A Quick Guide To Estimating Body-fat Percentage
Finding your body-fat percentage is an imprecise endeavor.
The methods that we generally have available to measure it range from being ‘acceptably accurate’ to ‘very poor,’ and they are nearly all useless for tracking changes over time. – They’re all marketed as being really accurate of course, but that’s just because people are after your dollars.
The lack of a guide on this site to help you find this has been something that has bugged me for some time, however it seems like I’ve found a fix for that – a quick calculation to help you to get a good estimation of your body-fat percentage that can be used to help you set up your diet more precisely.
Estimating Body-fat Percentage
It doesn’t really matter what your body-fat percentage is, you just need to get an estimation of it for your calorie intake calculations and setting your fat and protein intake targets. We don’t need to be exact, we just need to not be too far off.
Here’s a rundown of the methods we typically have available to us for measuring body fat, from most to least accurate. I’m basing this on their standard error of estimate (SEE) – which basically means how accurate they are for most people:
- Die in a non-too-messy way that requires an autopsy.
- DEXA scan, ~1-2% (Expensive, inconvenient.)
- The US Navy Equation based on body measurements below. ~3% (Convenient, free.)
- BodPod / Underwater weighing, ~3% (Expensive, inconvenient.)
- Body-fat caliper measurements. Skilled practitioner, ~3%. Non-skilled practitioner ~5%.
- BIA machines (Omron, Tanita), usually found in commercial gyms, 5-8% (Best avoided.)
Here’s that US navy formula. Take a look and then we’ll get into some practical recommendations:
US Navy Method of Body-fat Estimation
(For Men)
METRIC:
IMPERIAL:
Measurement Guidelines
- Height – Get someone else to do it for you if possible.
- Stomach – Measure at the navel. Have a relaxed stomach, exhaled but, don’t forcefully push it out!
- Neck – Keep your head straight, look forward. Keep your shoulders down and relaxed.
- Measure three times for each and average the three.
Simple as that.
Recommendations
The US Navy equation wins out on convenience and cost, this is what I’d recommend for most people. Notice that the reading is heavily determined by the measurement at the navel. Two important things to point out in regards to this:
- Fat comes off of the stomach/torso generally from the top down. Past the point of 10% body fat, the mid-stomach measurement (that at the navel, which the equation uses) will change very little as the person gets leaner, because most of the fat is coming off from the lower abs and back at this point. Therefore, if you’re already very lean, this won’t work well for you. It will be best to estimate from pictures, or use calipers.
- It will give those with particularly thick (well developed) abs, or a bloated abdomen higher readings than reality.
Now, let’s put vanity concerns aside for a moment and focus on practical concerns of getting body-fat percentage incorrect. What does that mean for your calculations?
Well, let’s say that you are a 80 kg (176 lb) male, at 15% body fat, but you calculated it to be 20%. You’ll underestimate your calorie needs by ~115 kcal. You’ll eat ~10g less protein than may be ideal, and your fat intake may be 5-8g lower than it should.
Is this a big deal? No. So I really wouldn’t worry about it too much.
Anticipated Questions
Do you have the calculation for women?
Sure thing. The reason it wasn’t included is because I know that the majority of the site’s readers are men. Here are the full, imperial calculations for both men and women:
- Men: Body-fat % = 86.010 x log10(abdomen – neck) – 70.041 x log10(height) + 36.76
- Women: Body-fat % = 163.205 x log10(waist + hip – neck) – 97.684 x log10(height) – 78.387
“This looks inaccurate for me…”
Yes, you may be right. It’ll work well for the majority of people, but you could be the exception to the rule. If this can be used to get close to what is correct, then that will do, because it’s as good as or better than whatever else most people have available.
If you believe that you have a better measurement from elsewhere then by all means, please use that. Remember, this is not being suggested a means of tracking your progress but just a start point for which to base calculations, which will need to be adjusted based on real world progress anyway.
Now, the comment section is going to develop a natural selection bias towards people complaining about this calculation being wrong because people rarely write to say that something worked out right, and no-one ever complains if they are told they are leaner than they thought. So, before writing to tell me that, here are some reasons for inaccuracies that may be worth considering:
- Measurement error,
- Comparison with some other measurement device that was also inaccurate.
- You are the relatively rare exception to the people that it will work well for.
If you’d like to read the full US MoD paper on the way the formulas were developed so you can make your own decision, you can get that here.
