3 ms·
These plots definitely align with my personal anecdotal experience. I own both a Fitbit Vivo Active 4S and a Garmin VivoSmart 4. In the daytime, as a watch, I
by yodon 3y ago
These plots definitely align with my personal anecdotal experience.
I own both a Fitbit Vivo Active 4S and a Garmin VivoSmart 4.
In the daytime, as a watch, I much prefer the Garmin, but at night, as a sleep tracker, which is the main thing I want from these watches, I only trust the data from the Fitbit.
If I'm concerned about actual total sleep times, I do generally need to edit the Fitbit data in the morning. Over the years, my Fitbit watches have fairly consistently confused idle time sitting on a couch watching videos or lying in bed reading on my phone before or after sleeping as sleep. The data edit tools in the app fortunately make it easy to correct this (I'm not someone who falls asleep with the video playing, so I generally know when I turned off the activity at night and when I woke for the morning).
The data during the night, which is what I care about, I've come to trust on Fitbit watches over the years. I can't know if it's precisely accurate, but it's definitely directionally correct (if there's an overall scale factor error or similar from true sleep quality and time numbers I don't really care as the measurements are consistent enough that I can use them and benefit from them).
My Garmin I view as a random number generator as far as sleep tracking. My perception is it's probably a bit better about not misclassifying idle time before and after sleep, but not accurate enough to be actually correct, and I have not been able to find any simple data editing feature in the app to fix the misclassifications, so the wrong data stays wrong.
The Garmin data during the night I have zero confidence in, as anecdotally it feels far less "believable" than what I see reported by my Fitbit.
Edit added: another commenter just mentioned that Garmin uses a learning algorithm that needs about two weeks of data to hit good data quality. I've done a couple periods of sleep tracking with my Garmin, but I'm guessing I've abandoned it due to poor data quality each time at around two weeks. Sounds like I should give it another try and force myself to wait a month or so before making any assessments.
- onetimeuse92304 3y agoI don't have experience with Fitbit, but I have been using Garmin Fenix 7x for the past two years and I can confirm that at least in my case, its measurement of my sleep quality is only loosely connected to reality. As I am an avid runner and I train daily, I pay close attention to all sleep, heart rate, stress, and running performance numbers. Unfortunately, it will frequently say I had good night of sleep when I definitely feel groggy or the opposite. As to misclassifying idle time before and after sleep, I recognise that simple lifestyle adjustment could fix it (just don't browse on your phone before sleep!) But that of course won't fix it if you try to lie still but can't fall asleep.
- lloeki 3y agoMy anecdata is different: sleep is mostly accurate, if sometimes a bit fuzzy, and occasionally misses the mark. Regarding grogginess or lack thereof, what I noticed is that sometimes I feel that way even though the watch says I had good rest, but it's only so at the beginning of the day; afterwards I'm okay. Conversely I may have had a bad night and still feel okay, but what I noticed is that if I don't pay attention to the watch I'm actually plowing through and don't notice I'm wearing myself out, which I then have to repay the following days. Also the watch gives a) objective measurements (whereas my feeling is subjective) and b) measures only a bunch of things but not everything (e.g I noticed eating pasta or something weighting on my mind and morale can both make me experience brain fog that the watch does not account for)
- petre 3y agoDrinking one or two cups of green tea helps with my grogginess. Or some light cardio exercise like 3 minutes of rope jumping if you're more inclined to do that rather than stimulents. I also drink a shot of espresso sometimes, but the tea does a much better job without risking a headache. It's Chinese green tea with low caffeine content, Chun Mee or other blends from Yunnan or Anhui. I like their method of processing more than what is used in Taiwan or Japan.
- smallerfish 3y agoI've been sleep tracking with my Garmin Fenix 7 for about 18 months, and to me it seems bang on. The HRV stat is particularly good - a decline over about a week predicted the onset of flu symptoms a couple months ago, for example. Body battery is also a really useful feedback measurement. Sleep onset and wake times are right; REM seems right; and deep/shallow measurement at least correlates with how I feel.