FAQ - Syncing, logging food & exercise, calorie adjustments, activity levels, accuracy
Replies
-
HR is likely going to be better, I really doubt RunKeeper incorporates elevation changes, unless they replace the bad GPS data with better online data, which Garmin does do, so it is possible.
Actually Runkeeper does incorporate the elevation/GPS data. I'm aware of the elevation issue you mention and I do their 'fix' which corrects the route and calorie burn. That said, I have no idea how accurate it is.Wondering though why you want to trust MFP's rough estimate of daily burn outside of exercise, when you have a device able to measure it much more accurately?
You can still manually log that workout on MFP if it must be seen logged in your exercise diary, and then that will show you that any adjustment that remains is exactly because the MFP estimate needed correction.
Have you ever looked at the adjustments that remain after manually logging the exercise to see how close MFP was in the first place?
Right now I have a good rhythm using MFP for food and adding/importing exercise to manage my net calorie goals. I got the device in the hopes of getting more accurate calorie burn for exercise. I am willing to switch over to using the device for full daily burn once I get better used to the device. I only got it yesterday!
I have not connected Fitbit to MFP yet. I am still hoping to learn if my logged activity walks have an accurate calorie burn given your statements about what the HR formula. Do you think the HR formula is accurate for power walks like mine? Or is it using estimated steps?
0 -
I'll bet your HR is getting high enough on power walks for it to be used. When you hit the button, it starts doing per sec logging of HR and calorie burn, and as long is steps is frequent enough it figures it is a workout.
If steps stop but HR is high, it figures malfunction (or perhaps stage fright) and stops either the activity or using HR formula.
But if Runkeeper is using corrected elevation profiles, those formula's are really accurate too, more than HRM actually, though steep incline starts causing more and more inaccuracy.
http://community.myfitnesspal.com/en/discussion/774337/how-to-test-hrm-for-how-accurate-calorie-burn-is
0 -
HR formula is used when you start an activity record with button press, and your steps are high enough to indicate the higher HR is indeed from exercise, or HR and steps go up to show exercise is being done. Otherwise it thinks elevated HR due to something else and just uses step calculation.
HR formula is only a decent calorie estimate when the exercise is steady-state aerobic, same HR for 2-4 min. Anything anaerobic, like good lifting or interval workout, which is also non-steady-state HR, is not valid use of formula and will result in inflated calorie burn. Step based would actually be best for intervals.
Is there Fitbit documentation that explains this? I'd like to know what value or modeling they use for to determine when "steps are high enough". Also why interval is not good use of HR formula if heart rate is taken at 1 sec intervals and should show my heart is working harder. I am still skeptical of the calorie burn values produced despite reading all the linked references and other posts. Fundamentally I would expect a HR-based calorie burn to be more accurate, but this issue of Fitbit choosing/applying varying formulae has me guessing at what is going on.
Even my walks have varying HR due to hills, and soon I will incorporate a little interval jogging / walking which will further increase the variable HR readings.0 -
@heybales - First, thank you for hitting on points that are often confusing with regard to the activity trackers and working them into the EM2WL program. I've been meaning to ask you about the variation / correction of weight training workouts in relation to my activity tracker (you often say the calories are underestimated by the AT). But by what I am reading on these posts/comments, it seems the FitBit is different in how it figures calories than how my Bodymedia Fit does. BMF not only tracks movement (steps) like FitBit, but also skin temperature, heat flux and skin response. According to BMF, these additional measurements are to record non-step activity.
My questions: how accurate, then, do you think the BMF is for weights and machine-based exercises (stairstepper, elliptical, crossfit-esque cardio machine, et cetera)? And do you think that it is still underestimating weights and I need to manually put in weight training, elliptical and such through MFP? So far it averages me at about 2200 calories, and going by the calculators, I'm figuring that is pretty close, if not over-estimated by a tad. Thoughts?0 -
HR formula is used when you start an activity record with button press, and your steps are high enough to indicate the higher HR is indeed from exercise, or HR and steps go up to show exercise is being done. Otherwise it thinks elevated HR due to something else and just uses step calculation.
HR formula is only a decent calorie estimate when the exercise is steady-state aerobic, same HR for 2-4 min. Anything anaerobic, like good lifting or interval workout, which is also non-steady-state HR, is not valid use of formula and will result in inflated calorie burn. Step based would actually be best for intervals.