How do you gauge body-fat percentage when setting things up for clients?
I just do it by eye, looking at photos, but I’m an experienced coach so I know what to look for and ask for when I get people to take those photos. Obviously most people reading this don’t have that available as an option, hence this guide.
If I post a photo link in the comments can you tell me what you think my body-fat percentage is?
Sorry, but no. I’ll just be flooded with requests if I do and that’s not how I want to spend my time. Besides, I think talking about specific numbers puts the focus on the wrong thing – what matters is how the body measurements combined with scale weight, strength and appearance are changing over time. Make sure you don’t screw up your chances of being successful, check out my detailed progress tracking guide here.
******
Hope this was helpful. Questions welcomed in the comments. – Andy
Get The RippedBody.com Starter Kit:
1. Macro calculator
2. 'The Complete Guide To Setting Up Your Diet' book
3. Email course on the 5 biggest set-up mistakes people make.
(Yes, it's all free.)
Find my 'Complete Guide To Setting Up Your Diet' book useful?
↓ Take your physique to the next level ↓
The Last Shred: How To Adjust Your Diet Like A Pro To Achieve Single Digit Body Fat
Stop second guessing yourself.
→ 77 pages, Real data from 5 clients guided to shreds
About the Author
49 Comments on “A Quick Guide To Estimating Body-fat Percentage”
Andy, all of your articles are extremely well written and full of great information. Definitely the most precise and enjoyable online coaching I’ve came across thus far. Thank you for your dedication!
Thanks for writing, A Rog. 🙂
PS: Sorry for the delay in replying. I had been unable to do so while the website went through a big update over August.
I can vouch for this calc. I had a DEXA scan and this calc was within 0.3% !
So, if you are a man with a typical fat distribution in the torso it is as accurate as DEXA and free !
Hi Andy,
For an optimal nutrition partitioning in realizing muscle mass, what bodyfat % do you recommend? Max 15%? for men?
Is this also a issue for women? If so, whats the maximum bodyfat %?
For men 10% bodyfast is pretty ripped, what is the equivalent for women?
Thanks in advance Andy.
Hi Teun, thanks for the questions. In order:
– The leaner the better, down to around 9-10%. Beyond that there is probably little benefit.
– Yes. / There is no maximum, it’s in degrees.
– Add 8%.
What does the ~ 3 % mean for the us navy method does that means add three to your bf percentage
Hi Seif, that’s the standard error of estimate (SEE), which basically means how accurate it is for most people. You can expect your estimate to be within 3% of your actual body fat percentage. Thank you for asking, I’ll rewrite them as ~3% to ~3%(+/-) which will be clearer.
Andy Hi,
I’m a little bit confused. Why the narrower the neck the higher the BF ? Shouldn’t be the opposite since a more narrow neck means a thinner neck and a thinner overall body ? Thanks.
Thicker necks usually correlate with more muscle.
Check out this method for estimating body fat: https://strongur.io/calculator.html
All it uses is height, weight, max bench & max squat. I’ve been helping beta test his new app (it’s pretty awesome), and the body fat calculator is amazing accurate. It uses your max lifts to estimate your lean mass, and then backs into your body fat the other way.
The calculations I get from it are right in the ballpark of the Navy method, only without the measuring.
[…] – Estimating BF% […]
Put me right at 14.7% which is pretty accurate. Thanks for this article man!
Most welcome Alex.
Hi Andy, I’m a little confused. The standard error estimates that you list above for each method of measuring body fat are the absolute errors in body fat percentage and not the percentage error for the body fat percentage?
Ie. for a measured body fat percentage of 12% for example an 8% measurement error seen on a BIA machine would mean that the actual body fat might be anywhere from 12 +/-8 = 4% to 20% (absolute error) rather than 12 +/- (12 * 0.08) = 12 +/- 0.96 = 11.04 to 12.96.
That’s right.
[…] – Estimating BF% […]
What is log10?
Karl, save me my fingers and just google it eh bud?
Hi Andy,
Based on the navy calc, in 12.6% body fat. I’m 5’8″ ,weight 154lbs male, and am 45yrs old
I have been cutting for 6 weeks
What’s your recommendation of BF% before I can should end my cut to transition to my slow bulk?
