The speed at which you turn around such thorough testing is unbelievable. Been waiting for this one before I part with my money. Best wearable channel of all time 👏
@@TheQuantifiedScientist Rob, hi! I just wanted to know if you have the Muse S video available yet. I think other people are interested to see it too. You are awesome!
@@eXqmpIe Hi Luka! Not yet. It takes much more effort to test this device, since I first need to build the reference PSG device. This is already partially underway, but due to circumstances was put on the backburner for a bit. I will continue with this though!
@@TheQuantifiedScientist thanks for your great work! I have been looking through your videos and finding it as one of the best sources for systematic and scientific examination of fitness and health trackers. I already asked this in another post, but do you think you would find time to look at the asus hc-a05 it is a very interesting piece of hardware that is not much covered in Europe
Hi Rob, you should make a video on which is the most accurate fitness tracker. This would be more beneficial in helping people choose a fitness tracker. Your videos are amazingly well done though. Thank yoU!
I've watched a lot of your videos - I f****ing love them! Do you have a ranking of all the devices you've tested? That video would be aaaawwweeesssooommmee!
Bought the Fitbit charge 5 because of this channel. Sleep tracking is the most important to me. Rob testing out wearables in such a scientific approach really helped a lot in my decision. This channel deserves more subs!
I've found that if you change your setting on what wrist you have your fitbit on in the fitbit app and put it on dominant either if you have it on right or left it will help to prevent overcounting steps. "For wrist-based devices, it's important to specify whether you wear the device on your dominant or non-dominant wrist during setup: The dominant wrist setting decreases the sensitivity of step counting and should reduce any over counting of steps when your body is not moving. Your dominant hand is the one you use for writing and eating. The non-dominant wrist setting increases the sensitivity of step counting and should reduce any under counting of steps." This is from fitbit site and I choose the dominant setting even though I use it on my non dominant hand because I use both hands pretty much every day.
Having worn my Charge 5 for just aweek or so now, I find it very optimistic about step counts over the whole day. Probably not too bad for the periods when I'm actually walking. Compared with the step counter on my phone (which I'm inclined to believe) Monday Fitbit 19164 steps 14.38km / Phone 14554 steps 10.35km Tuesday 22921 17.16km / 15683 11.62km Wednesday 21608 16.21km / 16832 12.06km Thursday 19349 14.51km / 14787 10.82km I haven't yet worked out exactly where it is overcounting so badly. For now, I'll rely on the phone for steps and distance, and use the fitbit for everything else that it does. (Fitbit on my right/dominant arm and set accordingly i'n the app.)
Man... the amount of effort you put into this, much better than the average "review" that just states superficial qualities, I really respect you for providing non-biased reliable reviews.
I recently bought a Charge 5 and get on well with it. I am not looking for massive accuracy and won't lose sleep if my step count is a couple of hundred off accurate over the space of a day... I bought it to help with general awareness of my activity and just so I can see in relative terms, what I am doing and how much I am moving around and it does it well. Great to see someone having taken the time to do all this research, very interesting to see!
Hi Rob, I was wondering if you have ever thought about doing a series comparison series? For example, I would be interested to see how the charge 5 compares to previous iterations of the model. Has the sensor accuracy improved? Or have the sensors remained the same with updates to the software?
If we didn't see your sleep tracking results I wouldn't believe you ever sleep! I don't know how you make such comprehensive and thorough videos while being so active and upbeat. Thanks for all you do for the rest of us mere mortals!
I hate to tell ya this, but it looks like the Polar device is the one reporting inaccurate numbers. There's no way your hear rate went from 160 down to 65 in less than 1 min according to the chart at 14:43. That's like going from full sprint heart rate to chilling on your couch heart rate. Your heart rate doesn't just bottom out as soon as you stop moving. The 160 to about 115 in less than a minute reported by the fitbit seems far more valid.
I have a Fitbit Sense and it isn't even close. I have had nothing but problems with it, first one I had wouldn't even turn on and it was fully charged!
