Body Fat Calculator

How to use this tool

Pick your sex, then enter your height, weight and age, and the neck and waist measurements from a tape measure. Women also need a hip measurement, and the box appears when you choose female. Height and the tape measurements share one unit selector, so if you measure your waist in inches, set height in inches too. The result updates as you type.

The headline figure comes from the Navy circumference equation by default, because it is the more accurate of the two for most people. The BMI and age equation runs alongside it as a second opinion, and you can make it the headline if you have no tape measure to hand. Both results are shown with the standard error their original papers reported, drawn as a bar around each estimate, because the error is part of the answer rather than a footnote to it.

Below the headline you get fat mass and lean mass in your chosen weight unit, a fat free mass index, and a table showing what you would weigh at other body fat levels if your lean mass stayed where it is. The waist table and the side panel show how much the result moves for each couple of centimetres of waist, which is the most useful thing to understand about this method: the waist does almost all of the work.

Where to put the tape

The whole calculation rests on two or three numbers from a tape measure, so the measurement technique matters more than anything else on this page. The protocol below follows the one the equations were built on.

Neck. Just below the larynx, the Adam’s apple, with the tape sloping slightly downwards towards the front. Keep your shoulders relaxed and look straight ahead. Do not flare the neck.

Waist, men. Horizontally around the navel, level with the floor, at the end of a normal breath out. Do not suck the stomach in and do not push it out.

Waist, women. At the narrowest point of the torso, usually a little above the navel, again horizontal and at the end of a normal breath out.

Hips, women. At the widest point of the buttocks, with the feet together and the tape horizontal all the way round.

For each measurement the tape should sit snug against the skin without pressing into it. Take two readings and use their average, and if they differ by more than half a centimetre, take a third. Measure in the morning before eating if you can, and always in the same state, since a large meal or a day of salty food can add a centimetre to the waist on its own.

The two equations, and what they actually say

The circumference equations were published in 1984 by Hodgdon and Beckett at the Naval Health Research Center, and they remain the basis of the US Navy body composition standard. They were fitted by measuring sailors with a tape and then weighing them underwater, which was the reference method of the time, and finding the formula that best predicted one from the other. The equations work in inches and use the logarithm of the circumference difference, which is why a centimetre at the waist matters more for a slim person than for a large one.

For men the equation is 86.010 times the log of waist minus neck, minus 70.041 times the log of height, plus 36.76. Take a man 180 cm tall with a 38 cm neck and an 88 cm waist. Waist minus neck is 50 cm, which is 19.7 inches, and his height is 70.9 inches. That gives 86.010 times log 19.7, minus 70.041 times log 70.9, plus 36.76, which comes to 18.5 percent.

For women it is 163.205 times the log of waist plus hips minus neck, minus 97.684 times the log of height, minus 78.387. A woman 165 cm tall with a 32 cm neck, a 75 cm waist and 98 cm hips has waist plus hips minus neck of 141 cm, which is 55.5 inches, and a height of 65 inches. The equation gives 29.2 percent.

The second estimate uses the equation published by Deurenberg and colleagues in 1991 in the British Journal of Nutrition, fitted on 747 adults. It is 1.20 times BMI, plus 0.23 times age, minus 10.8 for men or nothing for women, minus 5.4. The same 180 cm man at 82 kg and 35 years has a BMI of 25.3, and the equation gives 22.2 percent, about 3.8 points above the tape measure figure.

The error bar, and why one reading is nearly useless

Every prediction equation has a standard error, which is the typical gap between what it predicts and what the reference method measured on the people it was fitted on. For the Navy equations that figure is about 3.7 percentage points. For the Deurenberg equation it is 4.1. Roughly two thirds of people land within one standard error of the equation’s prediction, and roughly one in twenty lands more than two away.

Put a reading of 20 percent through that and the honest statement is that the underlying figure is most likely somewhere between about 16 and 24. That range covers two or three of the five point fitness bands at once, which is why this calculator draws the error bar on the chart instead of printing a single number and a category as though they were exact.

The good news is that most of that error is about the individual, not the occasion. If the equation reads you three points high this month because of how your fat is distributed, it will read you roughly three points high next month too. That makes the method much better at tracking change than at establishing an absolute level, and it is why a series of readings taken the same way is worth far more than any single one.