Because I’m 45yr old in want to make sure all the same rules apply. Since I’m gonna do several bulk/cut cycles I want to know if getting extra lean (10%) is a waste or would be very beneficial (so I can have longer bulks
Thanks
Dan
Hi Dan, thanks for the question. Hard to give a black and white answer. I’ve covered the greys in this article:
• How To Adjust Your Diet To Successfully Bulk
You left out one method: ultrasound.
https://www.hindawi.com/journals/jobe/2013/280713/
Frank, thanks. An SEE of ~2% is very good. I’ve just shot statistics whiz James Krieger a message to ask him about this to get his opinion.
James came back to me almost immediately, here’s the exchange:
Me: James, what do you think of ultra sound (US) as an alternative? Is there something I’m missing from that review, or can we take it at face value (SEE ~2%, thus very good alternative)?
James Krieger: “Ultimately US is going to have the same limitations as other methods. SEE doesn’t tell you the how big errors can get on an individual basis…you need Bland Altman plots for that. Some of the studies referenced in that review used criterion methods (like DEXA) that in and of themselves have errors. Ultrasound is basically going to be a slightly more precise and reproducible form of skinfolds really as that is what it is detecting, but ultimately individual error levels will still be similar to other methods.”
I love how “Die in a non-too-messy way that requires an autopsy” is not noted as inconvenient. 🙂
Different people will take it in different ways. Some may find it upsetting. 🙂
Hey Andy, I actually found this formula a few months back, great way to get an initial estimation. I believe Greg Nuckols says it has a margin of error of about 3%, which could account for some of the higher than expected readings. My calculation is skewing leaner than I realistically think I am, estimating 9.6%. I’d put myself more in the 11-12% range. Lyle McDonald had an article not long ago about how shredded stage-ready bodybuilders looking the typically 5% bf were actually reading 8-9% on a DEXA scan. Like you said, the actual REAL number is unknowable and unimportant. The only situation I’ve found it useful is in trying to determine if I was lean enough to start bulking (10-12%) without negatively affecting calorie partitioning. Thanks!
Scott, thanks for taking the time to comment. Yes, 3% sounds right. I’ll check with Greg (he never forgets anything). Couldn’t find that in the paper to confirm.
Just to buck the selection bias, looks like it is within 1-point for me 🙂
Haha, thank you Mike.
I’ve come out at 19.6% which i’m a little unsure about having googled images for 19-20%, but they do say everyone underestimates it. I seem to hold a lot of my fat around my mid section though so it may have thrown the calculation slightly.
Altogether not a bad calculator for getting a rough idea.
Wow, this just gave me 16% BF. My measurements are spot on. You can see all my abs, obliques and serratus anterior very clearly. I have been advised from an experienced fitness model that I’m about 8%. Just thought I’d give you my positive feedback as this is way off for me.
Thanks for the great site and information Andy, it’s worth it’s weight in gold. Using your guides I have went from about 15% – 8% in a few months. I will be posting you some pics soon. All the best.
Scott
Thanks Scott. Look forward to seeing the pics.
Hi Andy
Tried this and I’m not sure it works for a person who lost a lot of weight (and who have a lot of skin around the belly). Last time I went to the doc (in August), he used his scale and he said I was at 15.2% body fat. With the US Navy Method I’m at 24.6%. It’s like 10% difference. Odd, isn’t it ?
Cheers
Phil
The loose skin will definitely be getting in the way Philippe.
Do you have any advice on how getting “rid” of it ? (the loose skin)
I’ve started the The Big 3 Routine not so long ago.
Thanks Andy!
In all bud the most extreme cases, it’s just a matter of waiting for it to come tight again.
Dude, this article needs to be updated. You are missing out on an amazing piece of technology called Skulpt Aim which measures body fat very accurately and is convenient and easy. I have had 2 Dexa scans whilst using it and it has been very accurate!
Hi George, thanks for the comment. I’ve had a look at that site, here’s what I’ve determined:
1. It’s a $150 dollar piece of kit that attaches to a smart phone. – So it’s a significant amount of cash to be spending for most people.
2. It works through a similar mechanism to the BIA device, but has more measurement sites thus claims to be more precise at measuring body fat. – This is actually misleading because BIA machines don’t measure body fat, they measure electrical impedance and then use a formula to estimate body fat from that. Now, it’ll be more accurate at measuring the impedance, but will still have the same issues when it comes to applying that formula to determine body fat. See the first paragraph and then “Error Compounding On Error” section of this article by James Krieger for more on that. The best analogy I can think of is it being kind of like a World War I artillery sheller trying to fix his accuracy problems by making the degree of precision that the barrel angle is set at half a degree higher, when the issue is the ever-changing wind.