Interesting and informative video, loved the scientific approach. I just wanted to make you aware of a slight limitation to your heart rate testing if you were not already aware. ECG monitors calculate heart rate based on the electrical activity it senses and I think most fitness watches and finger probes sense the pulsatile flow of blood. The first small issue is that ECG devices can misidentify ECG distortion/artefact as a heart rate but this is well known and difficult to avoid. The other slight issue are ventricular ectopic beats, which are extra beats from the bottom chambers of your heart. Ectopics are seen in normal healthy hearts but can be associated with various conditions. It's worth noting that ventricular ectopics produce a very weak or no pulse (because the lower chambers have not been able to fill with blood) which usually does not produce a palpable pulse in the wrist. This means the ECG monitor counts the ectopic as a beat which the watch or probe is unable to sense and therefore does not count. As a side note I have looked after cardiac patients in "ventricular bigeminy" (every other beat is an ectopic) and the ECG monitor reads heart rate 84 bpm (for example) and the probe on the finger reads the heart rate as 42 bpm (this is an extreme example of this mismatch). Overall these are generally just small limitations but are dependent on how many ectopics you have and how much artefact there is in the ECG.
Great detailed analysis! Thank you. I bought a Charge 5 last night and used it this morning for HIIT on a Stair Stepper machine, while comparing it to results using a Polar H9. I was very disappointed to see a large lag between my HR as measured by the H9 and the Charge 5. When on a high intensity segment my HR would shoot up quickly above 150, but the Charge 5 would stay back around 130-134. Then when on a rest segment when my heart rate would get down to about 140 the Charge 5 would then catch up and then the two reading would be the same as HR continued to drop. I saw the same thing for six repetitions. My conclusion is that the Charge 5 will not read sudden and large increases in HR. This makes it useless for those wanting to do careful HR Zone training.
Your videos are so awesome. Nothing compares to your scientific approach in helping to decide which device to buy. Comfort, charge time, waterproof and things like that are helpful, too. But so far, I find no other videos comparing the actual accuracy scientifically like you do ! Awesome!
Just found your channel today, you're doing a superb job! What bothers me about Fitbit is that they require to be a premium subscriber for certain features and a lot of the sensor data cannot be exported at this time, not even via the developer API. It's really sad, because I'd love to see how the SpO2, skin temperature, EDA and 1-lead ECG sensors are performing (and of course, use the data myself, if it performs reasonably well).
Same. My Charge 3 is dying so I'm looking to change trackers because of this. Right now I'm mostly interested in sleep tracking and it seems like they are one of the best out there but I'm definitely not paying for a subscription to get my own data...
I recently watched a video testing of the Charge 5 where it was disclosed that during a run the sensors of heart rate and gps are i conflict, since the gps (due to the antenna location) only works with the watch band rather loose around the wrist while the heart rate sensor requires the band to fit rather firmly around the wrist. Have you noticed anything like that? Thank you for your nice job.
Great job, I was waiting for it. But I didn't expect it so soon. I'm already looking forward to the sequel. I have a Fitbit charge 4. I bought a better metal magnetic bracelet for it, it holds better and thanks to that the measurements are also a bit more accurate. Most silicone bracelets do not last longer than 6 months and are less comfortable to wear. I will not buy a new one at the moment, but I would do it in about a year. Thanks for testing ROB!
you are totally right, I don't see people saying that much. I have always gotten a 3rd party metal magnetic snap wrist band when available for my fitbit. They are more comfortable for all-day/night use with a few breaks here and there. And I THINK it may be more accurate but at least its more comfortable, eh?
Hi Rob,. Thanks for your excellent analysis series. In preparation for buying a new fitness tracker, I watched probably 10-12 of your analyses. After consideration of cost vs. accuracy vs. battery life, I got a charge 5 from Amazon. I'm pretty happy with sleep tracking but I notice what seems to be a slow HR response to increased exercise intensity. I have a polar F6 and will compare the F6 and the charge 5 when I get new batteries in the F6. I also note slight overcount in steps against actual count. Overall I am happy with my choice of the C5 thanks to your excellent review. Rick
Update to the review above. After 5 months with the charge 5 and adding an h10 to my arsenal, I've found that the charge 5 has significant lag in picking up an increase in heart rate; around 1-2 minutes. It also lags in picking up a decreasing HR. In between it was fairly close to the h10.
I have had 2 Fitbit 4. Within a 7 months, with the 1st one, the band broke. I made such a fuss with them (disabled Navy vet) that they sent me a new unit, no charge! In April I was camping in the Texas woods. Swimming in a deep spring lake. Guess what happened? YES! The band broke and the whole unit sank into the bottom of the lake! Fitbit was important to me and my doctors; I suffer from painsomnia; due to a genetic disease, which is now under treatment - because they saw the data, I was taken seriously. Before the Fitbit, the doctors dismissed me and wanted to send me for a psych eval! I don't know what to do; get the Charge 5 and duct tape it to my arm, or try another product. Any recommendations?
