing, at least in part, from your training regime.

WHOOP is positioned to sit alongside whatever you currently do. Triathletes’ will still use their existing Suunto/Polar/Garmin devices to execute, monitor and analyse their workouts.

To see the value in WHOOP you have to understand that, when training, it is the LATER ADAPTATION of your body to the STIMULATION of the exercise that results in a positive change to your fitness. WHOOP endeavours to make sure that you have information to enable you to adapt as best you can. Part of that involves recommending to you the amount of sleep you need ‘tonight’ and also assessing what you actually achieved and benefitted from in your sleep by the time you wake up ‘tomorrow’. It then recommends a level of strain/activity for the day.

WHOOP Strap 2.0What it is NOT

It’s not a heart rate monitor that you can link to ANYTHING else. It only works with the WHOOP iOS app (Android later this year)

It’s neither a ‘step counter’ nor a typical activity tracker. It does track calories but if that’s the sort of thing you are focussed on then WHOOP is probably not for you.

So What?

Maybe I’m not explaining this very well if you are not already excited.

In the original draft of this review I went off on a tangent here to give you my life story and how I tried to find a product such as this a couple of years ago…but couldn’t.

Suffice to say I am excited by the potential of a wrist-based product like this. As far as I know; it is unique.

If your sport requires that you care about your recovery and adaptation to exercise then you should, at least, consider WHOOP.

OK. You’ve got this far. Thank you. I’ll do the regular review stuff now. You will probably have already noted that I’ve sat on the fence quite a bit.

WHOOP is priced at around the £/Eur/$500 level. As much as I’d love you to buy one of these and support my blog with a small automatic commission that comes from that when you use the discount code, below, I need to do it in good conscience. Luckily I have quite a few gadgets to bring to bear to endeavour to test some of the claims made for WHOOP. I’ve used WHOOP continuously for well over 6 weeks and this review is based both on that usage and on several actual and emailed conversations with WHOOP staff. But it’s going to be tricky as WHOOP is a closed system in terms of data. As I’ve already said, I’ll do my best.

WHOOP Strap 2.0Contents

The pleasant cylindrical packaging contains: the WHOOP device; an interchangeable strap; a proprietary charging cradle; and a generic micro USB cable.





The WHOOP Device

WHOOP has a novel clasp. It’s very light, comfortable and not really noticeable when wearing it. There are no electronic pods inside rubberized straps that can pop out and get lost – like from some other brands.

The strap is secure in normal usage

There’s only one way the WHOOP can be inserted into the charger and, as you can see from the images below, the metal ports on the side of the WHOOP need to line up with the pins on the charger for it to charge. You can even charge it whilst wearing it.


This slideshow requires JavaScript.


Let’s look at WHOOP’s own-designed sensor array. For comparison I’ve shown it next to a Scosche BLE/ANT+ armband HR monitor and its sensor array.

This slideshow requires JavaScript.

There are 2x green photodiode LEDs. That’s similar to many optical sports sensors on the market today, although the Scosche has 3 (2 green, 1 yellow). Apparently green light produces optimal results for most people. WHOOP also has an orange LED but that has not yet been firmware enabled.

Sensor spacing can affect accuracy.  WHOOP and Scosche are broadly similar in that respect. In any case much of the ‘accuracy’ comes from the algorithms that you can’t see.

WHOOP also contains a 3-axis accelerometer which detects activity and the data from this accelerometer is used to cancel motion artefacts (errors caused by moving the wrist).

When you sleep, WHOOP continually takes high frequency HR readings (100/sec) but only at night are they used to determine the recovery score through heart rate variability (HRV) analyses. HRV analysis potentially could be determined during the day but this is not yet enabled. There are no features which could yet use that facility.

There is a lot of science behind HRV and recovery using the RMSSD calculation. Google it for more info. Alan Couzens blog (here) is a great place to start.

Inbuilt sensors also determine ambient temperature (a proxy for environmental temperature) and there is a touch sensor to determine if the device ie being worn or not.