When the two equations disagree

The BMI equation only knows your height, weight, age and sex. It has no way of telling whether weight above a typical frame is muscle or fat, so it assumes the average mix for the population it was fitted on. For the reference man above the two equations differ by 3.8 points, which is normal and well inside either error bar.

A gap of more than about six points is a signal. If the BMI figure is much higher, you are probably carrying more muscle than average, and the tape figure is the better estimate. If the BMI figure is much lower, you may be carrying less muscle than average, or the waist measurement may have been taken too tight. The comparison chart on the page shows both estimates with their bars so you can see at a glance whether they overlap.

The National Academies review of body composition standards in the armed services makes the wider point plainly: prediction equations are population specific, and different service equations produce substantially different estimates for the same person. That is not a flaw in any one equation. It is the nature of estimating something you cannot see from measurements taken on the outside.

Common measurement profiles

Exact arithmetic on both equations. The BMI column moves much less than the Navy column across a range of waists, because the BMI equation cannot see where the weight is carried.

Every row is a link that opens the calculator on that profile. Neck and hip measurements are typical for the waist and height shown; open a row to see them and change them.

Fat mass, lean mass and fat free mass index

A percentage on its own is hard to act on. Multiplied by body weight it becomes something you can: the reference man at 18.5 percent of 82 kg is carrying 15.1 kg of fat and 66.9 kg of everything else. Lean mass here means muscle, bone, organs, and the water in all of them, and it is the half of the reading most people ignore. On a diet it is the half worth watching, because losing weight while lean mass falls is a different outcome from losing weight while it holds.

Fat free mass index divides lean mass in kilograms by height in metres squared, the same shape as BMI, and it gives a way of comparing muscularity between people of different heights. The reference man comes out at 20.6. A normalised version adds 6.1 times the difference between 1.8 metres and your height, so short and tall people sit on the same scale. In a 1995 study of 157 male athletes, Kouri and colleagues found the normalised index among those who had not used steroids topped out at about 25. If this calculator gives you a normalised figure above that, the more likely explanation is a tape measurement that is a centimetre or two off.

What the scale would read at another body fat level

If lean mass stays where it is, the weight at any target body fat is lean mass divided by one minus the target. For the reference man, getting from 18.5 to 15 percent means reaching 78.7 kg, a loss of 3.3 kg or 7.4 lb, all of it fat. The table on the page does this for a range of targets.

In practice lean mass rarely holds perfectly still. A careful cut with plenty of protein and resistance training can keep it close; a crash diet without either will take some of it with the fat, and the scale will then overstate your progress. That is why the same measurement repeated through a diet is more informative than the scale alone: if weight is falling and the tape figure is falling with it, the weight coming off is mostly fat.

Mistakes that catch people out

Measuring the waist in the wrong place. Men at the navel, women at the narrowest point. Measuring a man at the narrowest point instead of the navel can take several points off the result.

Holding the stomach in. The equations were fitted on a relaxed waist at the end of a normal breath out. Anything else is a different measurement.

Pulling the tape tight. Snug against the skin, not compressing it. A tape pulled tight enough to dent the skin reads low by a centimetre or more.

Believing the second decimal place. One centimetre at the waist moves the reference man by 0.74 points and one inch moves him by 1.9. The error bar is several points wide. A reading of 18.4 and one of 18.9 are the same reading.

Comparing readings from different methods. A bathroom scale with foot electrodes, a gym machine, a set of calipers and a tape measure will all give different numbers for the same person on the same morning. Pick one method and stick with it.

Measuring too often. Body fat changes slowly. Every two to four weeks is plenty, and daily readings mostly measure water and lunch.

Popular measurement profiles

The combinations people look up most. Each opens the calculator on that profile with its own address, and every value can be changed once it is open.

Men, centimetres

Men, inches

Women, centimetres

Women, inches

The BMI and age method

How this was checked

The Navy equations were implemented from the published coefficients and checked two ways: once by converting centimetres to inches inside the calculation and once by feeding inches in directly, which gave identical results to the last decimal place. They were then run across men and women at heights from 150 to 200 cm and waists from 60 to 130 cm to confirm the output rises smoothly with waist, falls with neck and height, and never produces a figure where the logarithm is undefined. Where the waist is not larger than the neck, or for women where waist plus hips is not larger than the neck, the page says so instead of printing a number.