Excellent review. I really appreciate the time you put into giving your viewers in-depth information. The inaccuracy of the heart rate bmp and the inconsistency made my decision to return the Fitbit Charge 5. I could of accepted it being off 10 bmp but 35 to 47 bmp either to low or to high was unacceptable. It’s unfortunate because I liked all other things about the Fitbit.
I returned mine after discovering it couldn't be relied on to record an activity in the rain. Wet clothing sends the touchscreen into meltdown prematurely ending or cancelling a run or a ride. Shame as I really like the features and the design.
Another terrific video, thanks for all the hard work you put into each one. I look forward to the update as I am very interested in whether this (or any other band/watch) would be accurate at recording ECG, especially as from both your channel and personal experience I know that these wearables are pretty hopeless at SPO2 due to the restricted way they have to work as light can't be detected through the wrist, instead having to use reflected light. The fact Fitbit was forced to release this without one of its main features makes me concerned that ECG might be unreliable too. BTW, having to pay Fitbit a monthly fee to access the full sleep tracking data is outrageous, but at least ECG is not behind a paywall.
Thanks for your excellent reviews, they really do help, especially for those with sleep apnea like myself. However, I went to check the sleeping cycle confusion matrix posted in the Charge 4 review, since I have that tracker, to see if the Charge 5 has any improvement in the accuracy of sleep tracking and I found the two matrices not comparable since you changed the basis of how the percentages are calculated in each. I would, therefore, appreciate if you could comment on how the two trackers compare in sleep tracking and whether getting the Charge 5 for that purpose would be warranted if I already have the Charge 4, thanks!
@@krisbarton3249 What I'm referring to is the way the percentages in the confusion matrix were calculated and that changed between the two reviews so the results are not comparable.
I've been using the Charge 5 a few weeks now what I do like about it is the readiness score, I'm currently undergoing investigation for Rheumatoid arthritis and the readiness score show's when I have dips that coincide with the weather and how I'm feeling on the day without reporting it to the device.
Would be great to see a similar review done with the Pixel Watch - apparently the Heart Rate sensor is better. Keen to see if the performance during weight training improves.
Man, Man, Man!!! Don't have enough words. Really really really appreaciate your efforts. RU-vid, where was this channel till now?! Glad I found it now, best channel for wearables/trackers! Thanks much, scientists.
Please create video on ranking of fitness bands based on their accuracy. I am trying to find it through your all videos but in every video u have explained individual bands only. Please create a ranking video
I haven't bought the series 8 yet but your video showed exactly where my interest in the Apple Watch was, namely fitness and sleep. I now feel justified in spending my money on this watch. Thank you.
I got my Fitbit Charge 5 yesterday and had my first strength workout today. With each set of exercises my Garmin Forerunner with chest strap showed a heart rate of up to 155 to 162 bps. The Fitbit Charge 5 usual just showed 111-120. I don't know what to think about that and i am very disappointed. I know that a chest strap is usually more accurate, but the difference makes the Charge 5 unusable for tracking heart rate and strain during workouts
Dude your channel is amazing! The amount of work that you put on your videos is mind blowing 🤯 In your experience would you recommend me the Fitbit charge 5 or the inspire 2? Should I wait for the inspire 3 review?
Thank you. You gave me confidence in my charge 5. The little idiosyncrasies are not off-putting. I did wonder if I actually took all the steps, though, ha,ha.
Год назад
I guess the variant mismatched may be came from the on/off delay timing of sensor for capturing, it may impact to the battery life of device if sensor on in high frequency.
i just got my Fitbit Charge 5 and seems to work very well for me and this is my first tracker. I am not that fussy when it comes to stats so I take everything with a grain of salt. I also like the fitbit app which works on my Motorola android phone. It is nice to know what my resting heart rates are and also get periodic SPO2 readings which show I am doing very well. So far I have done 1 sleep cycle and see how it compares as nights go on. Will get into exercise modes in the weeks to follow. I got my Charge 5 for 119 dollars on sale at Fitbit. Normally 180 dollars. I will not get the premium option which is 79.00 annual cost. Things in the premium do not seem to interest me. Thanks for your scientific video on the Charge 5. take care. All in all I like the Charge 5 for watching my normal everyday activities not counting exercise modes - step counting, heart rate, and periodic Oxygen levels. i do have better heart rate and blood pressure devices to compare off and on. It may not be for others, but works for me. take care.
