Here is how I fixed inaccurate step counting!!

[cenetti]

Hopefully it'll help others having the same problem. 

Mine was off by about 2 miles. First thing is that you need to enter your stride lenght under settings.

Here is the guide I used to measure mine, first one :

https://www.walkingwithattitude.com/articles/features/how-to-measure-stride-or-step-length-for-your-...

Secondly, if you're using it on your non dominant hand, switch it to dominant under settings. That'll lower the sensitivity. No more inaccurate steps when I scratch my .. head 

Now blaze is right on the dot. I walk 4 miles everyday, to work to gym to home. It counts as it should, perfectly. 

Hope it helps... 

[MarkWesley]

Spot.IM Conversation

• I have the Fitbit HR, I have always suspected it UNDER counts the number of steps I do. On Saturday I did a brisk walk (not running) and it recorded I was doing 1,046 steps in the first 15 mins, and about the same rates for the next hour. However I did the same first leg of that walk, with my 12 year old son, at substantially less than half the pace, a few hours later, and it records me doing an intense workout at 1,616 steps in the 15 minutes - over the same distance !!! Not only is the intensity so different, when clearly I was moving briskly on my own, and I was NOT when walking with my son, but the steps recorded on his Fitbit are always far greater than I record over the same distance !! Do you know what adjustments are possible to record correctly, my number of steps, & the speed & intensity ?? My distance is not within cooee of being correct, my son's football ground is exactly 400 metres (about 440 yards) on the inside running track, and more on the outside, hence the staggered starts on each lane - However I walk around the outside lane & have done so hundreds of times & ALL measurements are 205 to 210 metres (say 220 yards). I put this down to incorrect stride length, but no changes have helped, so I gave up. But I hoped the number of steps would be right, now seems that is also severely undercounting.

[Maryann262570]

My Blaze is severely under-counting. I've tried it on the treadmill using Exercise, Treadmill. My treadmill read 1.59 and Blaze .87. I can't remember what the second one said, bit it was way off too. I work in retail and was wondering why I wasn't getting all of my miles in! I've tried shortening my stride, but nothing works. Not happy!

[Rich_Laue]

You would not want to shorten but increase the stride. 

If the treadmill say 1.59 and the blaze says 0.87 you would want to increase the stride my multiplying it by 1.83

Of course the first thing to check is the step count accuracy, not the distance

[MarkWesley]

I did hear that suggestion before. Didn't help in my case. It does change the kms, but the step count is roughly a third of what I do.

[Maryann262570]

Thanks, Rich! I'll test it on the treadmill tomorrow. I just realized that it only started miscounting after the update. It may just be a coincidence, and may have started earlier and I just noticed it, but I do know that both steps/distance used to be accurate when I did my treadmill workouts.

[TildeK]

If it is on my right hand I need to put: wearing it on left hand? 

[MarkWesley]

I have tested mine against others in a group, doing exactly the same walk. Same speed, same distance. It is wildly out, & much worse the faster I walk. Fast walking I get about one third of the people next to me. At a slower walk, I get just over a half the step count of others around me. Got to say, I've wasted my money, & no help whatsoever from Fitbit. I'll never buy Fitbit again. Bought the wife a Garmin to replace hers. I'll replace my son's, then mine. Seriously a bit of junk & a complete con.

[PNH0]

My Charge 2 is fine for tracking steps when walking, jogging or running slowly .  It's when I run quickly (more than 5 miles an hour) that it misses counting some steps.  It's off by 15 - 20%.  It doesn't matter whether it's indoors, outdoors, on a treadmill, on a trail, a track, going up a hill, down a hill, or any of that.  It has nothing to do with the settings, stride length or any of the other things people tell you to do.  Plain and simple is that if you manually count your steps (and yes, it takes a lot of effort and concentration to do it over a long period) or you compare it to any other pedometer, the Charge 2 just doesn't record all your steps when you run quickly.

I have done this enough times to say that 10 times out of 10 I get the same results, and that is, the device simply under counts steps when I run quickly.  It's always accurate when I move slower than 5 miles an hour and inaccurate when I move faster than 5 miles an hour.

My Fitbit one used to count steps very accurately for all types of stepping activities (walking, jogging, running, hopping, dancing, etc...). 

[MarkWesley]

Finally someone having exactly the same issue. I cannot say I know exactly the speed at which it becomes highly inaccurate, BUT if I walk briskly it is highly inaccurate, if I walk slowly it IS accurate. Mine is incorrect much worse than you have found though. The faster I go, the lower the count, in fact less than half. I have counted the steps to 100 many, many occasions, & you are right it takes a lot of concentration. Also I have compared my results with people I'm doing the same walk with, walking right alongside of me, & the faster we go, the lower is my count. By accident, I have found that by walking with a heavy bag of books in my left hand (I had just been to a book sale well out of town) and the count is much more accurate. Also I found being barefoot on sand is likewise more accurate. It seems better to jog or walk briskly with a weight in the left hand, to help it be accurate. 

