Fitbit Blaze step count inaccurate

[OmarLittle]

Hey there,

Today was my first full day rocking my Blaze. I went to an outdoor festival and had some solid step count, however, when I got home and checked my step count I noticed that the blaze tracked about 1/4 the steps I actually took. My iPhone health app logged about 10K and my Fitbit did about a 1/4 of that (2500). Also I noticed it did an auto workout and thought I had rode my bike. I think this is due to my sons stroller I was pushing. Any suggestions on how to calibrate? If I go for a run and use the GPS- will that help? Thanks!

[Cybdiver]

So much effort is being put into this.  I think I've solved the problem.  I just look at what I did today and try to do better tommorrow.

So many people of so many sizes with so many different size strides and activities.  If there is a more accurate device out of the box please let me know, I'm sure I can't afford it.

I like heart rate monitor its pretty close considering it's not a 10,000 dollar EKG machine.

My favorite is when it buzzes to let me know I've done 10,000 steps it's like a party on my wrist.

I going to increase that number.

[amandagal]

"So many people of so many sizes with so many different size strides and activities.  If there is a more accurate device out of the box please let me know, I'm sure I can't afford it."

Sure, I'll name several:  The Fitbit Charge, the Fitbit Charge HR, the Fitbit Zip, the Fitbit One, the Fitbit Ultra (which is oldschool) and the Fitbit Surge.  You may not be able to afford the Surge, but I'm pretty sure you could afford all the others since they are actually cheaper.  Cheers!

[sagegoose]

I have tested this out so many times it's not even funny. It doesn't matter on a track when I count out a 100 steps or 200 steps or 500 steps, it still only counts 60% of those steps. I can't tell you how many times I've tried to change things to make it better (changing wrists, changing my stride length, resetting it, etc....) but to no avail. I have received a replacement Blaze and it's the exact same as the last one.

[arkaycee]

No, the Fitbit Surge suffers many of the same issues.

BTW, Some here (and in the Surge topic) are incorrectly thinking that Stride Length settings affect step counts. They don't. Stride Length x number of steps = Fitbit-calculated distance. So if you have tiny or big steps, the Stride Length doesn't affect how any of the Fitbits calculate the number of steps. Without a GPS or when it's turned off, that's the only thing the Fitbit has to go on to calculate distance traveled.

In short, if you walk 1,000 steps and Fitbit only counts 700, adjusting the stride length to anything won't ever make the Fitbit record 1,000 steps, though I suppose you could input a stride length that's (1000/700) times your real stride length and at least your distance would be correct.

[sagegoose]

Thank you @arkaycee!! I am getting tired of fitbit's trouble shooting telling people to make sure it's on the right arm or that stupid stride length. They have told me to check that numerous times and the whole time I'm thinking "what difference does that make?". I don't think they get the fact that the stride length has no effect on your step count, just on your distance. And since my step count is off anyways than my distance is going to be off too. So in the end....it just plain old sucks!!! I thought this new firmware update was supposed to help?????

[Barbara_F]

I was one of the earlier complainers.

I don't know if the Blaze has gotten more accurate, or I've gotten more used to how it works, but I've had significant success in getting reasonable results with my Blaze in the last month or so. I'm averaging 15K a day now, which is at least 2k more than I was averaging before changing from my beloved One.

I am treating this as motivation. I'm not looking for 100% accuracy, because I think that's a dicey proposition anyway. I'm not expecting a reasonable stride count on machines that don't require some form of impact by feet and movement of arms. Sadly, this means the bicycle is relegated to secondary exercise.

I'm getting my expected step count when walking around my neighborhood. I have a 3.9k walk and 5.1k walk, and the Blaze counts to within 100 steps. Considering I'm much fitter now, with an improved stride and that I don't expect the same exact count on each walk, I'm accepting that as a good sign.

I've noticed at least three different flash upgrades.

I don't use the "exercise" feature, sadly, because I've had problems with that.

Other than that, I'm happy with my Blaze. It's not perfect, but it does what I want/need for the most part.

[Cybdiver]

I wear mine on the left arm it's working pretty good 😊

[RealBlaze]

If your not moving your arms it does not count well.  Now if you hook it to your iPhone via GPS it will do much better.  And if you are inside and your iPhone GPS is not working well and you need use of your arms say pushing a cart in a store.  Then I put the Blaze in my pocket.  It will not read your heartrate right.  But at least your steps will be accurate.

[arkaycee]

@Barbara_FI hope this doesn't sound voyeuristic, but when you had your One did you clip it to pants or sports bra or somewhere else? I was wondering because with a One clipped to my pants (not being a sports bra wearer 🙂 ) the One was always dead on +/- 1 step, so I was surprised when you said the Blaze, even when inaccurate, was more accurate.