3. It provides a degree of accuracy that is, quote: “Within 1-2% of gold standard, underwater weighing.” – Two issues with this. Firstly (and this is me neck-bearding a bit), DEXA is the new gold standard, but even that still has it’s issues. Secondly (and way more importantly), research shows that underwater weighing has pretty low errors on average, but individual error rate can get as high as nearly 5-6%, which is what is important to us, the individuals.
So, put that all together and people will be spending $150 for a piece of kit that may or may not be more accurate on any given day, which you’ll not be able to confirm unless you go get a DEXA scan (and thus spend more money and time).
Now, people go nuts for any gadget priced around $100 that they can convince themselves will help them towards their fitness goals – it gives the hedonistic pleasure of shopping, without the guilt – so I have no doubt that this will tell a ton, however I cannot recommend that people buy one.
I used this formula to track my fat percentage last year. The lowest I got to was 16.7% using the formula, yet at that point I had a Dexa scan which showed I was actually 11.9%. The Dexa scan quite clearly showed that most of my remaining fat was around my abdomen, and lopsided too. I noticed the formula wasn’t the best when I was visibly getting leaner everywhere but the stubborn abdomen fat wasn’t coming off so the calculation didn’t budge much. As a starting point though the calculations are decent enough.
Peter, thank you. Yes, I’d agree with that. The measurements at the navel will reach a lower threshold and then not move, despite getting leaner. This is because fat comes off of the lower abs/back the last. That is usually at the point where people get to 10% and lower though.
Thanks for the feedback, worth mentioning in the article.
Hi Andy,
Love the site and appreciate all the hard work you put into it. I just doubt the accuracy of this test. I had a DEXA scan done in January this year and it came back as 18% BF. Since then I’ve lost 6kg and inches off my waist while also significantly increasing my strength across all lifts. This calculation says I’m at 19.2% so saying I’ve gained fat!
I do plan on having another DEXA scan in another month or two and will give another comparison then.
Martin, firstly thank you for the comment. Glad to hear you find the site so helpful.
I’m sure that in your case you are right given what you have said – DEXA should be more accurate than this, and this should be considered an alternative, and an estimation only.
I’ve added a fuller answer in the FAQ section of the guide, so have a look above.
Andy.
[…] A Quick Guide To Estimating Body-fat Percentage […]
Way, way off in my opinion. I’m 5-11 182 lbs with visible abs and anterior delt/pec separation. With OrbiTape measurements and that calculation it has me at almost 19% body fat.
Pete, thanks for the feedback. Very interesting, I’d like to take a look to check. Can you link me to two pictures, front and side, arms by your side, tensed abs. No camera selfie done at arms length, set the camera up so that it is a distance away.
Also, tell me the measurements you inputted please. I’ll then have a look.
Note to others: This is just an offer for Pete.
I was wondering if you have seen any improvement in BIA devices? We have a fit bit Aria and one of the omron handheld devices. They are nearly always within 1% of each other (for me) and line up exactly with BF based on the US Navy calculation. I’m not advocating , just wondering if the tech is starting to improve.
The inaccuracies don’t come from flaws with the machines themselves, but the way in which they derive a calculation to estimate body fat based on what they measure. Hydration status & glycogen levels will affect the readouts, but as these things fluctuate from day to day and throughout the day the machines can’t be relied upon.
Hi Andy,
This is an interesting calculation which I wasn’t aware of. I’m from Holland, so maybe that is the reason. Can you share the formula (of the metric one)? I have build my own spreadsheet in which I keep all my statistics. And I would like to embed this formula in it too.
Thank you in advance,
Michel
Hi Michel, same, Greg Nuckols clued me into it a couple of weeks back when we were talking about this issue of helping people to find body-fat percentage.
Males.
% body fat = 86.010 x log10(abdomen – neck) – 70.041 x log10(height) + 36.76
Females.
% body fat = 163.205 x log10(waist + hip – neck) – 97.684 x log10(height) – 78.387
(All circumference and height measurements are in inches, so just multiply by 2.54 to put that into cm.)
Here’s link to the original US Department of Defense document for further reading.
Question welcomed. (Over 16,000 answered)
Comment rules: