Message Boards » Fitbit Users » Discussion

TOPIC: Fitbit Flex Way Undercounting Steps

 
Ic_disabled_photos
Topic has been inactive for 30 days or more and images have been disabled.
Display All Images
June 8, 2013 8:02 PM
I have had my new Flex for just 3 days after waiting forever for it. I have to say that I am very disappointed and disheartened as my steps per day have been grossly below my usual. Today, I wore my Fitbit ultra as well and last I checked my ultra was 5100 steps and the Flex just 3300! I am wearing the Flex on my left, non-dominant hand as my settings reflect. Any ideas or help?

Thanks.

Hannah
  13641631
August 15, 2013 9:38 PM
Hi,

I am facing the same issue with my new Fitbit Flex as well. Were you able to find a solution to this problem pl.?

Thanks,
Armin
August 17, 2013 5:34 AM
And my fitbit flex is not recording any activity at all. LIke all I do all day is sit still. It is on my non dominant wrist.
  24894917
August 19, 2013 10:56 AM
I just posted in a related topic - same issue here. This was my post - and no solution yet.

Has anyone else counted your steps to determine an average, and then compared it to what Fitbit Flex calculates? I have a pretty large step when I go on walks, and average 105 steps per minute. When I look at my Flex data when I get back, I generally only get credit for 3/4ths of my steps, or between 70 - 85 steps per minute.

A similar but different wrinkle is that my main exercise is the elliptical (it is a cross trainer with the handles that move for each step). Due to limitation on the machine, my step is 24 inches (smaller than my walking steps). It is slightly more accurate, but whereas that elliptical step averages 120 - 140 per minute, the Flex only counts about 100 - 110 per minute.

I don't know if mine is a faulty unit, or if others have experienced similar problems.
  7657856
August 23, 2013 8:38 PM
I sent an email to fitbit and asked about this. Mine is under reporting about 25%
September 9, 2013 8:19 AM
I tried wearing it on my non-dominant hand (my right hand, as I am a lefty). The Flex seemed to under-report my steps by a wide margin, compared to my pedometer (a New Balance Via Slim: seems fairly accurate). The flex gave me credit for around 16000 steps, while my pedometer had me at around 25000 (sorry, I didn't record the actual numbers). So it was only giving me credit for about 65% of my steps, assuming my pedometer is accurate.

Today, I've moved the Flex to my left hand, and so far, it seems to me more accurate. Here are the current readings (almost 1 pm):
Flex: 10134
Ped: 12776

This is about 80% -- still well under what I would expect. Is this just a bad unit, or do all the Flex behave this way?
September 9, 2013 8:57 AM
I find no problems with accuracy and I've had mine since it came out. One thing I'd ask, have you manually set your stride data on Fitbit's website or are you using what it's calculated for you? I know a fair number of folks have addressed an accuracy issue and this is always one solution that comes up.
  11948731
September 9, 2013 9:18 AM
I haven't had any issues with accuracy, I've only had it about a week and a half and wear it on my dominant hand.

To also double check the accuracy, I've had the iPhone app open while walking to watch the steps increase. I've increase and decreased speed while monitoring and it's bang on.

Hopefully you can find out what the problem is!
  18890694
September 9, 2013 1:51 PM
Have you guys measured your stride and all that?
September 9, 2013 6:14 PM
QUOTE:

I have had my new Flex for just 3 days after waiting forever for it. I have to say that I am very disappointed and disheartened as my steps per day have been grossly below my usual. Today, I wore my Fitbit ultra as well and last I checked my ultra was 5100 steps and the Flex just 3300! I am wearing the Flex on my left, non-dominant hand as my settings reflect. Any ideas or help?