Hi Rob for your professional testing of the Charge 5. I find that mine is very accurate in heart rate sensitivity. I am 78 year old male,6 ft in high tans weigh 184 lbs (BMI 24) and am reasonably fit. I walk 60-80 km a week at a good pace. When walking I swing my arms over a 90 deg arch and have a cadence of 60 steps per minute. When walking my Fitbit Charge 5 shows a heart rate in the 104-106 range. When I time check my heart rate it is generally 72. Interestingly if I stand still and just swing my arms at the same rate I get a reported heart rate of 106-120 and time check it at 60-65. Obviously if the reported heart rate is in that rate of error so are the calculated values for zone minutes. I find that information totally unreliable. The heart rate is generally very close to the time checked rate when at rest. I have written Fitbit about my observations but received no response.
Great work and videos. Saved me spending crazy money for what appears to be a very expensive inaccurate toy! How do they get away with misinforming people on serious health related statistics. Regulation needed!!
You did an excellent review. I'm older, 65, and working on losing weight. I am interested in which Fitbit to get to best track steps & other exercise. I'm less interested in sleep tracking bc I figure it is what it is & will likely improve as I lose & get more exercise.
Thanks for these reviews. They have helped me build a set of items to track various things. Apple Watch for heart rate and workout HR, Fitbit 5 for sleep
Fantastic work from a real conscientious scientist. I already decided the Charge 5 was not for me because I'm not interested in sleep tracking and I am interested in accurate HR. I am now about to watch your Garmin videos to see if there is an alternative to the chest strap.
I have Fenix 6 and wife has apple watch 6. Both are garbage and give different readings than my polar chest strap. Garmin has a crap way of maintaining it's data. The apple watch is nice but it's more of a smartwatch than a fitness tracker. I'm still looking for something myself
what about ECG, can you cover about ECG accuracy ( not only this device, any watch you would prefer ), please help us to know that, can we trust ECG in smartwatches...
Hi Rob, I’m new to your channel and love the content. I currently own a Fitbit Charge 2 and am thinking of ‘upgrading’ to the Charge 5. I was wondering if you’d consider testing the Charge series and seeing if they’ve improved over time. This would aid anyone using the Fitbit ecosystem to make the best upgrade choice. Regards, Phil, UK 🇬🇧
Really love the videos. This is what is needed and really much better than most of the other reviews that just tell you what is on the spec sheet which is basically pointless.
Thank you for the review. I would like to see both from reference methods how good fit for the sleep analysis. Due to the accuracy of the device, there would usually be slight deviations.
I'm looking for a daily heart rate tracker as good as the apple watch but for Android, do you have any recommendations? I found your channel by accident right before i ordered the charge 5, I decided to wait for your review and I'm glad I did. Thanks for the work you do!
Thank you for your review’s! I have the Fitbit 4. I appreciate the heart rate review about its highs and lows, it freaks me out sometimes at night when it shows my heart rate in the low 40’s then it would spike to 120. My husband told me not to freak because he doesn’t thinks it’s accurate. So thank you again for confirming that for me. Maybe I’ll sleep better now 😉😊
Different instances of the Charge 5 perform differently. I swapped out a Charge 5 with hopeless GPS for one with really accurate GPS. Product consistency would be good to measure.
Thank you so much for this video! A lot of the time people just give their personal opinion about how they experienced things like usability etc. and while that's certainly valuable, I missed the more objective comparisons. I really admire the amount of time and effort you put into this! As a fellow science-minded person I can definitely appreciate this, and it gave me exactly the information I was looking for! I was wondering if you could give me your thoughts on something. For me the two most important metrics are sleep and heart rate. I've been using a Fitbit Charge 3 for the past two years, which is my first and only wearable up till now. But I need a replacement. I don't have the budget for anything more expensive than the Charge 5, and generally I like to stick to the brand I'm using because that way I can keep all my data in the same app and have at least some consistency from one device to the next. Because of ME/CFS (chronic fatigue syndrome) and a bad back I'm mostly sedentary unfortunately and I often don't even leave my one-room apartment for days on end. But to get a little more active I've taken up modern dance once a week. My heart rate tends to spike up really fast, and a warming-up dance can easily get me to upwards of 140. Since you spoke about a lag in the registration of your heart rate, do you think that a quickly changing heart rate might pose a problem for the Charge 5? And considering the difference in accuracy between workouts, how do you expect it would perform in modern dance? It's mostly upright, sometimes we get down to the floor or jump but it's not the main thing. And would you say the Charge 5 has significant benefits over the Charge 3 when it comes to sleep tracking? Sleep is the bane of my existence, since I never wake up fully rested no matter what I do. I've had that all my life, and aside from an ME/CFS diagnosis doctors haven't been able to do much with it. I hope the Charge 5 could give me perhaps a little extra or more accurate information to work with so I can maybe distinguish a pattern myself. I would be very interested to hear your thoughts! Thanks in advance and keep up the great work!