Battery Life

Battery life is given as up to 44 hours ie not quite 2 days. That seemed about right from my experience and I was charging the device every other day, which was fine. Full charge time from empty was said to be about 90 minutes but I found that a good top up of 30 minutes whilst I was at a desk kept things going nicely. Unusually for wrist devices, both straps and watches, WHOOP can be charged whilst being worn.

LEDs indicate battery life:

  • 3 solid white lights = full Strap battery
  • 2 solid white lights = 60%
  • 1 solid white light = 20%
  • 1 solid red light = 10%

I occasionally ran out of battery without realising it. Grrr.

Data Transfer

A full transfer of 2/3 days worth of data over Bluetooth to iOS can take up to 5/6 minutes. That’s fairly slow. That transfer can be initiated manually but happens silently in the background if the app is open and if a live pairing exists. So, although it is fairly slow, you will never normally notice the transfer happening.

WHOOP Versions

The original v1.0 product was release in September 2015. The WHOOP 2.0 was released a year later in Oct/Nov 2016 with the addition of: automatic sleep-detection; the sleep coach; and more accurate sensors & algorithms.

WHOOP – Why it’s different in how it works

WHOOP can communicate over Bluetooth but will never broadcast as a ‘normal’ heart rate monitor over Bluetooth in a standard form.

Here’s why.

WHOOP was designed with the sole purpose of being ‘always on’ to capture raw, high frequency HR data and other sensor data and, generally, sending that for processing ‘elsewhere’. This enables battery conservation for extended periods.

So ALL the processing is either done in the WHOOP app or on the cloud service (and the results passed back to the app).

Contrast that with other commercially available, wrist-based, devices – none of which yet produce HRV data to my knowledge. They perhaps have to: handle ANT+ and BTLE channels; or have to incorporate a display and/or an on/off button; and/or be an activity/sleep tracker; and/or provide smartphone notifications. Lots of other functions, lots of other complications, lots of battery juice eaten.

Setting Up The App

The WHOOP logo to the right clicks through to the WHOOP app on iTunes.

I used version 1.6.3 which works fine on an iPAD.

You have to create an account and a profile. You know how to do that; it’s all standard stuff.

You have to pair the device to the iPhone/iPAD. Again, you know how to do that – it must be worn in order to pair.


Here is the manual: WHOOP v2 User Guide – March 20 – 2017

Here are setup, wearing & charging instructions: (this link).


You strap it on and you are good to go for about 2 days until the next recharge. Simple.

whoop review

WHOOP+HYDRO BAND shown worn away from the wrist

Well of course it’s NOT quite so simple.

  1. As shown in the image above the wearing position IS important, as it is with all optical HRMs. If you wear it too close to your wrist bone ie closer than shown, then you risk wrist movements affecting the accuracy of the readings. You could wear it on your upper arm for more accurate readings, strap size permitting, but it IS designed to be wrist-worn.
  2. The WHOOP app takes a while to build up your baseline data. Over 3 days. That’s normal for HRV-type apps. Once the baseline kicks in then so does the richness of information from the app.
  3. The app does show battery status but you will likely only look at the app a couple of times a day. You need to remember to charge it every other day. Surprisingly you can charge it while you wear it.
  4. Optical HR readings are notoriously susceptible to many factors causing inaccuracy from hairiness to the colour of your skin to movements. There’s a LOT of movements going on in the wrist area and they’re all potentially subtly different across different sports.

Suffice it to say, you may have a different experience to me.

A dark band was supplied with the model I had. It was like a strong elasticated sticking plaster-type material (hard to describe). This was perfectly fine for every usage EXCEPT swimming when it ‘flips’ and twists leaving the sensor facing outwards. If you intend to swim with it you will need the HYDRO-BAND which is a more rigid woven strap that will stay in place. That’s the white band shown in the previous image.


So that’s all well and good. We’ve got LEDs, sensors and an ‘app’. But most of us want actionable data, presented simply. This is precisely the intention of WHOOP. Let’s see how it does that.

