Treadmill distance and Fitbit distance are off

[mkbc59]

I walked on the treadmill the last two mornings, and the treadmill is saying that I am walking about a half mile more than the fitbit is saying.  Any ideas how to fix this problem?

[whome]

Have you set your stride length in your settings? That might account for the discrepancy

[Heybales]

Walk at average speed, not fast exercise walk, so maybe 3 - 3.5 mph.

When the treadmill ticks over to 0.4 miles starting counting every right foot step until it clicks over to 0.5.

Double that number. Steps per 1/10 mile, or 528 feet.

528 / steps = stride length

Now do the same at your normal comfortable jog.

Settings page has your stride length. Default is 0, which means calculated from your height.

As you lose weight, your stride may change.

[Mystres]

I log my walking through the Fitbit app.
Enter start time , how long, and distance.
So what I entered will over ride the steps done during that time.

[FB_Opus]

I've had the same issue.  I log my morning 7.5mi walks using the stats from RunKeeper which is slightly more accurate.  My FB Flex is set to go off at 7.5 and when RK logs me at anything above that, FB does not go off until after I log the activity (it's usually off by half a mile).

FB is not perfect but when you do walk/run long enough to make it go off while still being active it makes it even more exciting.

[JRyan]

My Fitbit Zip is always off on mileage - whether walking outside or on a treadmill.  I have very carefully calculated my stride length several times and input it to my profile.  When I walk around 5 miles (outside or treadmill) the Fitbit always measures over 6 miles. I always walk at the same pace - 3.5-3.7 mph. I'd like to be able to measure how far I've walked when I am somewhere without mileage markers. What am I doing wrong?

[UVcat]

---

@JRyan wrote:

My Fitbit Zip is always off on mileage - whether walking outside or on a treadmill.  I have very carefully calculated my stride length several times and input it to my profile.  When I walk around 5 miles (outside or treadmill) the Fitbit always measures over 6 miles. I always walk at the same pace - 3.5-3.7 mph. I'd like to be able to measure how far I've walked when I am somewhere without mileage markers. What am I doing wrong?

---

Can I ask how you calculated your stride length? I did mine by walking 0.1 miles on the treadmill while counting my own steps. I did this while wearing my flex, so I could also check how accurate its step count was. I found that as long as I was swinging my arms freely (more of an issue with the flex than the zip, I would assume), it counted my number of steps correctly (it was plus or minus a few steps from my count). Great. Then, I used my step count for 0.1 miles (at about 3.5 mph treadmill speed) to calculate my step length and inputted this into my settings.

The thing I don't fully understand how to account for is that my stride length definitely varies with how fast I'm walking. In order for my fitbit to be accurate for the "puttering around" steps I take around my home or at work, I think the stride length would need to be shorter (I'm not walking at "full tilt"). But, in order to correctly capture my distance when I go for purposefull walks, it needs to use my stride length for 3.5+ mph speed. I ended up going back to my settings a few months later and adjusting my stride length down because it seemed to be giving me over-long distances on hikes (where I take shorter steps because of going uphill and over uneven ground).

If you know your zip is over-estimating distance, could you use a walk with a known distance and the zip's measured number of steps to determine your stride length?

-c

[Heybales]

---

@JRyan wrote:

My Fitbit Zip is always off on mileage - whether walking outside or on a treadmill.  I have very carefully calculated my stride length several times and input it to my profile.  When I walk around 5 miles (outside or treadmill) the Fitbit always measures over 6 miles. I always walk at the same pace - 3.5-3.7 mph. I'd like to be able to measure how far I've walked when I am somewhere without mileage markers. What am I doing wrong?

---

If you really have walked a known distance because of mileage markers, than what you've done wrong is not just get correct stride length entered in, perhaps your method of getting that is wrong, perhaps not all steps is seen, ect. Several reasons could still throw off distance even with best measured stride length.

To gain 1 extra mile over the course of 5 miles, that's a rather lot actually. So it does need correction.

Now, Zip does make it a tad hard because it has no ability to do an activity timer and separate the steps, distance, calories in to it's own record for review.

So you have to calculate it. Here's lengthy process to do just once. Treadmill best for known distance.

Note start time per Fitbit, since you'll be using it's stats.

Do your workout to some exact distance, and note the duration minutes & seconds. Because you are going to manually log your own workout.

Now go sync with Fitbit, hopefully computer account since easier to do.

Write down what the existing Steps, Distance, calories is. This contains what Fitbit thought the workout was, and everything else. We need to get out just the workout.

Now go create a manual walking activity, use the correct start and duration time.

But distance is 1 mile, calories is 10. It'll calculate steps based on current stride length, note what it came up with in the entry.

Write down newly changed daily total Steps, Distance, Calories.

So now some math with those 3 sets of figures.

Take original values minus the workout figures minus the new figures = workout figures.

So if distance was 8 originally - 1 mile entered - 2 mile new figures = 5 miles for the workout.

Do the same for steps, and calories if curious.

So now you have the steps for that distance as Fitbit saw it. And now you have steps for the distance as you know it was.

