Optimum body fat 10% on DEXA or 10% on calipers?

So it seems DEXA scan body fat measurements are often way higher than caliper measurements. Most charts generally put a male healthy athlete around 10% body fat being about right. Now do they mean 10% on the calipers or 10% on DEXA? 10% on DEXA I assume is pretty ripped right? BTW… On non athlete mode my Tanita puts me at 19% and I can see all my abs in the right light, that can’t be right can it? On athlete mode it says 12% which is where calipers put me too…

As I understood it, DEXA usually measures higher because it measures ALL the fat in the body (including intramuscular and visceral fat) whereas calipers only measure skinfolds, so subcutaneous fat. Those charts probably refer to caliper measurements, because this is just a lot more common form of measurement (because cheap).

1 Like

I recently changed my Garmin scale from athlete profile 7 to profile 8. I know we aren’t advised to do this, but all the metrics just kept going up and it didn’t “seem” right. What started to throw me off is the Water % number. It kept dropping down to almost 60% and the Body Fat had climbed all the way to 15%. But I’ve got a 28" waist, weight 155 lbs (about 3 pounds heavy right now). I have upped my training from about 5 hours a week to 9-12 hours a week and haven’t missed a workout this year. I feel stronger and healthier than I have in years and yet the body fat kept going up every single day.

I drink almost a gallon of water every day. I’m very sensititve to dehydration and can get “hangover” headaches from dehydration very easy. So when I see a water % down towards 65% I’m normally also feeling sluggish or have a headache. When I’m properly hydrated I almost always show 68%. But in the past 2 months as I increased training and continued to eat meatless, I look better and better in the mirror and yet my body fat on the scale goes up every day and my water goes down.

I finally switched to profile 8 from 7 and for the past week it has my water back to around 67-68% the way it has been for years and my body fat% shows me back in the 7-8% range.

Maybe this is all vanity, but since we are making guesses and assumptions I think sometimes we have to use our personal experience and make an educated guess. The best use of these things is just to measure trends. And I just felt I couldn’t understand the trend anymore when I’m eating meatless, working out 10 or more hours a week, drinking plenty of water, and getting 8 hours of sleep a night.

1 Like

I’ve always wondered about this too. On my Nokia/Withings scale what determines whether i should use the “Athlete” setting or the normal setting? On Normal i am 17% fat. On Athlete i’m about half this. What science is behind this vs just vanity of me being able to use a number that sounds better or not?

I’m 1.85cm @ 76Kg and can see my ribs & abs, in a triathlete body, so decent arms as well as legs. Does this correlate with 17% fat?

1 Like

Did you use any calorie counting app (MyFitnessPal, for example) to try to understand trend? Just going meatless and upping training volume doesn’t give you all the necessary info.

For example, this winter I increased training volume. As I was eating by feel, I actually under-ate very slightly, unintentionally decreasing weight. Because I still felt sufficiently full, didn’t get the reason until started monitoring actual caloric input.

Beating machine until it gives you correct numbers isn’t very helpful in long term :slight_smile:
Worse, now you can’t compare your numbers before and after twiddling with scale.

I only trust Coach Greg’s Lazer Eye Vision ™ to measure body fat. lol.

They are all a bit of a crap shoot. If you watch a bunch of fitness YouTubers documenting their Dexa scan results you will see a huge variance in appearance / measured body fat and also just some BF% that would see these people in graves. Dexa results can vary based on your hydration level and and glycogen depletion, just as a scale using electrical impedance does. Does my BF change by pounds day to day as my scale indicates? No, just hydration levels are changing. MRI is probably the most accurate but unless you have tons of disposable income you probably can’t afford it on a regular basis.

If you can deal with the ranting and Gilbert Godfried voice:

Probably the most simple way is to use a skin fold caliper and just track measurements at various parts of your body. Don’t use the measurements to estimate bodyfat (there are different models for different distribution of body fat, which do you use?) just track the measurements for progress.

