Distance calculated using running stride

[yurek99]

I just stared using fitbit and noticed my daily distance is calculated based on a running step lenght (~90cm) although last time I run, it was 7 years ago. My walking stride is about 67 cm (I think, it was calculated based on my height), so why Fitbit is adding to my distance 30%. I don't need more fake news!

Any sugestion what I'm doing wrong?

Tnx

UPDATE:

Today I walked 16.28 km according to GPS, and no more activity, but Dashboard is showing 24056 steps and  21.71 km. Steps are OK but distance is WRONG ... more than 5 km added

Moderator edit: subject for clarity

[LiliyaFitbit]

Welcome to the Fitbit Community, @yurek99.

I appreciate your participation in the Forums and sharing your concern regarding your Fitbit device calculating distance accurately. Thank you for providing the additional details. Fitbit devices calculate distance by multiplying your walking steps and walking stride length. Similarly, your running steps are multiplied by your running stride length. This will not necessarily match a pre-measured distance, nor will it show exactly what a GPS device will show. I recommend confirming that your personal information is correct. Please also make sure your Fitbit device and app are up-to-date. 

As you mentioned correctly, we use height to estimate your walking and running stride lengths. You may want to measure and manually adjust these values if your legs are longer or shorter than average. For more information, see How do I measure and adjust stride length for my Fitbit device?

I'll be around if you need further assistance. 

[yurek99]

I think, we don't undertand each other...

Let's do simple math:

I walked ~24000 steps, and my walking step length is ~67 cm so;

24000 * .67 = 16 080  16.08 km

Close enough with GPS and map? OK for me.

But this is not what 'Today' distance is showing:

24000 * .90 = 21600 21.8 km

Now we are at home. Instead multiplying my steps by walking stride, is multipling it by RUNNING stride (67cm vs. 90 cm)

Question is... How this can be fixed? I don't need this kind of encouragement

[Doug]

Just as a user giving my perspective, what I found in my running days (which I really should get back to...) was my running stride really varied significantly with my pace.  And Fitbit often seemingly applied my running stride even when I was power walking above 4mph (or about 6kmph).

My "solution" was to lower my profile's running stride to nearly equal to my walking stride.  It does make it report my distance low when running, but I decided I'd personally rather error on the side of too low of distance than too much distance in those stats.  (I get my runnings status from other methods anyway.)

I should add that I use the Fitbit One, which does not have a heart rate monitor.  So the algorithm used to decide between walking and running may be much different than on other models.

This is not a "solution" per se; just my user perspective on one way to handle it.

[yurek99]

Yes, you're right.

When I changed my running stride (90 cm) with walking one (67 cm), distance reported is OK.

But this looks like FitBit app issue. Walking 5 km/h ( ~3 mi/h) is NOT Running. Even brisk walk is conciser above  4 mi/h. 

So... lets, programmers work hard, so user does not have to.

[Jack-O]

Hello! Every week, I hike with my friends (from. 10 to 25 people) at the same mountain for 7 to 12 km each time. We walk the same distances (4 circuits) every other week, three times every week and we have done so for over 10 years. My Charge 5 (with Charge 2, it was the same) reports a wrong number of kilometers if I compare with the map we are given and if I compare with my companions (3 or 4 different makes of watches). Fitbit gives 1.5 or 2km more than the other hikers whenever we compare. I tried having my iPhone with me and without, with the GPS activated on my watch. I also checked the stride length: here I would have to subtract a little on the length, which would give me more kilometers as a result.  I suspect my Fitbit Charge 2/5 do not calculate the numbers correctly with the satellite. I think this problem should be remedied. My companions always laugh at me with my ridiculous distances provided by my watch.

[SunsetRunner]

Hello @Jack-O,

here we have 2 problems: the first is that for some strange reason Fitbit is calculating the distance via stride rather than using GPS, as confirmed above by @LiliyaFitbit.This is already a problem when walking and running when stride is not always constant as it changes depending the pace.

The second problem is that you're hiking and in this case the strides are already very variable depending the terrain and, even more, considering that you might move your wrist for other reasons and thus giving more additional steps.

The issue, as you can see, has been present since forever but Fitbit is ignoring the complains. For my running sessions I preferred to move to another brand of more dedicated running watches and use my Charge 4 only for daily tracking.

[t.parker]

@SunsetRunner the problem of distance is a lot more complicated. GPS shouldn't be used as the only distance indicator. If you look at any map you'll see how many inaccuracies it has, hence using GPS only isn't the solution (with approx. 3m GPS accuracy, the instant pace would be anything but accurate). GPS, however, is good when it contributes to continuous calibration. I don't want to go too much into technical issues but will dive a little more into details. While using the accelerometer in order to compute distance is possible, the method hits a very serious wall. The algorithms perform double integration which results in huge drift. Within a few seconds, the drift exceeds a couple of meters. Other sensors (and GPS) can be used in order to correct that drift. The shorter the time period to integrate, the less drift. Some algorithms recalibrate every step (watching vertical oscillation to detect steps and computing the length of stride based on accelerometer data). This, however, won't be perfect. GPS (when good) might come in handy but the best algorithms use all the available sensors array. Each sensor has some issues while another sensor may correct them. In other words, sensors correct each other. If Fitbit really used anywhere (even for non-GPS) simple "stride x steps" anywhere then it would be the World's worst-ever pedometer (I really hope this isn't true on Fitbit but this would explain a lot, too). I believe, "steps x strides" is just a version for regular users. One thing I believe is that Fitbit very unlikely updates the distance computation algorithm between devices and may not really be using the full array of sensors (some devices have gyro). Also, having a magnetometer on board would help outdoors, too (magnetometer can be used as part of data correction but I don't think any Fitbit device has it).

One thing that Fitbit algorithms lack is consistency which is probably more important than accuracy as it gives the user a relative way of looking at metrics. The same route should result in the same distance while in Fitbit's world it's not always a case. Then you know, whether your pace is getting better/worse etc. Fitbit probably has a very outdated code of the pedometer(and distance based on it) which is carried over onto new models.

Cycling uses GPS but accuracy isn't as good as using a speed sensor on the bike that knows the size of the wheel and counts revolutions. There is no perfect method to compute distance. Best solutions involve everything that hardware has to offer so one sensor keeps an eye on another one.

[Jack-O]

Thank you for your explanation. Walking with (trekking) poles might be a factor too, I suppose, as I rotate my wrist somewhat twisting the watch regularly. The problem is that others too do the same and get different results with their watches.

[Jack-O]

Thank you for this very technical answer. I will print it and try to figure it out.