Actual Miles x 5280 / steps per Fitbit = feet per stride in decimal feet.

Decimal feet (say 2.18) - whole number (2) = inches part (0.18)

inches part (0.18) x 12 = decimal inches (2.16)

Already got the feet.

Now go change your setting. Come back to your daily total page, and see the difference after deleting your bogus activity.

You can redo that whole math section again to confirm the distance for your workout is right on 5 miles, or whatever it is.

[Heybales]

---

@UVcat wrote:

The thing I don't fully understand how to account for is that my stride length definitely varies with how fast I'm walking. In order for my fitbit to be accurate for the "puttering around" steps I take around my home or at work, I think the stride length would need to be shorter (I'm not walking at "full tilt"). But, in order to correctly capture my distance when I go for purposefull walks, it needs to use my stride length for 3.5+ mph speed. I ended up going back to my settings a few months later and adjusting my stride length down because it seemed to be giving me over-long distances on hikes (where I take shorter steps because of going uphill and over uneven ground).

---

Fitbit can read amount of impact.

Based on the stat of your weight, and stride length, it can decide if that was hard impact and long hang time to mean you ran, or low impact short hang time, short stride slow walking, ect.

So it can adjust what it uses as I call it active stride length based on impact and hang time - but it can only adjust so far from entered stride length, and it's best based on best stats.

So indeed, the walking test would be best using average non-exercise walking pace, that way it can adjust active stride down for casual walking, and up for exercise walking.

The problem of hiking is different - where you carrying more weight? That throws off that above math.

Coming down incline also is bigger impact even normal weight, throws it off too.

[UVcat]

That's interesting; powerful algorithm! When I say hike, I guess I mean trail walk: I don't carry a pack, but there's some significant elevation gain (~1000 feet in ~2-3 miles). Once you (I think?) posted a link on these forums to a nice site that would estimate your calories burned during a walk with various elevations. When I inputted a hike that I had done recently, I think I found that my flex was pretty close. Not surprisingly, it had the calories on the way up less (didn't know about climbing) and the calories on the way down more (faster pace and more impact), but such that it pretty much balanced out.

Thanks!

-c

[JRyan]

Thanks for the many helpful suggestions, which I will try. I should have
explained that I don't carry my Fitbit fob around with me all day. I only
carry it when I am "exercise walking." I wanted a device to measure the
distance I walk, and someone gave me the fitbit. Usually I walk on a
marked track - or when it's too cold, on my treadmill. I thought that I
had properly calculated my stride length, but clearly that is the problem.
I'll try your suggestions and see what happens.

[YellowFever]

I have a surge and use the GPS while walking on my normal path.  When it rains I use my treadmill.  I walk for the same length of time at a slightly faster pace, however, my distance went from 5.16 miles at the path but on the treadmill my fitbit says 3.02 miles.  My pace on the path was 3.5 and my treadmill pace was 3.7.  I've logged my stride already at 2.2 ft.  What's going on and how do I fix this.  Its hurting my stats. 

[slysam]

---

@UVcat wrote:

---

Can I ask how you calculated your stride length? I did mine by walking 0.1 miles on the treadmill while counting my own steps. I did this while wearing my flex, so I could also check how accurate its step count was. I found that as long as I was swinging my arms freely (more of an issue with the flex than the zip, I would assume), it counted my number of steps correctly (it was plus or minus a few steps from my count). Great. Then, I used my step count for 0.1 miles (at about 3.5 mph treadmill speed) to calculate my step length and inputted this into my settings.

The thing I don't fully understand how to account for is that my stride length definitely varies with how fast I'm walking. In order for my fitbit to be accurate for the "puttering around" steps I take around my home or at work, I think the stride length would need to be shorter (I'm not walking at "full tilt").

---

You can enter two stride lengths -- walking and running. If you don't regularly run or if you use a treadmill or GPS for your running distance then you can use these two strides for regular walk and fast walk. I am pretty sure my One would sometimes credit me for running when I was walking briskly. It use to overestimate and did it even after I calibrated my walking stride. This was fixed for me when I calibrated both (though it underestimates when I run since I chose a compromise between a jog and brisk walk for my running stride). I wrote a post about the method I used. I would suggest you do a longer distance like a full mile or half mile if you can to get a better average. I did mine on a track outside and it tends to be accurate outside (treadmills change our stride some, but it is usually close if I set it at my average speed). My post:  http://www.feelingfit.info/2014/08/how-to-measure-and-calibrate-your-fitbit-stride-settings/

[slysam]

---

@JRyan wrote:
Thanks for the many helpful suggestions, which I will try. I should have
explained that I don't carry my Fitbit fob around with me all day. I only
carry it when I am "exercise walking." I wanted a device to measure the
distance I walk, and someone gave me the fitbit. Usually I walk on a
marked track - or when it's too cold, on my treadmill. I thought that I
had properly calculated my stride length, but clearly that is the problem.
I'll try your suggestions and see what happens.