I don't have a Flex, I have a One. But what happens if you try keeping it on "non-dominant" setting but actually wear it on your right hand? I think that should in theory be the most sensitive setting. I think the "dominant" setting is intended to filter out extra movements that are not quite steps as it is assumed you move your dominant hand more. By extension, the non-dominat wouldn't filter some of the movements as much since people move their non-dom hand less. So, for some people, a non-dominant setting plus wearing it on the dominant hand should over count steps--since it is undercounting for you it might actually end up more accurate. ??? That is, if you can stand to wear the Flex on your right hand.
January 29, 2014 4:44 AM
This problem has carried over into the Force but, perhaps, with a twist. I have run many experiments with my force and have found it to be quite accurate at counting steps when walking slow or running at almost any speed I am capable of. HOWEVER, when walking at 4mph, the Force consistently under counts my steps by 20-30%. I know it makes no sense which is why I have run MANY experiments on treadmills, indoor tracks, and outdoor paths. The error is consistent and too large to ignore.

Several emails to Fitbit have resulted in responses that do not address the issue (like change your stride length) or simple denial.

When my Polar Loop and I get back to California tomorrow, I intend to do some bench marking of the two devices against each other. I dumped the Loop because it would not share data with MFP. Now I am reconsidering.
  43898712
January 29, 2014 8:12 AM
I have done a test on mine too - with my Iphone app on and it was also right on. I tested it while driving and no steps recorded. I tested it while sitting at my desk typing and no steps recorded while I was working at my desk for a hour. So I'm not sure what to tell people who are having problems. I have mine set to non-dominate hand and I wear it on my non-dominate hand. I left the stride as it was defaulted to since it seems like it is accurately counting my steps. I take mine off for showers because I just don't like wearing anything while showering jewelry wise, I don't put it back on until I get done with my hair so even though I know I lose a little steps right there, it doesn't bother me I still feel like I'm getting the calories burned right in order to create my deficit. I also quite wearing mine at night just because I really don't care what my nighttime activity is, but I do set it to sleep mode and then take it off and put it by my bedside dresser and then wake it up after I get dressed for work and ready to walk out the door the next morning.

Also, I don't log any activity manually, I let the fitbit do its own adjustment on MFP. I think it just makes it easier than trying to figure out if its double logging, etc. I guess I feel if I just wear it all day and let it do its own logging of my calories burned versus my food logged in to MFP that at long as I hit the "Nailed It" on my fitbit at the end of the day I should be pretty accurate on creating a deficit. Oh I also did set up my fit bit to lose - I think at the most agressive setting which should be a deficit of 500 calories a day. So if I am over by a bit - a hundred calories or so, I'm still creating a deficit.


So far I love mine and have had no problems with it.
February 1, 2014 10:07 AM
QUOTE:

This problem has carried over into the Force but, perhaps, with a twist. I have run many experiments with my force and have found it to be quite accurate at counting steps when walking slow or running at almost any speed I am capable of. HOWEVER, when walking at 4mph, the Force consistently under counts my steps by 20-30%. I know it makes no sense which is why I have run MANY experiments on treadmills, indoor tracks, and outdoor paths. The error is consistent and too large to ignore.

Several emails to Fitbit have resulted in responses that do not address the issue (like change your stride length) or simple denial.

When my Polar Loop and I get back to California tomorrow, I intend to do some bench marking of the two devices against each other. I dumped the Loop because it would not share data with MFP. Now I am reconsidering.



OK. I'm back in California and have run a series of experiments. All were conducted on a three mile outdoor course while wearing a Fitbit Force and a Polar Loop on my right (non-dominant) wrist. The goal was to cover this course at three different speeds and compare the step count registered by the Force against the Loop. A further goal was to calculate the indicated step distance (distance covered / steps recorded) and test for reasonableness.

The net result is that both activity monitors do a good job at 3 and six miles per hour. The Force totally falls apart at 4mph reporting 3,490 steps yielding an ABSURD 55 inch step length compared to the Loop's 5,585 steps and 34 inch step length. This is quite consistent with a lot of other less rigid experiments I have run. Several emails to Fitbit have resulted in nonsensical responses. They have gone quiet lately. Hopefully that means the issue has finally been kicked upstairs to the adult leadership and the problem with their algorithm, will be addressed in some future firmware update.
  43898712

Message Boards » Fitbit Users

Posts by members, moderators and admins should not be considered medical advice and no guarantee is made against accuracy.