[Corvettekid]

"BTW, Some here (and in the Surge topic) are incorrectly thinking that Stride Length settings affect step counts. They don't. Stride Length x number of steps = Fitbit-calculated distance. So if you have tiny or big steps, the Stride Length doesn't affect how any of the Fitbits calculate the number of steps. Without a GPS or when it's turned off, that's the only thing the Fitbit has to go on to calculate distance traveled."

Unfortunately, it doesn't appear to be using the Stride Length to calculate distance -- that, or the arm motion detecting steps is way off (20-30%).

"In short, if you walk 1,000 steps and Fitbit only counts 700, adjusting the stride length to anything won't ever make the Fitbit record 1,000 steps, though I suppose you could input a stride length that's (1000/700) times your real stride length and at least your distance would be correct."

Well, that's the $64,000,000,000 question we have been asking for weeks/months:  if the distance is correctly using the default or inputed Stride Length, then why is the arm motion or steps off by 30% ?

What's going on ?  Why can't FitBit support get back to us or tell us they have re-created the problem, are aware of it, and are working on a fix ?

[arkaycee]

Exactly. I've found that having *no* arm motion (limp at my side, hand in pocket, holding a water bottle) gives pretty accurate step count. Moving my arms normally only gives me 70-80% of the steps, and therefore with an accurately input stride length (not using GPS) only 70-80% of the distance.

Part of me wonders if this is the nature of the beast and wondering if any solely wrist-based tracker is/could be accurate as to steps if your arms are moving, or if there's a way to improve the algorithm. It does strike me that using arms to measure leg movement may not be the best way to do things in general. That's likely why the One worked perfectly for me for its limited abilities -- it was on my front pants pocket at all times, and was never off by more than a step or two either way whenever I counted along with it. I switched away from the One mostly for the extra features (heart rate, gps), and also because it was black and I came perilously close to not seeing it and putting it through the laundry. 🙂

---

@Corvettekid wrote:

"BTW, Some here (and in the Surge topic) are incorrectly thinking that Stride Length settings affect step counts. They don't. Stride Length x number of steps = Fitbit-calculated distance. So if you have tiny or big steps, the Stride Length doesn't affect how any of the Fitbits calculate the number of steps. Without a GPS or when it's turned off, that's the only thing the Fitbit has to go on to calculate distance traveled."

Unfortunately, it doesn't appear to be using the Stride Length to calculate distance -- that, or the arm motion detecting steps is way off (20-30%).

...

[Corvettekid]

Ark, that's the weird thing:  you would think it would be less accurate if your arm WEREN'T moving, but that's apparently not the case.

Only thing offsetting this is that I get "steps" just from being at my PC and through random arm motion so it somewhat cancels out (undercounging vs. overcounting).

Still, I expect better accuracy for $200.  No way that the Blaze should measure a 1/4 mile High School track as 0.18 miles -- twice.

[arkaycee]

I don't know that I get 'em from typing, but I play music keyboards and playing hard will give me "steps," also enthusiastic applause will get me some as well.

[OmarLittle]

The best kind of appluses if you ask me.

[Suzeeh]

I drove five hours today and my Blaze says I walked 21,000 steps on my phone and log on the site. However on the Blaze itself it says 4,149 which is accurate. This never happened with my Flex or Zip. Very strange.

[sagegoose]

I see that fitbit has come out with a new product. Kind of funny seeing as they can't fix the products they currently have out first.

[ChrissyA]

Until they send a patch that fixes the sensitivity issue to get a more accurate step count I am wearing the Blaze on my dominant hand and I have it set up as non dominant. I have tested it and it counts 90% of my activity.

[sagegoose]

I've tried that and it didn't work for me. I've pretty much tried everything and it still comes out the same. I've even gone as far as changing my profile (my height and weight) and that didn't work. I still think it has something to do with that firmware update they had because my husband's is very accurate (within a step or two) and I'm guessing he didn't have that update because he never syncs his blaze.

[Barbara_F]

Sorry for the delay in responding, but I always wore my One on my bra. Day in, day out.

My reason for the change to the Blaze was a desire for the heart rate monitoring. I knew I'd be losing a bit of accuracy as a result of going to a wrist pedometer. They're never as accurate as something like the One. It's an inherent situation that would be hard to change.

Btw, having reread your post, it's not "more accurate" than the One. it's more accurate than it was in the past. 🙂

[sagegoose]

Okay, so I got a replacement Blaze and at first it was counting only 60% of my steps. It was accurate at that at least but for some odd reason it seems to be getting worse. Now I've been doing many step counts and every time it's counting only 52 to 53%. I would REALLY like to know what's going on here.