Body Fat scales?

Options
wijody
wijody Posts: 26 Member
in Health and Weight Loss
I got one recently and when it was set to 'athletic' it said I only have 13% bodyfat. Well, my workouts are mostly cardio, so I then set it to regular (not athletic) and it was still low - 18%. This scale has good reviews and seems to give a more believable reading for my husband. What about my body could give false readings? Over-hydration? (I don't think I'm over-hydrated) or maybe my calves are muscular and it's reading mostly my lower body because of the probes being on the bottom of the foot?

Replies

  • jwdieter
    jwdieter Posts: 2,582 Member
    Options
    This article does a good job of explaining the problems with body fat scales:
    http://weightology.net/weightologyweekly/?page_id=218
  • tquill
    tquill Posts: 300 Member
    Options
    Overall, they're not all that accurate. From what I understand, they can vary from a number of things (dirty or wet feet, ect...)

    If it's just on your feet, then I think it's only measuring across your legs... so you're upper body may be out of the calculation. They make some with hand and feet sensors... so I've heard they're a bit more accurate.

    I'm not entirely certain how these work... but the fact that you have to put in body type (regular vs athletic), I'm inclined to think some of the calculation has to be based on standard calculations or tables of what the manufacturer defines as "regular" or "athletic."

    If it were me, I wouldn't be so concerned with the actual number itself... but rather the trend over time. The reading may not be accurate... but if it's somewhat repeatable, you might get a good idea of improvement over time. I would just be consistent of when and how you measure (mornings, before or after shower, ect...)
  • Leonidas_meets_Spartacus
    Leonidas_meets_Spartacus Posts: 6,198 Member
    Options
    Save your money and buy a calipers. It will get some time and effort to get used but much better than the body fat scales. The most accurate readings you can get are from DEXA scanning, Bod pod( NFL combine and pro athletes use it ) and ultrasound. Try to see if you can find a bod pod testing near, it usually costs 50$. You wouldn't test every day but before and after makes sense.
  • alibee88
    alibee88 Posts: 31
    Options
    Overall, they're not all that accurate. From what I understand, they can vary from a number of things (dirty or wet feet, ect...)

    If it's just on your feet, then I think it's only measuring across your legs... so you're upper body may be out of the calculation. They make some with hand and feet sensors... so I've heard they're a bit more accurate.

    I'm not entirely certain how these work... but the fact that you have to put in body type (regular vs athletic), I'm inclined to think some of the calculation has to be based on standard calculations or tables of what the manufacturer defines as "regular" or "athletic."

    If it were me, I wouldn't be so concerned with the actual number itself... but rather the trend over time. The reading may not be accurate... but if it's somewhat repeatable, you might get a good idea of improvement over time. I would just be consistent of when and how you measure (mornings, before or after shower, ect...)

    I don't think they just measure your legs because I'm majorly apple shaped and would struggle to even pinch any fat on my legs anywhere right up to my hips. But my scale measures about 35% body fat, which seems about right for me overall but definitely doesn't reflect the level of fat in my legs/lower half.
  • Fishshtick
    Fishshtick Posts: 120 Member
    Options
    People say these body fat scales are highly inaccurate and imprecise, but my own data suggests that is not entirely true. I can't attest to absolute accuracy, but my fitbit Aria has always been within a few percentage points of calipers or hand scans. However, for most of us accuracy is less important than precision because we are tracking trends. Despite all the comments I see here that these scales are imprecise and all over the place depending on your hydration etc, my empirical data suggests exactly the opposite.

    The day to day variation I see in my Arias' BF% measures is actually FAR less than in my weights! Seriously, my BF% plot looks like a ski slope with fairly subtle ripples and the occasion week long blips (no comment...I'm a less than perfect dieter). By all appearances it shows what looks like very stable data. I'm talking maybe 0.2% variation day to day (less within a day based on the few times I have tried). In comparison, my day to day weights look like a jagged saw blade and it is not uncommon for me to see 2lb changes day to day depending on my sodium intake and other less seemly causes (mostly shall we say bowel related).

    So how do folks reconcile these long term daily observations with the pervasive claims that these scales are too imprecise to be useful? Again, I am not saying these devices are more accurate than other methods, just that I don't see a reason to not believe the trend. Indeed, I'll even put the suggestion out there that my data would perhaps indicate that the BF% data is perhaps a more reliable indicator of what is actually happening with my body composition than the weight data.

    I can think of one possibility. Maybe the Aria scale is using some sort of trend smoothing algorithm on the BF% data to look more precise? To be honest I have even suspected this for some of the weight data when I have gone like 4 days at the same weight to the last decimal place or when I have gotten off and on the scale repeatedly and it always goes to the same 0.1 lb after hopping around a bit. I say this because I work with high end balances for my job and most such balances will show some variation at the last decimal place upon repeated measures due to subtle effects of sample position etc. A foot scale should easily show variation of 0.1 lbs upon repeated measure just due to wobble from our attempts to balance while standing. But the scale seems to filter that wobble out amazingly well...maybe too well. It would not be difficult to add a bit of a 'statistical' prior into a scales's software so that it would appear more precise than it really is.

Categories