The Deurenberg equation was checked against the published form, with the sex term coded as 1 for men and 0 for women as in the original paper, and the standard errors of 3.7 and 4.1 points were taken from the source reviews rather than from secondary summaries. The sensitivity figures in the diagram were computed from the equations, not estimated: for the reference man, one centimetre at the waist moves the result by 0.74 points and one centimetre at the neck by 0.75 in the other direction. The live page was then opened at every address linked above and each was checked for a distinct title, the right headline, charts that rendered, and clean CSV, image and print output containing only that profile, at desktop, tablet and phone widths.

Common questions

How accurate is the Navy body fat method?

About as accurate as anything you can do at home without equipment, which is to say it gives a useful estimate with a real margin around it. The equations were fitted against hydrostatic weighing and carry a standard error of about 3.7 percentage points. In plain terms, a reading of 20 percent means the figure a lab would find is most likely somewhere between about 16 and 24. That is good enough to tell you roughly where you are and very good at telling you which direction you are moving, provided you measure the same way each time. It is not good enough to tell 18 percent from 20.

Where exactly do I measure my waist for body fat?

For men, horizontally around the navel. For women, at the narrowest point of the torso, which is usually a little above the navel. In both cases the tape goes all the way round level with the floor, snug against the skin without pressing in, and the reading is taken at the end of a normal breath out rather than with the stomach held in. The neck is measured just below the larynx with the tape sloping slightly downwards at the front, and for women the hips are measured at the widest point of the buttocks. Take two readings of each and use the average; if they differ by more than half a centimetre, take a third.

What is a healthy body fat percentage?

There is no single answer, but the classification most fitness professionals use puts the average range at roughly 18 to 25 percent for men and 25 to 32 percent for women, with the fitness range below that and the athletic range below that again. Essential fat, the minimum the body needs to function, is around 2 to 5 percent for men and 10 to 13 percent for women. These bands are conventions rather than medical thresholds, and a measurement error of three or four points means most people cannot place themselves precisely inside one of them anyway.

Why does the BMI method give me a different number?

Because it is working from different information. The BMI equation only knows your height, weight, age and sex. It has no way of telling whether the weight above a typical frame is muscle or fat, so it assumes the average mix for the people it was fitted on. If you lift and carry more muscle than average, it will read high. If you carry very little muscle, it will read low. The circumference equation at least measures where fat is actually stored, which is why it is the better of the two when they disagree.

What is fat free mass index?

Fat free mass index is lean mass in kilograms divided by height in metres squared, the same shape as BMI but applied only to the part of you that is not fat. It is a way of comparing muscularity across people of different heights. A normalised version adds 6.1 times the difference between 1.8 metres and your height, so short and tall people can be compared on the same scale. In a 1995 study of 157 male athletes, Kouri and colleagues found the normalised index of those who had not used steroids topped out at about 25, which is why that figure is often quoted as a practical ceiling.

How often should I measure my body fat?

Every two to four weeks is plenty. Body fat changes slowly, and a tape measure cannot resolve a change of less than a point or so from one reading to the next. What it can do is show a trend: measure at the same time of day, in the same state, with the same tape, write the numbers down, and after three months the direction will be obvious even though any single reading has an error bar several points wide.

Related Calculators and Reference

• TDEE Calculator: take the lean mass from this page into the Katch McArdle and Cunningham formulas, which are the more accurate ones once you know it.
• Macro Calculator: add your body fat figure and protein is set from lean mass rather than total weight.
• One Rep Max Calculator: for the training that keeps lean mass on while the fat comes off.
• Steps to Miles Converter: the daily activity side of the energy budget.
• Grams to Ounces Converter: for switching between kilograms and pounds of fat and lean mass.
• BMI Calculator: where your height and weight sit on the adult categories.
• BMR Calculator: the resting figure at the bottom of your daily burn, on five equations.
• Protein Intake Calculator: grams of protein a day for your weight and goal.
• Ideal Weight Calculator: ideal body weight for your height.
• Lean Body Mass Calculator: lean mass from four predictive equations, with no tape measurements needed.
• Waist to Hip Ratio Calculator: waist to hip and waist to height ratios from the same tape measurements.
• Deurenberg and others, British Journal of Nutrition 1991: the paper behind the BMI and age estimate.