I did, I use myfitnesspal connected to my Garmin. So my Garmin 945 tracks the rest of the “movement” and “steps” and all day HR for me. Which gives a little better look at total calories burned from movement. I had a personal trainer work up a BMR for me a couple years ago and that’s my baseline of calories. Most days I need anywhere from 2,800-3,400 calories to avoid a deficit. I didn’t just start meatless, I’ve been 95% meatless for years now. So I haven’t changed my diet at all, the only variable change was the increase in volume. With the increase in volume and running a calorie deficit most days (rarely more than a few hundred calories deficit) I have increased FTP every ramp test, increased endurance, lowered all my heart rate metrics, and increased strength since I’m weight lifting and running too. The only change is that my scale ended up nearly 10% higher on BF% than my last 2 BodPod results and my old tests when I was only doing about 5-6 hours a week.

Maybe with a calorie deficit and an increase in training I have gone up 7-10% in BF. It just didn’t seem likely. Whenever I changed the athlete profile from 7 to 8 it mirrored my BodPod results and brought the lines on the graph back towards the nearly straight line average I had for the 12 months prior to the last 2-3 months of continuous BF% increase.

Like you said, might not matter. I just had to make a decision because the last 2 months started to look like the outlier on my graphs compared to all the rest of the data. What was throwing me off the most is the fact that my Water% was down towards 60% and that just didn’t make sense to me since that number hadn’t changed in years and I have stayed on my nearly gallon of water a day regime the entire time.

Ok. Maybe you’re missing some other variable then?

What I’m saying is that you are right, those scales might not give correct numbers but you still can monitor long-term trends. If trend (not single number) is unexpected and before it was seemingly valid, then either you miss something else or scale is broken. Turning some knob doesn’t fix underlying issue.

When you get VERY low body fat (sub 6-8% caliper) the amount of water in your body actually is so low that the electrical impedance that is used to measure/estimate body fat no longer works properly. As such it will throw weird readings, so they needed to develop a different algorirthm for the calcs.
Based on what you have described you do not need athlete mode, and very few people do.

1 Like

Generally these days, it refers to Dexa or another method of measuring that is accurate and consistent.

Impedance scales are consistent but not necessarily accurate. Good for tracking changes over time, but you will probably want something more accurate eventually, even if just to calibrate mentally.

FWIW, once I dropped below 10% on the DEXA, my Withings scale just straight up stopped giving me a bodyfat reading.

My 2 cents as an exercise physiologist familiar with all of these measures:

10% on the normative charts almost always refers to hydrostatic weigh, skinfold measurements, or bodpod. These usually have very similar measurements with each other. All of these measures have fairly similar accuracies and reliabilities assuming the tester has done an appropriate job controlling conditions.

In the research community, we avoid any and all commercial impedance scales. We have ‘nicer’ impedance measures but we typically don’t trust what they say. They typically are not accurate or reliable unless you have perfect conditions. You can try your best to control your hydration but inflammation from hard workouts, water retention from carbs, and built-in sensitivity issues make them too difficult to use for most people.

DEXA is great and we have used it a lot, but for practical reasons. It always reads higher than any other method, but it is very consistent (reliable). Its accuracy is argued all of the time, but for studies where you are looking at changes in composition, its reliability is great.

From the top of my head, when the TR guys experimented with this, they found that the athlete setting closely tracks the body fat %age as determined by calipers whereas the normal mode is close to the numbers they got from Dexa scans. Of course, N = 3 or 4, so that doesn’t make a study, but gives an indication. As others have said, calipers only measure subcutaneous fat whereas Dexa gets everything.

10 % body fat sounds quite low. When I searched for this a while ago, there was no good agreement on what “good” ranges are, but if you get to the single digits, IMHO it is too low. Especially now when you have to watch your health and racing is less of a concern, I’d err on the side of a slightly higher body fat percentage.

Depends on the person, but I’ve found 9-10% Dexa to be pretty sustainable. Most offseason stable bf % for a lot of pros is around the 9-10%, dropping towards 7-8% for peak season.

I think every BF measurement methodology has its own inaccuracies. It’s like power meters - they are not necessarily accurate, but what’s more important is consistency over time so you can measure progress.

I’ve been within +/- 0.5% of 5% BF in three bodpod measurements over a 5 year period. Do I think that’s “right”? No. But it’s consistent, and the “mirror check” over that time period has been pretty consistent also, as well as my overall weight measured using the same digital scale.

I think most people generally have a good sense of if they need to lose some weight or not. So given the inherent inaccuracy of BF scales, I’d suggest just using the same scale every day/week/month, and track trends over time - with the goal obviously of moving the needle in the direction you want.

1 Like