[MarkWesley]

I should have said "very slowly" in my post above, when I say it can be accurate. A good walk, briskly or jogging & it is seriously inaccurate. 

[PNH0]

Mark you are spot on.  The faster I go the more inaccurate it gets.  I think it has to do with arm motion.  The faster you run, the more you swing your arms.  I think it just can't handle the increased arm speed or motion.  You are right that if I don't swing my arms when running (at least the arm the device is on) the count improves.  If I run holding something with my arm either in a vertical position down by my pocket or bent by my waist but not moving too widely, the step counts tend to be more accurate.  But who runs like that?

A couple of times I ran a half mile (known track distance) and counted my steps as I ran (yes, I am that serious)... and so I know exactly how many steps I actually took (@750 give or take 30).  Then I ran the same track with my arms not swinging normally as described above (at least the arm with the fitbit) and the count was much closer to the right number (still not correct but closer).  But when I just ran normally swinging my arms as I do when I run, the count was way off.  And the faster I run, the bigger the discrepancy.

Despite what they say I don't believe the device is capable of accurately recording steps when your arms are moving that quickly.  And unless you are built funny, when you run you swing your arms in cadence with your steps for the most part.

I just came to terms with the fact that I won't be able to have an accurate step count when I run fast and I live with it.  🙂

[PNH0]

Also depending on the speed you are moving, your actual step count will vary over the same distance because your stride changes with speed or terrain (going up hill or down hill versus flat surface or treadmill or off road; your point about the sand on the beach)  So all that settings tweaking that they claim will work for some people is meaningless unless you accept the notion that your running stride length is fixed and the same when moving at different speeds or running on different terrains.

Yes it works for walking and slow jogging because for the most part your walking stride length doesn't vary that much to make a difference. 

But when it comes to running...?  I know for a fact that the device under-records my arm swings when I am running fast.  So it's not a stride length thing or even a setting issue.  Plus you can't tweak the settings to accommodate for this under-counting every time I run at a different speed or on a different terrain, etc. because it's not uniform.

I do miss my Fitbit One for this reason because I usually clipped in on my body or waistband or put it in my pocket and it seemed to work fine (within 95% of actual).  I should try the Charge 2 for a few weeks in a clip of some sort or in my pocket and see if it makes a difference.  🙂

Anyway thanks for letting me know your experiences with the device.

[MarkWesley]

Must tell you something funny. I often do a brisk walk through the Botanic Gardens in Sydney for lunch. I have noticed a few girls walking with this really odd arm movement, sort of holding their arms in a horizontal plane to their shoulders. It looks odd, & I have wondered why some do it, whereas most do not. Once this girl was walking quite quickly in front of me doing that, & she stopped at the bubbler near the half way around Farm Cove, so I also stopped for a quick drink too. So I asked her, & she just tapped her fitbit, or whatever brand she had, it was too quick to tell, & said, "it makes this count better," & she was gone. I don't know if it's true, or did someone tell her to try it, but that's what she said, that's her theory anyway 🙂

[PNH0]

Well there you go.  Perhaps we are meant to walk fast or run with that funny motion so that our counters work.  It's pretty funny when technology drives life to the point where we adjust to fit the technology instead of them making the technology to fit our world.  Ha ha!  Maybe this is the beginning of the tech devices taking over and making us all into zombies or such.  You watch those movies and think it's far fetched and then you experience this...

News flash -- Poor girl now walks funny because her device works better that way!  And the number is slowly increasing daily.  Just go check out the botanical gardens in Sydney.   LOL!

[AlejandraFitbit]

Hello all, I hope you are doing fine!

If you are experiencing issues with your steps information, I recommend restarting your trackers, walk 100 steps and let me know how many the tracker registered. Also, you can adjust your stride length information before doing this.

Hope to hear from you soon. 

[meresydotes]

@AlejandraFitbit I'm sure everyone having these problems has restarted their trackers many times, since the problems have been in place in various wrist worn Fitbits since before the Blaze was even released. Also, while it's nice to have the Stride Length set correctly, that should only affect distance measured. Steps should be counted accurately, regardless. 

Perhaps engineering should take a look at the issue.

Also, Stride Length is almost impossible to get correct, as someone's Stride Length will be different in different locations, at different times of the day, on different surfaces, etc.

I've also found that Stride Length does not help for treadmill workouts, as Fitbit trackers don't appear to be able to tell the difference between walking and running on the treadmill, especially if you do both in a single workout.

