Logging activity accurately

[GymRat1]

What sources do folks use when logging activity manually?  For example,, I jogged around the small track at the gym for 30 minutes (about 30 times around), and have no idea how fast I was going.  I'm not a runner so I estimated by selecting the slowest speed available (5 mph).  The calorie burn calculated by FitBit was 178 which seemed tremendously low given my level of effort.  What is a more accurate method (without having to wear another device)?

I also wonder about Zumba which is very high impact, as well as the other activities I perform at the gym which involves a lot of HIIT.  I'm not feeling confident that burn for these activities is being calculated accurately. 

Thanks

[Vicki67]

How do I log my actitives? I thought I did but it doesnt show them.

[Heybales]

---

@Vicki67 wrote:

How do I log my actitives? I thought I did but it doesnt show them.

---

Website - Log - Activities tab - at the bottom.

You can log an activity record with start/end times if you want to make a snapshot what the Fitbit stats were during the workout time. This is what the button press on some models does automatically.

This way they aren't buried in the daily stats, and you can review workouts down the road for comparison.

If you know the calories will be wrong by Fitbit and need to manually log your own, log the activity, if not one of the icons shown, type in the name. Provide start/duration to over-write Fitbit calorie burn with whatever you enter.

I'll let someone else mention where on the app - but likely the same location area, and same 2 options for what you can log.

[bzymummy]

This helps so much. I am new to this and wasn't sure if I needed to log my walking. I walk for 2.9 miles everyday during my lunch, it takes me 30 minutes but I wasn't sure how or if I needed to record this.

[Soxfan58]

When i enter manually non-step based activities such as swimming and spinning, is a step amount added toward my daily step total or just the minutes etc.? 

[Heybales]

Well, that's why they are called non-step based exercise.

If they even have steps - they are not related to walking or running steps
that Fitbit uses the formula for calculating calorie burn from.

That's why even elliptical is best manually logged on non-HR device, the
steps seen, even if all of them - have no relation to walking or running at
whatever distance Fitbit thinks it sees by the impact (or lack of) from the
movements.

[hockey_magnet]

---

@Heybales wrote:
Well, that's why they are called non-step based exercise.

If they even have steps - they are not related to walking or running steps
that Fitbit uses the formula for calculating calorie burn from.

That's why even elliptical is best manually logged on non-HR device, the
steps seen, even if all of them - have no relation to walking or running at
whatever distance Fitbit thinks it sees by the impact (or lack of) from the
movements.

---

Interestingly my Charge HR indicates results that are very close to the stats reported by the elliptical machine I use at the gym. After a one hour workout the calorie burn shown on the Fitbit is about 10% lower than the machine shows. I attribute this to the fitbit not taking into account the resistance level the elliptical is set to. The higher the resistance level, the greater the discrepancy. The fitbit step count is very close to elliptical movements. 

[Heybales]

"That's why even elliptical is best manually logged on non-HR device"

You are using a HR-device, so above advice doesn't apply.

But, HRM formula for calorie burn based on HR is ONLY valid for aerobic
exercise that is steady-state, same HR for 2-4 minutes. Farther away from
that you are, the less chance of accuracy it has.

So strength training and intervals should be manually logged, as that is
anaerobic and no where near steady state - if done correctly.

[hockey_magnet]

I certainly agree re anaerobic activities, almost anything that doesn’t involve running or walking BUT even my flex matched elliptical calorie burn closely enough that it didn’t really matter but as I said, the more resistance you dial in obviously the greater the discrepancy

[Heybales]

You lucked out with pure coincidence then, and others may too, but how
would they know without a HR type device to compare the calorie burn to.

And that should disappear as you get more fit and can do it stronger,
burning more calories at same weight.

So blanket advice should be as Fitbit FAQ's mention - non-step based cardio
is more accurately logged manually with non-HR devices.
HR-devices is fine for cardio, better estimate especially on non-step based
stuff - in general.

If you happen to match - buy lottery tickets.

[hockey_magnet]

Have to disagree. the elliptical is a step based motion with associated arm movements so it's not "luck". Flex counts elliptical "steps", calculates distance and calories. No different than walking

Sent from my iPhone

[Heybales]

---

@hockey_magnet wrote:
Have to disagree. the elliptical is a step based motion with associated arm movements so it's not "luck". Flex counts elliptical "steps", calculates distance and calories. No different than walking

Sent from my iPhone

---

Except the distance from steps is based on impact - and most would be sorely pressed to get an elliptical step impact to equal a walking or running step impact - representing the same distance.