I’m going to look at WHOOP+me. Here’s a bit about me so you know where I’m coming from: I train for TRIATHLON  between 40 minutes a day and 5 hours a day, sometimes more. I have managed 10-20 hours a week almost every week for years. I sleep between 5 and 8 hours a day. My diet is generally good and I only drink alcohol occasionally and I neither smoke nor take non-prescription drugs. I have a partner. I train for triathlon multi-sports including, occasionally, gym. I would consider myself an occasionally half-decent age-group triathlete. So I am by no means anywhere near a pro level. BUT I take my training seriously even if nobody else does 😉

I’d say WHOOP is aimed at people similar to me (or better) who consider themselves an athlete of some sort, probably training 7 hours a week or more for triathlon. WHOOP is looking at giving someone at MY level ‘notable gains’ but much more serious athletes would be looking for the fabled ‘marginal’ gains. Important gains in either case and potentially EASY gains in either case.

For WHOOP to work it automatically records your sleep and automatically records your ‘strain’ (exercise). If you didn’t wear it for the rest of the day its predictive algorithms should still broadly work; although you might be surprised at the impact of the strain of a meeting or the strain of walking to work.

WHOOP is not invasive to your lifestyle. Really, you only need to look at the WHOOP app 2-3 times a day

  1. When you wake up – enter qualitative data and review your sleep performance – how recovered and ready for training are you? WHOOP suggests an appropriate level of STRAIN/LOAD for you for the day. Adapt your plan accordingly.
  2. Only if you intend to train a second time during a day might you look at how strained you were after the first session. Adapt your next workout that day accordingly towards WHOOP’s daily strain target for you.
  3. After your last exercise of the day – enter qualitative data about your workouts and WHOOP then determines how much sleep you need to recover. Adapt you evening and night plans accordingly.

Of course you can ignore what it says entirely or look at it more or less frequently.

TRAINING: You CAN use the WHOOP app to show your HR and you can probably record GPS with your smartphone. I never did that. Pointless. You are a vaguely serious athlete, right?  WHOOP just slots alongside the Garmin/Polar/Suunto regime you already have (I know as a triathlete that you DO have one!). You want to know how fast your HR is or how fast or powerful you are right now during your training? You look at your normal sports watch. That’s why you bought your normal sports watch. You want to check if you hit those power durations or paces during training? look at Training Peaks/SportTracks/GoldenCheetah, that’s why you have them. Nothing changes! You just wear the strap all the time and very occasionally tap on a few things on the app.

You use WHOOP to recommend SLEEP and to RECOMMEND LOAD/STRAIN. Don’t try to use WHOOP for something it’s not intended for.

The App

I didn’t like the app when I first used it. It seemed basic. But as more data was gathered, so more knowledge and insight was returned. Many of the areas of the screen are multi-functioned so it delivers a little more than you might, at first, think with little snippets of information popping up from time-to-time about how you are progressing.

I’d personally prefer a rich web interface (there is an online dashboard for trends, I hardly used it) but I’d admit it makes better sense to interact with the app. Shame there is no Android app yet. Grrrr.


Sleep is auto-detected. It does a good job, in my opinion, and seems to correctly identify my sleep durations. I tend to go to bed too tired and fall asleep quickly. Maybe if you sat up in bed reading for an hour you’d fool it. Either way you can manually over-ride sleep times.

As you can see in the following slide show WHOOP takes some responses to standard questions as well as showing your sleep performance against what it calculated you needed before you went to bed, based on your strain for the previous day. You can ‘drill down’ to look at the sleep stats in a bit more detail.

If you’ve ever used a sleep monitor before you probably found it IS very interesting to look at all the various sleep phases and other stats. But after a while the novelty wears off. Quantified introspection takes time! Really it’s the actionable data you want ie “you didn’t sleep enough last night” or “sleep 8 hours tonight” and that’s what WHOOP delivers. Simple, actionable stuff.

This slideshow requires JavaScript.



Activity is NOT auto-detected which is a bit of a faff. For my American readers ‘faff’ means ‘annoying‘!! or perhaps more generously ‘more time consuming than it needs to be‘.

Activity auto-detection was stated by WHOOP to be delivered in early Q2.2017 and that will be a welcome addition.