Is there Fitbit documentation that explains this? I'd like to know what value or modeling they use for to determine when "steps are high enough". Also why interval is not good use of HR formula if heart rate is taken at 1 sec intervals and should show my heart is working harder. I am still skeptical of the calorie burn values produced despite reading all the linked references and other posts. Fundamentally I would expect a HR-based calorie burn to be more accurate, but this issue of Fitbit choosing/applying varying formulae has me guessing at what is going on.
Even my walks have varying HR due to hills, and soon I will incorporate a little interval jogging / walking which will further increase the variable HR readings.
I've not seen Fitbit doc's - they'd probably like to avoid the confusion.
It's the nature of what increased HR means when doing cardio, to supply more oxygen for the more carbs that are being burned for the bigger calorie burn.
That's the only correlation between HR and calorie burn.
And it only applies in the aerobic exercise range, which for men and women and been found to start at about 90 bpm (HR flex point). And it ends when you go anaerobic, which varies. But HRM doesn't even attempt to figure that out, it keeps on applying the formula up to whatever level.
I don't know if they use 90, or something more personalized like % of HRmax they calculated.
While HR could shoot up from fright, it's not because of a need for more oxygen. Hormones are prepping in case there is and you must run, but at the moment, there isn't. So not a bigger calorie burn, except the heart going faster, but that doesn't burn nearly as much as the increased HR would doing cardio, and what a HRM would think you are burning.
Same as lifting, HR shoots up from exertion, not because of need for oxygen, energy source is ATP stored in muscle ready to be used without oxygen. And after it shoots up doing some good hard squats say, how long to come back down, while only standing there resting? Does it ever reaching standing resting HR level?
Here's some details on it. In the lower section called Worse Math. Only if really curious.
http://www.myfitnesspal.com/blog/heybales?month=201301
http://www.ncbi.nlm.nih.gov/pubmed/21904287
http://www.ncbi.nlm.nih.gov/pubmed/2706223
The problem with intervals and the fact Fitbit would use HR based formula, is again invalid use of such formula.
Walk 4 mph for 5 min and notice what the avg HR is.
Now jog for 5 min and see what the HR gets up to.
Now Walk 4 mph for 5 min. Did the HR instantly go down to the prior level that was actually needed for that pace?
No, and likely the lag time to go high was much shorter than the lag time to drop - so it's not a wash out.
I'd be curious if the HR even got back down to walking speed in 5 min.
So that whole time it was elevated HR for supplying the amount of oxygen actually needed for walking right then.
The HR is high for other things, not because more oxygen is still required for level of effort.
Now, the greater the difference and worse the fitness level, the worse the HR calorie burn estimate because of that effect. But good fitness level and smaller difference between HR levels, the better accuracy because of the wash effect.
It would be an interesting test on treadmill, wish I could do it. Fitbit would have to be proved out to be decently accurate on distance, and stride length corrected if need be.
Start walking slow, don't start an activity, right on an even 5 min mark according to Fitbit's time.
Every 10 min increase the pace so that HR goes up by 10 bpm.
Keep doing that until reaching your powerwalking pace and HR for 10 min.
On review in the 5 min daily graph, looking at calorie burn and HR, should see the jumps in HR.
Now, I'm betting for the 2 different 5 min blocks of calorie burn at the same pace, the HR is actually different averages.
The question will be, if calorie burn for each 5 min block is by distance (based on steps of course), they would match if distance did. If done by HR, then they should vary.
It may be required to make an activity record for each 5 min segment to compare when the level should be the same.
This would probably also make it evident what point the calorie burn went to HR based.
So you aren't going to get as accurate as you might hope.
The best would be in a metabolic chamber doing your workout.
Second would be a metabolic cart hookup with face mask and EKG HR unit.
Third would be a VO2max test of the step up variety to give 2 formulas to use later, steady-state and non-steady state formula, to be used for other workouts.
Fourth would be a proxy for VO2max test by doing a running test and developing the same formula.
Ha! - http://www.myfitnesspal.com/blog/heybales?month=201405
All will have limitations when actually applied to real life use - just like the Fitbit.
So with caveats to use, it may be the best you can get for now.