Also, the formula for turning distance and time and weight in to calorie burn is only for walking and running.

You will find NO good formula for elliptical's and calorie burn.

Not only is walking and running on treadmill the MOST tested form of exercise to get correct calorie burn on in a lab environment, thereby producing formulas that are highly accurate - elliptical's have been shown to be way too variable in the different settings and the way people can use them to give any decent formula's.

So when I say step-based, as represented by Fitbit and any other activity tracker that is measuring step impact walking or running - elliptical is not in that context.

So very different from walking in what is being discussed here. Calorie burn.

[hockey_magnet]

either way it’s just splitting hairs. what do you mean: "Except the distance from steps is based on impact - and most would be sorely pressed to get an elliptical step impact to equal a walking or running step impact - representing the same distance.”? The flex does not Measure impact at all, it uses the movement of your arm to record number of steps which are then translated to distance which is then combined with your body weight to calculate “work” (calories burned). So depending on your relative elliptical settings and the way you use it, flex is counting steps accurately and then it’s just a matter of having the resistance setting set up correctly and you will get approximately the same results as walking on a treadmill or otherwise. If you’re using a treadmill on anything other than level, your treadmill results will also be inaccurate as Fitbit does not take inclination into account.

[Heybales]

---

@hockey_magnet wrote:
either way it’s just splitting hairs. what do you mean: "Except the distance from steps is based on impact - and most would be sorely pressed to get an elliptical step impact to equal a walking or running step impact - representing the same distance.”? The flex does not Measure impact at all, it uses the movement of your arm to record number of steps which are then translated to distance which is then combined with your body weight to calculate “work” (calories burned). So depending on your relative elliptical settings and the way you use it, flex is counting steps accurately and then it’s just a matter of having the resistance setting set up correctly and you will get approximately the same results as walking on a treadmill or otherwise. If you’re using a treadmill on anything other than level, your treadmill results will also be inaccurate as Fitbit does not take inclination into account.

---

Incorrect on arm swing making steps.

It is attempting to measure the impact of your steps DESPITE your arm moving, hence the difficulty for some people that swing their arm at just the right (or wrong) moment.

You can read up on tri-axis accelerometers and how they work and how they are used.

Walk without swinging your arm - you'll still get steps.

Just swing your arms like walking with no steps, smooth transitions - probably little to no steps.

That's what most people get.

You are correct though - you can attempt to adjust the settings on elliptical such that your seen steps translate to a distance that is given a calorie burn based on formulas for walking or running - that happen to match what the elliptical is actually making you burn.

But - how in the world could someone know what you are actually burning to make that match happen?

And why in the world would you want to screw around like that with the settings probably making the elliptical workout too easy merely to get that match for easy translation to calorie burn.

And you are correct - incline as it gets steeper and steeper is same issue as elliptical - the personal way of doing it causes the formula's for walking & running at incline to start being more inaccurate as the incline goes up.

But even more so as you mention - Fitbit has no idea that's what is happening.

It actually has a bad direction associated with it and inclines.

Since the step up is shorter, the impact is softer than level walking, and usually more frequent going up.

Those 2 things make the steps appear easier and shorter distance - leading to even less calorie burn - when it's actually more.

And downhill is opposite effect, more impact and usually longer stride, appearing as longer distance than it is - and bigger calorie burn - when it's not.

These can all be tested out. And I have on my Zip, and asked others with various step-based devices. Same results.

One lady was able to have her jogging in place register the correct distance on Flex, when compared to treadmill test doing same HR using Polar - and step-based calorie burn matched.

That was my first surprise it worked so well.

Then my tests taking different stride lengths at same pace, still came up with correct distance to within 1%, despite big difference in steps to make that happen, and with same distance, about same calorie burn too.