You can either manually assign a sport type to an already-completed activity period or press ‘start’ and ‘stop’ before and after selecting a workout type.

So in this image I had 78% of the sleep I needed; delivering only 38% recovery. I woke at 9:00am and then I had a swim from 10:30 to 12:00 (not auto-detected).

So in this case I manually added a swimming activity over the HR track giving start & end times as well as adding qualitative impressions about how I felt I performed. These qualitative data are not used to determine any component of ‘strain’. I would, however, have hoped for an estimated figure for strain in the scenario where no HR data was gathered eg ‘flat battery’ or eg ‘forgot to wear’.

WHOOP identifies the strain automatically. It just doesn’t automatically ask you to assign it to a sport. You do NOT NEED to assign it.

This slideshow requires JavaScript.

Anyway, you enter your exercise parameters and WHOOP goes off and calculates AND SHOWS your workout strain score; in this case coming back with 11.5/21 By the end of day you can see a very, very slightly higher strain of 11.9.

WHOOP review

This is actually a nice example for other reasons. There are some peaks around 6:00pm that are probably false and also slightly raised HR around 9pm. But they make little or no material difference on the daily strain IMO. I’ve seen some comments on the net about users taking issue with the unexpected peaks. Yes the peaks are probably wrong but what’s important is the area under the curve, that is (sort of) the training load/strain. There is virtually no ‘area’ underneath those peaks so their effect is highly likely to be trivial.

Two possible downsides for triathletes in the calculation of strain are:

  1. WHOOP calculates strain (training load) from HRmax and the day’s HRrest. Creating zones based on the HRreserve. It’s neat that a varying HRrest is used but many triathletes and Garmin users are becoming used to setting zones based on the AnT/LTHR (Lactate Threshold Heart Rate – strictly the LT Turnpoint)
  2. Furthermore some athletes use different Ant/LTHR for each sport resulting in potentially very different zones. So the 140-150bpm, or thereabouts in the above chart is probably Z4 for me when swimming but for running the same HR level would be Z2.

Whilst those factors, even if included, might not make too much of a difference I would expect that most data-orientated triathletes would expect to be able to define their own training zones. Against that WHOOP might, in part, argue that such training zones are static as and are inaccurate as they do not reflect changes to your resting HR. Which IS a good point.

Indeed on this point WHOOP specifically asked prior to publication that I include the fact that they initially determine HR zones based on clustering algorithms (which include age) and that their algorithms refine the data over time, with usage.

WHOOP – APP Trends/Summaries

Trends are not shown on the app as such (see the online dashboard). However there is a limited, but useful, calendar view where each type of measurement is shown, colour-coded, for each day of the month.

The most interesting view, from my personal perspective, was the ‘Readiness’ view. In the examples shown below I am following the same training structure each week with Monday as a rest day and Tuesday PLUS one weekend day as  hard days.

Clearly the images below show that on most Wednesdays I had a low readiness (red). Tuesdays was always a fairly long run and a hard evening swim. Perhaps suggesting that I should either: train easier on Tuesdays (probably not, as I performed well in the swim); or that I should take it easier than what I did in Wednesday’s Brick training where I did often feel tired; or to improve recovery by actually following what the sleep coach suggested I should have done and got at least one hour more sleep on Tuesday night! Probably the latter!

Regardless of whether or not I made the correct training decision, WHOOP’s views of my training and of my recovery seemed to quantify how I felt.

This slideshow requires JavaScript.

WHOOP – Online Dashboard

There is also some additional, albeit limited, trending functionality online. The colour-coded red/yellow/green recovery line is platted against the blue-only strain/load line. Maybe I should have been performing high load workouts on days of green recovery?…maybe. Conversely my load/strain should have been low on the days of the 5 red dots. They weren’t.


Maybe I made some bad decisions? It is sometimes easier to rigidly follow a plan. Doing less training feels intuitively wrong.

But we probably know in our hearts that it is better to sometimes ease off. WHOOP gives you that information. Ignore it or follow it. It’s your call.


I consider myself to be reasonably aware of my sleep quality and very aware of my training load. I’ve used several disparate tools to measure those over the years.