0 -
@heybales - First, thank you for hitting on points that are often confusing with regard to the activity trackers and working them into the EM2WL program. I've been meaning to ask you about the variation / correction of weight training workouts in relation to my activity tracker (you often say the calories are underestimated by the AT). But by what I am reading on these posts/comments, it seems the FitBit is different in how it figures calories than how my Bodymedia Fit does. BMF not only tracks movement (steps) like FitBit, but also skin temperature, heat flux and skin response. According to BMF, these additional measurements are to record non-step activity.
My questions: how accurate, then, do you think the BMF is for weights and machine-based exercises (stairstepper, elliptical, crossfit-esque cardio machine, et cetera)? And do you think that it is still underestimating weights and I need to manually put in weight training, elliptical and such through MFP? So far it averages me at about 2200 calories, and going by the calculators, I'm figuring that is pretty close, if not over-estimated by a tad. Thoughts?
Step-based trackers are badly underestimated.
Squats for 45 seconds and maybe 1 step seen - that would be one slow pace, short distance, and low calorie count if even given above BMR level burn.
HR-based trackers will inflate the calorie burn by some amount. Explanation above this pots.
BMF can be decent, if the sensors work well for you. And caveats there to potential accuracy.
Is a hot run outside at 6 mph level burning the same as in the gym cooler air at 6 mph? Yes.
But guess what the sensors tell the BMF because of the extra heat your body generates in the heat.
Does that quarter inch square sensor on your tricep see an increase in temp when doing leg day for strength training?
The galvanic sweat sensor is really just an on/off button, I called BodyMedia about it when mine showed on body the whole day - when I forgot to put it on after the shower, and the sensors were touching the strap. That was enough to complete the circuit, making it appear it was on body. But with no movement, sleeping level burn. And since it was on-body, no way to manually correct the calories.
So it may or may not, easiest to test by eating more and more until you actually gain weight.
2 week 250 test is best. If you think you are eating at TDEE already, then eat 250 more daily for 2 weeks.
Should slowly gain only 1 pound.
If more and faster, you weren't at TDEE then, and just topped off glucose stores with water. Which wouldn't be depleted to any level if really eating at TDEE.
So cardio depends on you testing out sensors on your body, and if seems good, calorie burn probably pretty good. The motion helps too which it sees and is a bigger factor in calorie burn anyway.
Weights - I'd log them.0 -
Thank you, sir! Right now I'm doing -15% (working my way up), but I keep adding calories as I watch my BMF showing higher daily burns. When I get close to my estimated TDEE, I'll do that with the 250. I'll start logging my weights as well.
Thanks, again.
0 -
okay, I have read the entire thread looking for an answer to my problem. & I haven't found one. I just need to know how to log activities on my Fitbit and get accurate calories. I am talking about no one step activities with a Fitbit flex.
the other day I logged 39 minutes on my stationary bikeand Fitbit only estimated 51 calories burned for 39 minutes. MyFitnessPal would have given me a hundred and seventy calories burned. I know that my fitness.pal isn't considered very accurate but I assume it should be more than what I would normally get for walking.
even if I tried to log in exercise in my Fitbit it wants to know how many calories I've burned but I thought Fitbit would estimate that based on the data it receives and my weight and what not.
I don't have a heart rate monitor because it isn't in the budget right now but everything I've read says that the Flex is great for tracking even non-stop activity. I just can't figure out how to input it.
I have MyFitnessPal set to sedentary and have allowed for the negative calories because I want to get a more accurate picture of my activities during the day. This is why I'm stressing over getting the manually entered calories correct in my fitbit.
,any help or insight would be greatly appreciated.0 -
okay, I have read the entire thread looking for an answer to my problem. & I haven't found one. I just need to know how to log activities on my Fitbit and get accurate calories. I am talking about no one step activities with a Fitbit flex.
the other day I logged 39 minutes on my stationary bikeand Fitbit only estimated 51 calories burned for 39 minutes. MyFitnessPal would have given me a hundred and seventy calories burned. I know that my fitness.pal isn't considered very accurate but I assume it should be more than what I would normally get for walking.
even if I tried to log in exercise in my Fitbit it wants to know how many calories I've burned but I thought Fitbit would estimate that based on the data it receives and my weight and what not.
I don't have a heart rate monitor because it isn't in the budget right now but everything I've read says that the Flex is great for tracking even non-stop activity. I just can't figure out how to input it.
I have MyFitnessPal set to sedentary and have allowed for the negative calories because I want to get a more accurate picture of my activities during the day. This is why I'm stressing over getting the manually entered calories correct in my fitbit.
