2013-06-30
in
7 min read
My dad had asked for a Jawbone Up for Father’s Day. I had been going back and forth on which tracker to buy between the Jawbone Up, the Fitbit Flex and the Basis B1. Since the B1 is still out of stock and my dad got the Up, I figured I’d dive in and try the Flex. I was curious how accurate it would be at tracking steps/distances and calories. By far the thing I was most looking forward to was giving indication of how well I was sleeping. While overall I’m finding the data, and the corresponding Fitbit website, useful and interesting it isn’t all that I’m looking for in a personal tracker.
My interest in biohacking and quantified self, even before I knew those terms existed, has grown as the tools available have become more affordable and easier to use. The computer geek in me loves collecting data, tracking metrics and looking for how they change over the time and then coming up with potentially correlations as to why. Something like a device which can track activity levels and the way I move over the day and night has always been a dream for someone like me. While first generation versions of these devices left much to be desired, the current crop are getting better and better. The Flex and the Up being wearable and having actual metrics, versus some sort of points metric like others have, while also being somewhat stylized fit perfectly, assuming they work well.
Two weeks in I’ve gotten used to having something on my wrist. I wear no watches, wristbands/bracelets, rings or other things; nor have I for years or decades. It therefore did take a little getting used to wearing this thing all day every day. Since I want to track sleep I even wear it to bed. It’s perfectly comfortable, but when you aren’t used to having something on your wrist, things like typing or wearing a longsleeve shirt can be annoying. This isn’t anything particularly bad about the device, it’s profile is pretty much the same as a watch, but for people like me who ditched watches a long time ago it will be an adjustment.
Setup of the device was as simple as could be. You have to create an account on the FitBit website. The supposedly let you log in through Facebook authentication or with a username/password combination. I tried the Facebook way first but I couldn’t get it working so opted for the username/password route instead. That turned out to be more than fine as the first week I had it included the night that Facebook went down for our area. The website itself is mostly user friendly. You have to nose around a bit to figure out how to adjust some customizations. When in doubt, a quick Google search saved the day but there really is no need to have to do that if the site went through some minor changes to make it more user friendly.
The site allows you to link in with other sites like RunKeeper and Withings. If you are using these sorts of sites for tracking weight, activities, et cetera, then this is perfect. Who wants to enter the same thing multiple times? FitBit has made it a point to work through an API to allow for this interoperability. Other sites would do well to follow the lead and allow for this sort of integration. I know I’m becoming more and more frustrated by the lack of it in my nutrition tracker, FitDay.
Once setup all you have to do is sync the device through the computer or your phone. I don’t want to tie it to my phone so I tried to synchronize it through the Bluetooth on my iMac. For whatever reason FitBit has decided that even if you have the appropriate Bluetooth configuration you still must use the stubby dongle that they ship with the device. That’s one strike against the darn thing since I’m low on USB ports, but I guess if I cared that much I could just have it synchronize with my phone. The charging “station” uses another USB port but isn’t needed as frequently.
To charge the device you squeeze the actual sensor out of the wristband and put it in the dock. Sit it on the dock for a few minutes while you are using the computer and it should stay charged up. I let it drain almost all the way to zero and it was fully charged in under 45 minutes. Besides charging I’ve found that I need to remove the device from the band to let it and the band dry out from time to time. It is water resistant, not water proof. Which means you can shower and exercise with it, but don’t go swimming with it on! Even from just shower and exercise it gets a little water logged. Because of that I have started taking it off when I use the shower. This has cut down on the problem considerably.
Activity tracking wise I’m pretty impressed by the accuracy. My first trial was to wear it along with my Garmin Forerunner 110 for a nice 10 mile run. The Garmin very accurately tracks distance, heart rate and therefore also calories burned. I was curious how the FitBit would compare against it. At the end of the run I came in and the estimated distance was almost spot on the distance from the Garmin. The calories burned during that period were likewise very close. Between calorie estimates in FitDay and from activities from the Garmin the FitBit website is estimating my daily calorie expenditure to within 100 calories each day, often estimating lower than the FitDay site.
FitBit does have some problems with exercise estimation that doesn’t include a lot of steps, however. Depending on what type of yoga I’m doing it sometimes doesn’t register much if any activity. A good fast paced workout with lots of motion and it comes up with a pretty good bump in calorie burn, as expected. A slow methodical workout that gets my heart rate going but has lots of static poses however, and the estimates show me as not doing much. The same is true if I am on a stationary bike or if I’m doing a swim workout when I can’t wear the watch. Some of this can be accounted for by logging the activities in RunKeeper which will then log it in FitBit (or if you don’t use RunKeeper just log it directly on the site). I can’t fault the design for any of this. It has no concept of anything exception motion, so if my wrist isn’t being moved around a lot it can’t really tell what’s going on.
This gets me to the sleep function. Supposedly it should be able to tell how well you sleep based on the motion throughout the night. You tap on the device repeatedly for a second or two and it vibrates to let you know it’s gone into sleep mode. You also get two flashing lights on either side of the band to let you know you’ve put it in sleep mode too. Based on the motion through the night it can tell if you are sleeping well or not, how much you get up et cetera. That’s the theory anyway. For me, I think it is claiming I’m sleeping far more soundly than I am. I’ve tested it out on naps where I know I’m trying to sleep but not making much progress actually conking out. According to the watch over that 45 minutes or so I slept 100 percent of the time. I know this to be false. I can put it a more sensitive mode, which I think often over-represents small motions, but even in cases such as the nap experiment it didn’t actually change things much. Again, this is the limitation of what this device is measuring. By basing it solely on motion it can’t really get any other indication of what is going on. For what it’s trying to do, it works pretty well.
In the end I’m pretty pleased with the device and will continue to use it. For day-to-day stuff it seems to track accurately enough and the website works well, especially from an integration perspective. Still, I’d like to get better metrics about my sleep patterns. For this I need a device which is taking in other measurements. That device would be the Basis B1. As soon as they come back into stock I’ll be trying to get my hands on one to see how well it works on a daily basis.