So I wasn’t really expecting to learn too much.

However WHOOP’s recommendations for the amount of sleep I needed were eye-opening. Routinely I was 2 hours short of my recommended sleep each night. I have managed to get that down to one hour and I even scored a 100% sleep on one night (just one!). But I then do feel better for that extra sleep. Kinda stating the obvious. But so obvious I haven’t done it for years.

The other main insight comes with alcohol consumption. There have been a few occasions when I have consumed relatively small amounts of alcohol but then the next day the recovery shown on WHOOP is very small. Again it’s stating the obvious that Olympic athletes are not prescribed alcohol during training for a reason. It’s not great for athletic performance. Seeing that with your own data can be a game-changer.



So all that looks great. But how accurate is it? Indeed what is accuracy in this context?

PLEASE!!: I have to point out that I am not claiming any kind of scientific accuracy in my ‘tests’.

I have over 50 sessions’ worth of data with WHOOP alongside ‘something else’. Many of my training sessions were similar but I chose a few, below, that seemed representative of different types of training and asked WHOOP for the HR data. It was very time-consuming to put all this together but I think the results are worth it.

I’ve endeavoured to cover the normal range of my exercise-based heart rates; that’s from 120-170bpm or thereabouts. I’ve done that over the three triathlon sports and, where possible, I’ve tried to compare to two other devices one being an alternative source of optically derived HR. For the higher levels of exertion in intervals I’ve also included different types of recovery activity (standing vs. jogging vs. Z2).

WHOOP were kind enough to send me CSV files of the raw data and Matthias Krallmann, publisher of what I consider to be the ‘gold standard’ of sports file data conversions helped to get the data into FIT format that I then performed comparison suing sporttracks desktop (data conversion with the Fit File Repair Tool).

Here’s the first example comprising: Scosche (optical arm band) paired with the Lezyne; Suunto SPARTAN and Suunto chest strap; and the Polar M200 with inbuilt optical HR. For this ‘test’ the WHOOP was also worn on the upper arm like the Scosche (it’s an easier place to get better readings than the wrist).


Two things to ignore: Fristly I wore the WHOOP to run to the starting point of my test, that’s the bump at the start. Secondly the Polar tails off at 1 hour 20, that’s just because I fiddled with it.

Clearly, visually, there is a great deal of tracking between the four lots of data for the test period between 20:00 and 1:45:00

The following chart is for a more demanding session with a mile of fairly hard effort leading into 3x 1 minutes fast with different kinds of recoveries. WHOOP didn’t track quite so well here. We’ll come back to the interpretation later.


Cycling poses issues around filtering out the ‘noise’ caused by movements of both the wrist in different riding positions as well as vibrations from the road surface. WHOOP and Polar both suffered here for short bursts but over the duration of 3 hours of Z2/Z3 yet both broadly tracked the Suunto. You can consider the Suunto to be ‘correct’.


Swimming is not supported by many of the optical HR sensors. Indeed up until 2017 Garmin did not enable optical HR for swimming on any device. I’ve found that Polar’s M200 or M600 either perform very well in water OR have a ‘bad day’, The following image is a good day for the M200 as I tracked through varying intensities and cadences of front crawl for 20 minutes or so. The green line is the WHOOP this time and, again, assume the Suunto is correct.




Further test using the app for indoor cycling: I used the WHOOP app to get a feel for the ‘live’ HR and how it visually compared to the instant HR from my Garmin. There were some notable, short spikes with the WHOOP. Generally it seemed ‘about right’.

So what does all this mean?

I showed another set of my data to WHOOP and their feedback on the variation to the WHOOP data was that it was: “its consistent enough throughout such that our Strain and rest of the system will be able to identify the important physiological aspects of the workout.” Source: JC, WHOOP.

My ‘take’ on WHOOP’s view is that: they believe they are getting the ‘time-in-zone’ correct enough to meaningfully record strain/training load.

Our sceptical view might be: “Well they would say that, wouldn’t they?

So here is what Sporttracks calculated the training load to be. The load is based on TRIMP. These are some of the examples shown above and look pretty good to me.