,any help or insight would be greatly appreciated.
Section 1 of the FAQ - second question.
Logging exercise calories.
You did NOT manually log a workout on Fitbit, you either started an activity record with the button, or you logged an activity record merely viewing the Fitbit stats.
Neither is manually logging a workout.
Where you enter in start time, duration, intensity or distance, and let Fitbit calculate calories for you, which then replaces whatever the device saw and estimated based on steps.
Where did you read that Flex is great for tracking non-step (unless you really meant non-stop) activity?
I think you've proven to yourself that's not true on the bike.2 -
I meant non step. Exercise that isn't step based.
And actually, I went back and deleted the record and added it back and it logged correctly so it it must have been a glitch.
thanks for the reply.0 -
I figured out that I have to add manually from the computer not the app for some reason.0
-
Really? I went on the laptop last night and fund my way around (a little) and everything I had on MFP was there on the fitbit site as well as my phone. Unless im missing something. lol, which is most likely the case.
Im going to try the link someone posted recently on the other page.
0 -
You can indeed add a manual activity from the app. And an activity record too, incase you forgot to hit the button on your device that has a button.
And I knew you were talking about non-step based activity, and that question in the FAQ answered that question.0 -
Very informative post. However, 1 - I cannot find where it addresses logging food input into fitbit, as opposed to MFP. If food is logged into Fitbit, are there any issues
2 - I am trying out the charge HR, but prefer to wear a wahoo chest strap HRM for better accurancy during cardio workouts. The Wahoo app transfers workout info including times to MFP, and I assume based on the information in the post that MFP will then transfer that data to Fitbit which will replace its data with the new data from MFP based on logged time interval. IS this a correct assumption? The original heartrate data will likely not be ported but I can live with that.
3 - Finally I use a non fitbit compatible scale from wahoo and it sends weight data directly to MFP, and apple health for that matter. However, the weight synch between Fitbit and MFP appears to only be one way and fitbit never gets my weight data. any work around for this?
I apologize for posting the last item here but have asked fitbit support and got 2 responses already dodging the actual question with no answer, which I assume means it likely they don't have a solution.0 -
1 - 3rd question - where do I log my food? Yes there would be issues logging on Fitbit, food doesn't sync from Fitbit to MFP.
2 - Yes. and correct on HR data. Merely a workout of time and calorie burn.
3 - It should, may be time lag. Mine always does. Within seconds using the web account.0 -
So your weight data goes from MFP to fitbit? That is not working for me. Everything eslse seems to be working as described.0
-
In addition to confirming that The weight is synching from MFP to Fitbit, and not just the other way for others, when an excercise from MFP ports to Fitbit, and fitbit uses that data for the activity for the times in question, does this effect the steps taken during that interval? DOes fitbit zero them out or make any adjustment?0
-
Any workout that syncs over from other sites (except for a few they support distance with, MFP not being one), the steps and distance are untouched, only calories changed.
On their own site, only walking & running take your claimed distance, divided by your profile stride length, and recalc the steps to match.
Any other manually entered workout is same as imported - only calories replaced.
Unless driving is entered, then steps and distance is wiped out.0 -
Hello - I've read lots around this thread but can't seem to find a clear answer to the following. I'd like to:
1) Log food in MFP (much more comprehensive food database) - if connected to Fitbit, will the food data sync over to Fitbit?
2) All other data I'd like in Fitbit - exercise, calories, sleep, etc. I want to see everything in Fitbit and don't care about seeing anything in MFP other than logging food to feed my Fitbit data.
3) Don't want to log exercise manually, just let Fitbit track.
Is this possible? If so, what are the specific settings I need to get this done?
Thank you in advance.
1 -
vanessaescalona wrote: »Hello - I've read lots around this thread but can't seem to find a clear answer to the following. I'd like to:
1) Log food in MFP (much more comprehensive food database) - if connected to Fitbit, will the food data sync over to Fitbit?
2) All other data I'd like in Fitbit - exercise, calories, sleep, etc. I want to see everything in Fitbit and don't care about seeing anything in MFP other than logging food to feed my Fitbit data.
3) Don't want to log exercise manually, just let Fitbit track.
Is this possible? If so, what are the specific settings I need to get this done?
Thank you in advance.
That's exactly how I use Fitbit and MFP. Track food in MFP and exercise in Fitbit. Fitbit syncs calories burned to MFP.0
