Solved

Body composition innacuracy


Hi there.
I've been using the TomTom Touch for just over a week now, but have serious reservations about the accuracy of the body composition analysis.
I am a fit athlete with very, very little spare flesh on me at all, yet my Touch has been recording a consistent body fat percentage of 19% for the last week.
Very approximately, using my height, weight and waist size, I'd expect this to be much nearer to 9% - 12%.
I ensure that my arms are in the correct position as all the instructions show, so is there anything that could be making it so very far out?
icon

Best answer by tfarabaugh 8 February 2017, 19:28

Simon11R wrote:

Hi there.
I've been using the TomTom Touch for just over a week now, but have serious reservations about the accuracy of the body composition analysis.
I am a fit athlete with very, very little spare flesh on me at all, yet my Touch has been recording a consistent body fat percentage of 19% for the last week.
Very approximately, using my height, weight and waist size, I'd expect this to be much nearer to 9% - 12%.
I ensure that my arms are in the correct position as all the instructions show, so is there anything that could be making it so very far out?



Every body fat composition device takes a measure differently, some measure the whole body (if you stand on sensor pads and hold them in your hand as well), others (like the Touch) are predominantly looking at the upper body as they are using the hands (arms) only. Since the torso has a much higher fat content than the rest of the body (this is where most people keep fat stores, plus the organs are all protected by visceral fat), a torso dominated measurement will generally be a higher number. The important thing to look at is not the actual number itself, but at the trend. Is the number going up and down over time (looking at weeks and months, not days)? That is the key, to look at the trend, not at a specific number. The measurement is also heavily influenced by the time of day you take it, your hydration level, where you are on your menses cycle (for women) and other factors that it will vary day by day. By looking at it over the long term and seeing how it is trending you will get value out of it.

I am a competitive athlete with body fat in the single digits (8%-9% generally) and the Touch shows me around 15%. But it is consistently off by that percent, so knowing that it is always going to be offset, I do not worry about the specific number but more if the number is improving or getting worse.

I hope this helped answer your question. If so, please mark it as a solution so others can look for it if they have the same question.

View original comment

11 replies

Userlevel 4
Badge +2
Simon11R wrote:

Hi there.
I've been using the TomTom Touch for just over a week now, but have serious reservations about the accuracy of the body composition analysis.
I am a fit athlete with very, very little spare flesh on me at all, yet my Touch has been recording a consistent body fat percentage of 19% for the last week.
Very approximately, using my height, weight and waist size, I'd expect this to be much nearer to 9% - 12%.
I ensure that my arms are in the correct position as all the instructions show, so is there anything that could be making it so very far out?



Every body fat composition device takes a measure differently, some measure the whole body (if you stand on sensor pads and hold them in your hand as well), others (like the Touch) are predominantly looking at the upper body as they are using the hands (arms) only. Since the torso has a much higher fat content than the rest of the body (this is where most people keep fat stores, plus the organs are all protected by visceral fat), a torso dominated measurement will generally be a higher number. The important thing to look at is not the actual number itself, but at the trend. Is the number going up and down over time (looking at weeks and months, not days)? That is the key, to look at the trend, not at a specific number. The measurement is also heavily influenced by the time of day you take it, your hydration level, where you are on your menses cycle (for women) and other factors that it will vary day by day. By looking at it over the long term and seeing how it is trending you will get value out of it.

I am a competitive athlete with body fat in the single digits (8%-9% generally) and the Touch shows me around 15%. But it is consistently off by that percent, so knowing that it is always going to be offset, I do not worry about the specific number but more if the number is improving or getting worse.

I hope this helped answer your question. If so, please mark it as a solution so others can look for it if they have the same question.
Thanks so much for the feedback.
I'll take what you've said onboard, and look for trends rather than obsess about the numbers themselves!
The mine is consistently showing 26.5%. It looks like the Touch is off by 6 - 9 %. Do you have any statistics about the avarage discrepancy?
Userlevel 4
Badge +2
6% - 9% as compared to what? How do you know the other measurement is the accurate one and the Touch is wrong? Again, the actual number is irrelevant, it is the trend over time that matters.
By 6% - 9% I mean that my actual/real BF% is around 19% - 20% and the Touch is showing 26.5%.
The trend is important yes but the right presentation as well. In pecentages the touch is off by some 40% - 50%
Userlevel 4
Badge +2
How are you measuring your actual BF%? How accurate is that measurement? The Touch is impacted by the fact that it is using the arms, as it takes a more torso-impacted reading, which is where most people store much of their fat, versus one going from feet to hands, which brings in the leaner legs. I have never seen a study that correlates a set difference between the measurement types unfortunately, but I also found a 6%-9% difference (my historical "accurate" number if 8%-9% and the Touch reads around 15%).
Normaly I do a 3 point measurement on the chest, belly and upper leg.
Userlevel 4
Badge +2
Caliper testing is even less accurate than electrical impedance. It is too open to user error (not getting the same spot every time, squeezing a bit too hard or not enough, misreading the calipers, etc.) and three points is too small a sample for accuracy. Unless you are testing under clinical supervision, using water dunking or body scanning, you are as good as guessing (the difference between methods is how educated the guess is). In your case I would actually guess the watch is more accurate, but there is no way to tell.
tfarabaugh wrote:

Caliper testing is even less accurate than electrical impedance. It is too open to user error (not getting the same spot every time, squeezing a bit too hard or not enough, misreading the calipers, etc.) and three points is too small a sample for accuracy. Unless you are testing under clinical supervision, using water dunking or body scanning, you are as good as guessing (the difference between methods is how educated the guess is). In your case I would actually guess the watch is more accurate, but there is no way to tell.


If that was true than everybody (dietist, bodybuilders and other athletes) would use this great device called TomTom instead of a multipoint measurements.
i was losing body fat. i have done alot of bodybuilding over the years but quitting smoking caused be to get some puppy fat. the tom tom was telling me that my body fat was dropping and muscle was increasing. then all of a sudden it jumped up 5% body fat in a week. i had lost weight that week so its not an eating thing. the fat composition is now identical to what my bmi would be for height and weight of an average build person. does the body composition still work? in my case, the tom tom reading should not be identical to the "norm" its like the body composition part was given up on in an update and height and weight measurements used instead.
in addition. if i change the weight dramatically, it changes the percentage of fat. it does that without even redoing the test. i connect the tomtom and it syncs with the same readings and the fat/muscle amount changes alot. shouldnt the fat reading be the same?

Reply