That looks reasonable to me.

If you bought WHOOP and solely wanted to use it as an instantaneous source of HR for your sport then there are definitely LOTS of better solutions out that for that purpose. You probably already have one with the word Garmin written on it!

I am the first to admit my tests are rudimentary and non-scientific. IF you wanted to conduct a scientific experiment to validate WHOOP then my suggestion would be to try to validate the ‘overall usefulness’ through a measure of training load/strain like TRIMP.

Initially I was concerned about some of WHOOP’s HR spikes when viewed live. However, after looking at the training load calculations, such as those shown above, my view changed to becoming generally happy with what WHOOP does on the STRAIN side of its calculations.


Really we are trying to measure the accuracy of RECOVERY not sleep.

First of all put aside your Fitbits and your Garmin wrist bands. They either measure motion or simple HR and try to infer a LOT from those simple measurements. Most of them probably infer an AWFUL lot.

A simple HRrest/HRmin in the morning might generally have some value in indicating recovery. Might. Let’s leave it at that.

Here are two nights’ worth of my sleep with a simple HR chart plotted for each night comparing EMFIT to WHOOP. I’m certainly NOT implying EMFIT is the gold standard for medical-grade night-HR accuracy.

This slideshow requires JavaScript.

The general HR trend for both nights with the two devices seems similar enough.

BUT that is nothing to do with HRV.

I buy in to the broad argument that a rising trend in RMSSD throughout the night indicates a positive recovery/adaptation. RMSSD is calculated from the beat-to-beat variation of your HR ie HRV/IB/RR – whatever you want to call it.

The key thing that WHOOP has that your Garmin lacks is that it can measure your HRV at night from your wrist without the need for a chest strap.

I can’t sleep with a chest strap and years ago I used to use a 3 minute waking HRV measurement in the morning from a chest strap. this takes up a lot of time over the year and I eventually moved to QS EMFIT which involves a sensitive metal strip underneath a mattress to measure all-night HRV.

EMFIT produces a RMSSD calculation of my night’s HRV data. It then plots a line of best fit. voila. It should then look something like this:

12March 2017-A good night’s sleep with QS.EMFIT


WHOOP did provide me with beat-by-beat data to perform my own RMSSD calculations but I was unable to import it.

So I am somewhat of a dead-end for that line of investigation for the time being. Sorry.

What I do have, however, are these two overviews of trending RMSSD in February – one from EMFIT and one from WHOOP. EMFIT has a very slick interface and feels more accurate and there is more depth on analysis of the data.

However I *KNOW* that on Wednesday morning of each week I was very tired from two multi-sports workouts each and every Tuesday. The red dots on the chart, below, from WHOOP identify nearly all of these ‘tired’ Wednesdays. WHOOP also identify that Mondays were my day off in this time period, EMFIT does not measure strain/training load in any way. So whilst EMFIT might be a slicker interface, WHOOP does seem to correctly identify some key weekly moments in my training that push me somewhere close to ‘the edge’.

EMFIT – RMSSD Recovery – Proprietary Measure over Feb 2017

WHOOP – Coloured Dotted Plot IS Recovery


Pondering over WHOOP’s recovery stats I’m erring towards feeling that it’s ‘about right’.

RESEARCH: (Here) are some of WHOOP’s validations and white papers


  • Auto sleep detection worked for me but auto workout detection is needed to make the product much more user-friendly. This feature is due soon.
  • As a triathlete I expect to be able to enter my own zones and to enter different zones per sport.
  • When I manually enter a workout with the various subjective attributes BUT NO HR DATA I still expect to see a strain score.
  • An Android version.

Others could argue that better integration with the existing sports data sources like Garmin/Polar would be useful. I would disagree with that. WHOOP needs to keep its focus on doing what it does.

Others would argue that WHOOP should broadcast or make available HR data.

  • Broadcasting of ANT+ HR is unlikely to happen for hardware reasons. Although it would make it easier for a triathlete-user if only one HR device needs to be worn DURING EXERCISE. An alternative, that I don’t think will happen, could be that WHOOP links to external sources to pull-in exercise HR data from Garmin & co.
  • Sharing could happen if the demand were there.


