Garmin Vivosport underestimating my calories burned.

Options
I'm currently the new owner of a garmin vivosport tracker, which seems to vastly calculate my calories burned far lower than my previous fitbit charge 2! Has anybody else experienced this problem and how to rectify it? Thank you.

Replies

  • bjoshua55
    bjoshua55 Posts: 1 Member
    Options
    I have had both brands and, while it all is an estimate, Garmin seems to be the most consistent. When I use various equipment at the gym that tracks calories, my Garmin is usually within 2 - 5% consistently. The Fitbit was all over the place. Some days it was 20%higher or lower, other days it was less than 5% than what the gym equipment showed. Not sure if it is the programming or the sensors.

    Also found the Garmin to be in line with Map My Fitness app when I went on walks/runs outside more consistently than Fitbit.
  • collectingblues
    collectingblues Posts: 2,541 Member
    Options
    Fitbits have a reputation for being generous and overestimating calories burned. Is there a reason you think it’s the Fitbit, and not the garmin, that’s accurate?
  • Theoldguy1
    Theoldguy1 Posts: 2,478 Member
    Options
    PAV8888 wrote: »
    Hold on just a second everyone.

    Why can it NOT be that the Garmin under-estimates and the Fitbit is spot on for the person in question?

    While yes, in general, the Garmins give lower values for calories burned, there are many of us who have found their Fitbits to be fairly accurate for them. Which would make a lower value from Garmin, potentially, inaccurately low.

    I would assume the OP thinks their Fitbit is right because they have been losing weight as predicted by their purported deficit while using it.

    I would check my setup for Garmin and / or discuss with their support...

    Just curious, what makes you think your Fitbit is that accurate?
  • Theoldguy1
    Theoldguy1 Posts: 2,478 Member
    edited June 2019
    Options
    PAV8888 wrote: »
    Theoldguy1 wrote: »
    Just curious, what makes you think your Fitbit is that accurate?

    TL;dr: Observed over-estimation of 2.7% over 1256 days.

    A base assumption is that a lb of weight change requires 3500 Cal, a value that may not be absolutely true but which is widely used.

    During the 1256 days I kept fairly accurate MFP records and monitored my weight trend via Trendweight. My MFP food logs and 86.8lbs of weight change account for an average TDEE of 3,051 Cal a day.

    In the same time period, my various Fitbits estimated my average TDEE at 3135 Cal a day.

    A difference of 85 Cal or 2.7% of my TDEE as estimated by Fitbit.

    I also note that during the same time period I did not find any two+ month time period where divergence exceeded 6%.

    On a full year basis the smallest divergence was less than 0.25% for year one, increasing to 3.14% for year two, 4.77% for year three, and was hovering at 2.66% by the end of the first five months of year four.

    Subsequent to that changes to MFP (harder to capture past food logs) and reduced personal incentive at maintenance made it too much work for me to continue analyzing my data in similar detail.

    My activities were well suited for capture by a tracker. My logging was detailed. My deficit did not exceed 25% of TDEE while obese or 20% of TDEE while below mid overweight.

    Looks like you've done a lot of record keeping.

    To me there are factors such as record keeping lapses and, the bigger elephant in the room, food label calories can be off +/- 20% that make me question "accuracy" claims.

    The FDA allows food companies wide latitude in the accuracy of the calories listed on package labels—20 percent in either direction. That means if a label says 200 calories per serving, it could be 240 calories or 160 calories or anything in between.

    https://www.berkeleywellness.com/healthy-eating/nutrition/article/can-you-trust-calorie-counts

    If a tracker as a tool helps one hit goals, great. But to me, "accuracy" needs to be taken with a bag of salt. Any measurement such as calories burned that incorporates multiple factors can only be as "accurate" as it's weakest link.
  • PAV8888
    PAV8888 Posts: 13,940 Member
    Options
    Theoldguy1 wrote: »
    PAV8888 wrote: »
    Theoldguy1 wrote: »
    Just curious, what makes you think your Fitbit is that accurate?

    TL;dr: Observed over-estimation of 2.7% over 1256 days.

    A base assumption is that a lb of weight change requires 3500 Cal, a value that may not be absolutely true but which is widely used.

    During the 1256 days I kept fairly accurate MFP records and monitored my weight trend via Trendweight. My MFP food logs and 86.8lbs of weight change account for an average TDEE of 3,051 Cal a day.

    In the same time period, my various Fitbits estimated my average TDEE at 3135 Cal a day.

    A difference of 85 Cal or 2.7% of my TDEE as estimated by Fitbit.

    I also note that during the same time period I did not find any two+ month time period where divergence exceeded 6%.

    On a full year basis the smallest divergence was less than 0.25% for year one, increasing to 3.14% for year two, 4.77% for year three, and was hovering at 2.66% by the end of the first five months of year four.

    Subsequent to that changes to MFP (harder to capture past food logs) and reduced personal incentive at maintenance made it too much work for me to continue analyzing my data in similar detail.

    My activities were well suited for capture by a tracker. My logging was detailed. My deficit did not exceed 25% of TDEE while obese or 20% of TDEE while below mid overweight.

    Looks like you've done a lot of record keeping.

    To me there are factors such as record keeping lapses and, the bigger elephant in the room, food label calories can be off +/- 20% that make me question "accuracy" claims.