---

If you are using a track it chould be fine. I found the method I used (I posted a comment or few back) works fine for me when I measured it on a track. I did one mile for my walking stride to get a good average. You want to factor in the actual distance with the fitbit device counted steps. That way if it over or undercredits your steps it may still get a good "average" as long as the error is pretty consistent. I really should do this test again with my Surge, but my Fitbit distance is still pretty good (I tested on a One as that is what I wear outside of exercise).

Be sure and calibrate both walking and running even if you don't run (in that case use your brisker walking pace). I am not sure about the treadmill because originally I calibrated on a treadmill and the distance would always come up short in real life. Then when I calibrated on a track, it was fine except sometimes it is long when I use a treadmill. I think the treadmill regulates our stride so it isn't always natural (at least if we only ocassionally use it, it may be different for those who often use a treadmill).

[DominicJ]

Its not really a *problem*

It happens because to go "faster" you dont take more steps as much as you increase your step distance.

Usain Bolt has done a 100m sprint in 41 steps, more than two metres PER STEP, I can barely jump that far.

He doesnt do that distance per step on a stroll though.

If its within 10% I wouldnt worry about it.

[M0PAX]

I have a surge, walking on the treadmill i have have to hold on to the bars not able to swing arms.

I have set the stride length from default to set stride. I walk 2 miles and both times the surge has said i have only done 0.12 miles.

calories are about the same on the treadmill as the surge.

whats wrong? is it i am not swing my arms?

thank's

Graham

[Heybales]

It's not the swing of the arms. The device actually needs to try to see past
that to the actual impact of the steps.

Which a strong grip on anything (shopping cart, stair rails, treadmill)
hides that impact to some extent, a lot in your case.

You could put it in your pocket if you wanted it to still see steps and
distance over your HR and calorie burn figured from that.

If distance was seen correctly, it would likely be about the same calorie
burn.

So need to decide what's most important - HR and HR based calorie burn, or
steps and distance and maybe get same calorie burn.

Oh, stride length should be set to average daily stride, not exercise level
pace - because the vast majority of your day outside of not moving at all is
NOT exercise level pace, but daily pace. It'll adjust as best it can based
on the impact it sees - but if you start with it assuming exercise level
pace and stride length, it may not adjust down correctly to grocery store
level pace. So then much of your day (which is based on steps and distance
NOT HR) calorie burn would be inflated.

[Mr_TV]

In my humble opinion, counting steps is both very difficult and not all that meaningful (sorry).  What really matters (more) is how far you walk (steps x stride length).  You can take a step that is 1 foot or a long jump step of 10 feet,  It stills just registers as 1 step.  Every one uses different stride lengths throughout the day..

Also, wearing a pedometer on your wrist is not very meaning full either.  It should be on your ankle or at least on your hip (iPhone in your pocket) to detect a step.  If a pianist played Rachmaninoff's 3rd piano concerto with a fitbit on his wrist, it would probably chalk up 50 miles!!!  https://www.youtube.com/watch?v=9AmxZnlRa6Q

GPS versions have the best potential for accuracy but not for small distances and steps around the house, office or supermarket.

[Heybales]

---

@Mr_TV wrote:

In my humble opinion, counting steps is both very difficult and not all that meaningful (sorry).  What really matters (more) is how far you walk (steps x stride length).  You can take a step that is 1 foot or a long jump step of 10 feet,  It stills just registers as 1 step.  Every one uses different stride lengths throughout the day..

Also, wearing a pedometer on your wrist is not very meaning full either.  It should be on your ankle or at least on your hip (iPhone in your pocket) to detect a step.  If a pianist played Rachmaninoff's 3rd piano concerto with a fitbit on his wrist, it would probably chalk up 50 miles!!!  https://www.youtube.com/watch?v=9AmxZnlRa6Q

GPS versions have the best potential for accuracy but not for small distances and steps around the house, office or supermarket.

---

Well, you actually described what Fitbit is doing, except for the wrist-worn is attempting to see impacts.

Your weight and the stat for stride length give an expected impact and hang time for an expected step.

The actual impact and hang time seen is compared and gives an instant dynamic stride length for that step.

That's why with a properly setup stride length, you can walk a mile at different paces and stride lengths and it'll still come up with a mile.

Best practice therefore is for stride length to be middle of daily paces - not grocery store shuffle, not exercise level pace. Middle of average daily pace so it can dynamically adjust up and down pretty accurately.

On the wrist it's trying to see those impacts despite the swing of the arm, it can be really well done.

Ankle is actually worse unless running, it usually won't see opposite leg impact since with walking both feet are actually on the ground for each impact.

And the "impacts" from false steps playing the piano, which wouldn't all be registered anyway, would be seen as such a minor distance, and therefore calorie burn.

These are more than just pedometers with the accelerometer chipsets that have been available for some years now.

[JRyan]

Can't believe my inquiry is still getting responses! Gave my Fitbit back to my daughter, who had given it to me as a gift. I just try to walk/run by time and, if able, by distance. 5 miles. I found the whole thing about steps to be - I want to say "dumb" but I'll say instead not useful for me.

Sent from my iPad