[hockey_magnet]

---

@Heybales wrote:

---

@hockey_magnet wrote:
either way it’s just splitting hairs. what do you mean: "Except the distance from steps is based on impact - and most would be sorely pressed to get an elliptical step impact to equal a walking or running step impact - representing the same distance.”? The flex does not Measure impact at all, it uses the movement of your arm to record number of steps which are then translated to distance which is then combined with your body weight to calculate “work” (calories burned). So depending on your relative elliptical settings and the way you use it, flex is counting steps accurately and then it’s just a matter of having the resistance setting set up correctly and you will get approximately the same results as walking on a treadmill or otherwise. If you’re using a treadmill on anything other than level, your treadmill results will also be inaccurate as Fitbit does not take inclination into account.

---

Incorrect on arm swing making steps.

It is attempting to measure the impact of your steps DESPITE your arm moving, hence the difficulty for some people that swing their arm at just the right (or wrong) moment.

You can read up on tri-axis accelerometers and how they work and how they are used.

I ALREADY KNOW THAT, NEITHER YOU NOR I HAVE ANY KNOWLEDGE OF FITBIT'S ALGORITHMS AND DATA ANALYSIS

Walk without swinging your arm - you'll still get steps.

Just swing your arms like walking with no steps, smooth transitions - probably little to no steps.

That's what most people get.

THIS IS TOTALLY INCORRECT. HOW MANY WRIST MOUNTED DEVICES HAVE YOU TESTED? I CAN TELL YOU WITH HAVING USED A FORCE, 2 FLEXES AND NOW A CHARGE OVER A 2 YEAR PERIOD THAT THE EXACT OPPOSITE IS TRUE. SWING YOUR ARMS, STEPS COUNT. WALK NORMALLY AND SMOOTHLY WITH ARM HELD FIRMLY AGAINST CHEST, NO STEPS. RUN OR WALK RAPIDLY WHICH GENERATES MORE MOVEMENT EVEN IF YOU TRY AND HOLD  YOUR STATIONARY AND YOU WILL GET A STEP COUNT. THAT IS WHY MANY PEOPLE COMPLAIN THEY DO NOT GET A PROPER STEP COUNT WHEN PUSHING A STROLLER, ETC. 

You are correct though - you can attempt to adjust the settings on elliptical such that your seen steps translate to a distance that is given a calorie burn based on formulas for walking or running - that happen to match what the elliptical is actually making you burn.

But - how in the world could someone know what you are actually burning to make that match happen?

YOU TEST IT AGAINST THE PROFESSIONAL ELLIPTICAL RESULTS AND OR HEART RATE MONITOR ALGORITHMS

And why in the world would you want to screw around like that with the settings probably making the elliptical workout too easy merely to get that match for easy translation to calorie burn.

THIS MAKES NO SENSE, IN THE SAME VEIN, WHY USE A FITBIT WITH A TREADMILL.? 

And you are correct - incline as it gets steeper and steeper is same issue as elliptical - the personal way of doing it causes the formula's for walking & running at incline to start being more inaccurate as the incline goes up.

But even more so as you mention - Fitbit has no idea that's what is happening.

It actually has a bad direction associated with it and inclines.

Since the step up is shorter, the impact is softer than level walking, and usually more frequent going up.

Those 2 things make the steps appear easier and shorter distance - leading to even less calorie burn - when it's actually more.

And downhill is opposite effect, more impact and usually longer stride, appearing as longer distance than it is - and bigger calorie burn - when it's not.

These can all be tested out. And I have on my Zip, and asked others with various step-based devices. Same results.

I'M SURE YOU REALIZE THE ZIP IS NOT THE SAME AS A WRIST MOUNTED DEVICE WHICH IS WHAT I'M REFERRING TO. EVEN IF THE BASIC TECH IS THE SAME, MONITOR PLACEMENT,  DATA ANALYSIS AND APPLIED ALGORITHMS WILL DIFFER. 

One lady was able to have her jogging in place register the correct distance on Flex, when compared to treadmill test doing same HR using Polar - and step-based calorie burn matched.

That was my first surprise it worked so well.

THAT'S SIMPLY BECAUSE IMPACT DOES NOT MATTER, STRIDE LENGTH DOESN'T MATTER EITHER SINCE IT IS USING THE STRIDE LENGTH YOU INPUT OR THE DEFAULT TO CALCULATE DISTANCE WHICH PROVES MY POINT

Then my tests taking different stride lengths at same pace, still came up with correct distance to within 1%, despite big difference in steps to make that happen, and with same distance, about same calorie burn too.