It’s a neat product concept. I wanted EXACTLY this product 3, or so, years ago. But it didn’t exist.

Now it does! It has no direct competitors for a wrist-based measuring solution.

WHOOP bridges the gap that the likes of: waking HRV products like ithlete and BioforceHRV; and non-intrusive HRV sleep monitors like Beddit and QS EMFIT, do not yet fill. In that WHOOP provides a combined picture of recovery and strain. Rather than just one side of the picture. AND WHOOP provides that insight through a non-intrusive wrist-wearable.

In my opinion WHOOP does seem to be guiding me in the right direction towards how hard/long I should be training and how long I should be sleeping.

The sensor readings do show incorrect HR spikes from time-to-time during the day. Most probably I attribute this to either me incorrectly wearing the device and/or how WHOOP handles motion artefacts such as a twisted wrist on a bike on a bumpy road.

I’ve looked at my trainignload calculations based on HR on a near-daily basis for almost 10 years. The spikes from WHOOP are rarely long enough to have affected the overall load calculation too much from what I can see. The resultant score or recommendation would likely be little-changed even with perfect data. As to whether or not the recommendation is scientifically ‘correct’ and well founded, then “I don’t know”.

Yet many of us have looked at HR spikes on our Garmin (and other) running watches with optical sensors and bemoaned their inaccuracy. I’d suggest that is a slightly different use-case. When you are running hard for a minute then you want an accurate figure for that minute. It DOES matter to you when you are training and WOULD make you do something different if you believed the data. That minute won’t make much of a difference to what WHOOP does. BUT that will be of no comfort for those amongst you who expect a 99.9% accurate product all the time. If that’s what you want don’t buy the product.

If you want an easy-to-wear holistic solution for recovery/strain measurements and recommendations then I can see no single product alternative to WHOOP.


There are several other views of WHOOP based on studies of ‘other sports’. WHOOP is also available as a TEAM solution.




You can buy WHOOP only directly from the company at present (here).

You can use the following discount code on the WHOOP site THE5KRUNNER10 to get a 10% discount and that also helps to support this blog. Thank you!.



9 thoughts on “Detailed Review : WHOOP – an athlete recovery monitor/advisor and HRV-enabled oHRM for the wrist.

  1. Interesting stuff, well done.

    I kinda got what I think is the same feeling from my discussions on accuracy with them. It was interesting, I then brought up similar variation type questions in a separate meeting to the FirstBeat folks regarding how spikes/drops/etc might impact recovery/load scores and/or things based on HRV/RR data. In general they felt it was pretty significant when sensors (of any type) missed those events. It’s sorta like when folks have an inaccurate power meter but assume because the average is the same, it all worked out.

    I guess I’m still struggling a bit to see how the device is worth $500+ without what feels like any proven scientific data accuracy studies by a 3rd party. I think it’s one thing for a company like Suunto/Polar/Garmin/Fitbit/whomever to add-on recovery metrics onto an existing product that may be of questionable value, but then it’s a totally different beast to have a $500 product whose sole purpose in life is to create those metrics…and have them be of questionable value.

  2. this is not rocket science. its sports science. trimp doesn’t need a high level of accuracy for “strain” as you call it. as you say, to work out recovery from a trend in rmssd can’t be that hard either. ok that will require a more accurate sensor than normal but you also say that puleon do a similar one so it cant be that hard. i can’t see why whoop is not accurate enough to do what it claims providing their formulae behind the scens are up to the job. nice review thougk

  3. Much of the HR “noise” is eliminated by using their bicep strap and wearing it on the forearm (or bicep). The improvement in signal/noise is significant and is similar to the Sosche Rhythm +, also worn on the forearm or bicep.

  4. So, does the device measure your HR more accurately while on the upper arm compared to the wrist? if so, what is the possition of the sensor on the upper arm?

    • pretty much ANY optical hr sensor (including if part of a watch) will measure more accurately over the bicep. with arms by your side the sensor would be on the outside.

Leave a Reply