[MarkWesley]

I have restarted mine many times. Mine depending on the speed of my walking, just cannot count the steps accurately. The faster I go, the more inaccurate, the slower and it's closer. Very slow & deliberate steps it's accurate. But a fast walk, it counts about a third of the actual number. I've tested this over short & long distances & always the same, quick walking is about a third. I've actually counted the steps, and I've compared it to others I'm walking with, & always get the same, my count is about a third of those walking next to me, over the 9 kms around the Bay Run, as an example we have done many times. I have walked around my son's footy oval literally 100's of times whilst he does training, on every occasion that I've tested the distance covered, I get 200 to 220 metres, but the inside lane is 400 metres & I'm walking the outside lane which must be over 440 metres. The distance covered is bizarely out, but I would expect the number of steps to be a bit better than a third of what I'm really doing. As I said above, others have Garmin, & others I don't recall the brands, but ALL get higher readings than me, we are walking side by side around the same Bay Run or whatever walk we do that day, & always the same, the faster we walk the lower my number, the slower days & I get a bit higher. I'd love to hear someone come up with something, it's seriously disappointing. It's meant to motivate us to do more steps & a bit of competition with the wife , my son & friends to see who can do the most steps. But how can I be motivated when I'm standing next to a group, who all register say 10,000, & I'm walking alongside them & only show 3,600, & it's always the same. The only way I can get anywhere near my friends count is to let them go fast & I dawdle - & that defeats the whole purpose. Silly gizmo, it really is disappointing.

[mhaider]

---

@PNH0wrote:

Mark you are spot on.  The faster I go the more inaccurate it gets.  I think it has to do with arm motion.  The faster you run, the more you swing your arms.  I think it just can't handle the increased arm speed or motion.  You are right that if I don't swing my arms when running (at least the arm the device is on) the count improves.  If I run holding something with my arm either in a vertical position down by my pocket or bent by my waist but not moving too widely, the step counts tend to be more accurate.  But who runs like that?

A couple of times I ran a half mile (known track distance) and counted my steps as I ran (yes, I am that serious)... and so I know exactly how many steps I actually took (@750 give or take 30).  Then I ran the same track with my arms not swinging normally as described above (at least the arm with the fitbit) and the count was much closer to the right number (still not correct but closer).  But when I just ran normally swinging my arms as I do when I run, the count was way off.  And the faster I run, the bigger the discrepancy.

Despite what they say I don't believe the device is capable of accurately recording steps when your arms are moving that quickly.  And unless you are built funny, when you run you swing your arms in cadence with your steps for the most part.

I just came to terms with the fact that I won't be able to have an accurate step count when I run fast and I live with it.  🙂

---

I wrote a more extensive post but it keeps getting censored! In short, ditto for me too

[MarkWesley]

I agree exactly. That is the issue, the faster you go, be it running, or just walking faster, the lower the count & the higher the discrepancy. The moderator above has suggested I restart the Fitbit. I have restarted it many times, but deliberately restarted with his two suggestions above & counted again. With a very slow deliberate walk, I walked 100 steps & the count was 93. A near normal slow walk I got 82 for the actual 100 steps done. And a brisker walk it counted 57 for the actual 100. We've had a lot of rain here this week, but I'll test it with a few runs when the weather clears. My general experience is it counts between 36 & 50 for each 100 steps. Carrying something heavy in your left hand does get higher readings, but that may be because you have to go slower then. I find it annoying that I can we walking alongside other people who all get about the same, but I'm constantly under a half of their count. 

Also you are correct in noting that a bizarre arm movement does seem to help. You'll see above my comments about seeing just a few girls with this odd looking arm movement, of holding their arms very high, sort of level at their shoulders. I seriously just thought they were just "hey look at me" types, but a 2 second conversation at a bubler & a girl tapped her Fitbit or whatever she had, & gasped, it makes this count better. One day when I can be sure no one is within 500 metres & thus no one can see me doing such an odd arm movement, I'll test her theory. But in the meantime, I just accept, my count is about a third of reality. It still does motivate me, I've been trying since I got it to crack 30,000 steps in a day, meaning I must do over 90,000 to get that reading, but I just cannot get there. We do Sydney's Bay Run many times, on days when I've gone around twice & it's 9 kms (its actually 7 kms, but we must park away from it on our side), so I've done 18 kms, those I'm with have done over 12,000, & my best ever around there is 7,000. If I walk all day, I've got to 26,000 & 25,000 a few times, but I have to do way over 50 kms to get to that. 

[mhaider]

I bought my original Blaze 5 months ago, Fitbit sent me a replacement under warranty, got it this past Saturday and I'm basically having the same problem as the original with the same amount of inaccuracy.