PRETTY DUBIOUS I'D QUESTION YOUR TESTING METHODOLOGY - WHY DOES FITBIT WANT YOU  TO ENTER YOUR STRIDE LENGTH, CHANGING THE DEFAULT TO ENSURE BETTER DISTANCE ACCURACY? OBVIOUSLY BECAUSE THAT'S WHAT THEY USE TO CALCULATE DISTANCE -NO. OF STRIDES X STRIDE LENGTH INPUT. YOU'RE GIVING FITBIT WAY TOO MUCH CREDIT FOR COMPLEX ENGINEERING AND PROCESSING THAT SIMPLY DOES NOT EXIST IN THESE DEVICES

---

[slysam]

Impact does matter with the Fitbit and accelerometers in general. I agree it is very different between the torso worn (one and zip) and wrist worn devices (I happen to have a few Fitbit's -- a one, charge, and Surge).

I am not sure whether it is exactly impact, but yet I get a higher Fitbit estimated calorie burn and active minutes with higher impact. (I know they measure gravity and aceleration and track movement relative in three directions -- up/down, right/left, forward/back). I am just not sure with how Fitbit is programed whether it is using more the gravity/acceleration or the up/down motion.

Anyway

*I see a higher calorie burn going down the stairs than up when moderating my pace with a metronome

*I see higher downhill rather than uphill

*I always get more active minutes with high impact activities (though running in place has enough impact)

*I tested a walking workout video doing it once normal and once with a weighted vest (10 pounds). They had very similar step count and the exact same distance. But the weighted vest was credited with higher calorie burn and more active minutes. I did not log anything or change my settings. I felt slugish and had trouble staying on beat while wearing the vest. I expected it to have been credited lower. I was really surprised.

I am not sure how stride length fits in with calorie burn if Fitbit tracked. I thought stride length was mainly to estimate distance. However, I think impact (or up and down motion) might have something to do with whether the walking stride or running stride is applied (and speed too, but my jump roping always counted as "very active" under the old way but the pace was slower than my walking that was counted as "moderately active"). I can only guess it identifies jumproping as running because of either the impact or the up and down motion.

The wrist worn are totally different in how they feel impact though. I miss steps holding on to things like shopping carts. I actually get credited with fewer steps with the wrist worn than my One. I have to move my arms a lot to get false step credit. But it is apples to oranges in some ways.

Of course, I don't know Fitbit's secret formula and a lot of it is in how it is programmed to recognize movement. Have you seen the article from the New York Times "What Does Your Activity Tracker See and Not See?" It is really informative, but it wasn't a Fitbit and it was posted a couple years ago... But the draphs make it look like the accelerometer they use responds a little to impact (Look at the weight lifting example):

http://well.blogs.nytimes.com/projects/2014/03/accelerometers.html

[Heybales]

First - every step is NOT given your static stride length.

This is easily tested by what I described doing - same pace but different turnover or cadence used, therefore different stride length is used.

You can most obviously test out merely by your daily stats that the total distance seen is NOT a result of steps x stride length.

It's exactly why people complain that something must be broken when they have big step days but less calorie burn overall compared to smaller step days and big burn.

The stride-length merely gives an expected impact for a step with your mass.

If less impact, then stride was shorter, if more impact then stride was longer, if over a certain amount then running.

In simple terms that's what it is doing.

It's impact and to a lesser extent "hang time".

So see, that little point you have not undestood just proves to me you actually don't understand how this works.

And yes, mine is Zip - did you miss the point I worked with others and their devices too? Plenty have been willing to test things out for me, like lady jogging in place.

Yes, hold your arms to your chest, also hold on to shopping cart with rock steady grip - no impact from steps seen in the wrist. No swing either obviously, but it's not the swing - which doesn't provide nearly enough info to estimate a dynamic distance for each step - which is what is happening.

Fitbit's very specific algorithms may not be known - but guess what - the chipsets they buy and others that have the accelerometers in them - already come with algorithms built in. There ain't that much super secret sauce to what Fitbit is doing, though they may be tweaking a few things.

Did you know that people take their wrist devices and mount them on their ankles too - and guess what it detects - impact and steps.

For some that have tested for me, it saw impact on only 1 step of attached foot when walking - but both feet when running. I'll let you think about that one and why that makes perfect sense.