    The FDA allows food companies wide latitude in the accuracy of the calories listed on package labels—20 percent in either direction. That means if a label says 200 calories per serving, it could be 240 calories or 160 calories or anything in between.

    https://www.berkeleywellness.com/healthy-eating/nutrition/article/can-you-trust-calorie-counts

    If a tracker as a tool helps one hit goals, great. But to me, "accuracy" needs to be taken with a bag of salt. Any measurement such as calories burned that incorporates multiple factors can only be as "accurate" as it's weakest link.

    Non biased errors in sufficient quantity cancel as opposed to reinforce.

    Nutrients are off at times in all directions in terms of nutritional content. But generally not in a single direction.

    Weight (quantity consumed) is in your power to measure to within a gram regardless of packaging errors.

    One part where I do "cheat" is that I consider my caloric intake equal to my macros multiplied minus 50% of fiber.

    This avoids the systematic rounding down of calories that many labels use and accounts for the fact that often fiber is partially digested yielding calories as opposed to fully undigested and counting for zero (looking at you older quest bar labels). I don't drink sufficient alcohol for this to matter (since alcohol macros aren't recorded)

    Having said that, the point is that as long as you have *consistency* in your estimation all the above discussion about accuracy is *irrelevant*

    You can correct your goals based on 2.7% of TDEE or 27% of TDEE if the measurements are consistent.

    Your food logging is up to you. My first three years on MFP I did not have any record lapses and did break down restaurant meals from pictures of my plate when I didn't have a scale with me.

    Activity tracking is more subtle. What feels difficult one day may be less than what feels easy on another. Beginning and ending times are not always what we think and we often pause during activity.

    *I* have found that I prefer to off load the activity tracking to a third party service instead of trying to figure out and record everything.

    The problem with your "labels are out 20% therefore calorie counting will never be accurate" therefore there is no point in attempting to accurately count my calories in in the first place, is the exact same one I had for almost 30 years. Long term weight loss is impossible--so why try?

    In both cases I have found that trying may just work if the moons align, and that striving to be accurate (even if I'm not) makes me more *consistent* which allows me to make meaningful corrections.

  • AnnPT77
    AnnPT77 Posts: 32,872 Member
    Options
    It takes a tracker a while to learn us - few weeks.

    Explore every user setting; check them; set them accurately

    Especially important: Stride length, max heart rate (for heart rate devices). (Age estimates of max HR are inaccurate, way often.)

    Do any self-tests it offers (like VO2max estimate, or sub-max HR, or 'fitness' test).

    Even then, it can be off. It's just a statistical estimate. There are a few outliers; It's possible that you're one (but statistically not very likely).

    They're close for most people (statistics! ;) ).

    My Garmin Vivoactive 3 is 25-30% low for me (so is MFP), compared with almost 4 years of careful logging data. It happens.

    If it happens for you, one option is to estimate the average error percent, and adjust accordingly. Or, just go off logging and the scale. In maintenance, I do the latter.
  • lorrpb
    lorrpb Posts: 11,464 Member
    Options
    OP Can you give some examples of the difference you’re concerned about? You say they are “vastly” different but that could mean 50 calories to one person and 1000 to the next.
  • NorthCascades
    NorthCascades Posts: 10,968 Member
    Options
    Theoldguy1 wrote: »
    PAV8888 wrote: »
    Theoldguy1 wrote: »
    Just curious, what makes you think your Fitbit is that accurate?

    TL;dr: Observed over-estimation of 2.7% over 1256 days.

    A base assumption is that a lb of weight change requires 3500 Cal, a value that may not be absolutely true but which is widely used.

    During the 1256 days I kept fairly accurate MFP records and monitored my weight trend via Trendweight. My MFP food logs and 86.8lbs of weight change account for an average TDEE of 3,051 Cal a day.

    In the same time period, my various Fitbits estimated my average TDEE at 3135 Cal a day.

    A difference of 85 Cal or 2.7% of my TDEE as estimated by Fitbit.

    I also note that during the same time period I did not find any two+ month time period where divergence exceeded 6%.

    On a full year basis the smallest divergence was less than 0.25% for year one, increasing to 3.14% for year two, 4.77% for year three, and was hovering at 2.66% by the end of the first five months of year four.

    Subsequent to that changes to MFP (harder to capture past food logs) and reduced personal incentive at maintenance made it too much work for me to continue analyzing my data in similar detail.

    My activities were well suited for capture by a tracker. My logging was detailed. My deficit did not exceed 25% of TDEE while obese or 20% of TDEE while below mid overweight.

    Looks like you've done a lot of record keeping.

    To me there are factors such as record keeping lapses and, the bigger elephant in the room, food label calories can be off +/- 20% that make me question "accuracy" claims.

    The FDA allows food companies wide latitude in the accuracy of the calories listed on package labels—20 percent in either direction. That means if a label says 200 calories per serving, it could be 240 calories or 160 calories or anything in between.

    https://www.berkeleywellness.com/healthy-eating/nutrition/article/can-you-trust-calorie-counts

    If a tracker as a tool helps one hit goals, great. But to me, "accuracy" needs to be taken with a bag of salt. Any measurement such as calories burned that incorporates multiple factors can only be as "accurate" as it's weakest link.

    Agreed.