Hi Merel, i developed CFS too and tested as Cardio and Health coach Xiaomi/Fitbit/Huawei Devices too. As Rob mentioned in his Videos, nearly no tracker detects HR spikes well. Everything that includes movement of arms and Wrist as well as gripping sth isnt good for 95% of the trackers. Only Apple watch and the new Huawei Watch 3 GT (Robs new test) are pretty good in HR. Maybe for you would Pricewise a priceworthy Garmin Watch (venu2 maybe) + Heartbelt for the Dancing Time a good idea. The fitbit isnt compatible to Heartbelts. I like Sleeptracking also but Healthwise to be honest, its a Gimmick. What counts the most, is the starting time (constantly), duration, and most important the feeling at waking up. At the moment there is no device that can track sleep & hr equally well. Fitbit is generally very good for sleep tracking, but the charge 4 is practically equal to 5. If you do Breath excercise, which is good with CFS, you should be able to get a feeling when your stress / pulse goes up faster than the most trackers. Im interested in Charge 5 too, but the most features , even measuring and looking back some days, are only aviable with Premium subscription for 8€/month.
Got Charge 4 a few month ago. This thing has no problem showing me in any stage of sleep while watching tv 1.5 hours. The rest is a combination of X% of light Y% of deep etc. and very similar day to day. So if have to tell it when I go to sleep and get up, apply an algorithm and that may be the most of the data. Some measurement may be a small part of this. I have no way of actually measure what is happening, but also point out that if you are healthy and predictable, the result may be close to right. Give the unit to someone who is not well, light sleeper etc. and then it would really show if the result are data driven or a good guess.
much anticipated review. thanks you! I think I'm going to wait longer and watch more reviews but I'm considering upgrading to the charge 5 from my inspire 2.
Impressive and thorough testing, thanks so much. Your work confirmed some of my anecdotal observations/suspicions (mainly missing heart rate peaks and somewhat overcounting steps) , but overall, I'm happy with the Charge 5.
Thank you @QuantifiedRob! We see glimpses in your videos of data that you've captured from these devices through vendor API's. Have you considered doing a video about these API's and the tools you use to process the data? I think many of us would find that interesting! It looks like you're a Base-R fan!
Congratulations on your hard work. One of the few I trust. Comparing the Fitbitcharger 5 device: Zmax EEG Regarding Deep FitBitcharger 5: 92% accuracy, deviations considered light 7.8% and 0% Rem 0% Awake. Regarding Light FitBitcharger 5: 61.8% accuracy deviations considered 19.1% as deep 13% rem and 6.1% awake Regarding Rem Fitbitcharger 5: 66.1% accuracy deviation 23% as light 0% as deep and 10.8% awake Regarding AWAKE Fitibitcharger 5: 94.7% accuracy deviation 0% rem 0% 0% light and 5.3% as deep. Dream2EEG Regarding Deep FitBitcharger 5: 96.2% accuracy, deviations considered light 1% and 2.9% Rem 0% Awake. Regarding Light FitBitcharger 5: 79.5% correct deviations considered 15.5% as deep 2.7% as rem and 2.4% as awake Regarding Rem Fitbitcharger 5: 50.2% accuracy deviation 23% as light 0% as deep and 10.8% awake Regarding AWAKE Fitibitcharger 5: 89.5% accuracy deviation 40.5% rem 0.6% as deep and 8.78% light Based on your testing you can control variables without third party data. Technically, the fitbit charger 5 should have a better score. If you also consider the smart wake of the fitbit charger 5 it would be extremely accurate since it can achieve zmagx eeg 94.7 for wake and 92% for deep i.e. for this feature it would be extremely accurate. You could test for agent.
Rob! Great video as always, I usually look for your perspective on these things, and would love to know what you think of the HRV capabilities of the Charge 5, thanks!
it lost only for 1 to 3 month, though i got replacement throughout the 1st year 5 times but still at the end of the year i got this device stopped working and did not get the replacement 6th time as the warranty expired. My comment is unreliability got into heavy dosage.
The reason the GPS is not working during some testing is because it needs to be loose on your wrist to pickup GPS. The antenna is under there and if the watch is too tight you won't have a GPS signal.
You should test keeping it in pant pocket for step counts. This tracks motion of leg. I found overestimating step counts dropped when i put Charge 2 fitbit in pants pocket. It's also terrible during bike rides and cannot be worn during bike rides as it ruins accurate step count. Also, some have noted issues with GPS when using HR at same time.
