Body Fat Calculator (US Navy Method)

Calculator

What the tape method measures, and why circumferences predict fat at all

Nothing here measures fat. A tape measures the outside of you, and the equations convert that shape into an estimate of body density: how heavy you are for your volume. Fat is less dense than muscle and bone, roughly 0.9 against 1.1 grams per cubic centimetre, so a body with more fat in it floats better and reads as a lower density. The Siri equation then turns density into a fat percentage. That two-step chain is why the method is properly called a body density estimate, and why it inherits every assumption in the density model.

Circumferences work as a predictor because fat is not stored uniformly. In most adults the abdomen is where surplus fat accumulates first and visibly, while the neck stays comparatively stable, being mostly muscle, airway and bone. The difference between a waist and a neck therefore tracks stored fat more faithfully than either measurement on its own, and it does so while cancelling out a good deal of individual frame size. Height enters the equation to scale that difference against body length, since a 90 cm waist means something different on a 160 cm frame than on a 195 cm one.

For women the equation adds the hip circumference, because female fat storage is more subcutaneous and more gluteofemoral: distributed around hips and thighs where an abdominal tape cannot see it. That is not an assumption about anyone's build; it is what improved the statistical fit when Hodgdon and Beckett regressed circumferences against hydrostatic weighing in their female sample.

How to measure: this is where the accuracy lives

Neck: wrap the tape just below the larynx (Adam's apple), sloping slightly downward to the front. Keep it snug but not compressing the skin. Waist: measure at the navel, at the end of a normal exhalation, not sucked in, not pushed out. Hips (women): around the widest point of the buttocks, tape level with the floor. Use a flexible fibreglass tape, take each measurement two or three times and average.

Bad tape placement is the single biggest source of error in this method. One centimetre of optimism at the waist moves a man's result by roughly seven-tenths of a percentage point and a woman's by about half a point, which is how the same person can appear to shed two points of body fat between a Monday and a Friday without anything changing. Measure the same way, at the same time of day, and the trend over weeks becomes genuinely useful even if the absolute number is a few points off.

Do it first thing in the morning, after the bathroom and before eating or training. Waist circumference moves two to three centimetres across an ordinary day on food and fluid alone, which is more than a month of real fat loss usually shows. If you can, have someone else pull the tape: self-measurement at the back is where the tape rides up, and a tape that is 1 cm higher at the spine than at the navel reads low.

• Flexible tape only: a steel builder's tape cannot follow the curve and a stretched cloth tape reads short.
• Snug enough to stay put, loose enough not to dent the skin. Indentation means the reading is too small.
• Check in a mirror that the tape is level all the way round, front and back.
• Average two or three passes, and re-measure from scratch each time rather than nudging the tape.
• Record the raw centimetres, not just the percentage, so you can re-run any equation later.

Where the equations come from

James Hodgdon and Mary Beth Beckett published the circumference equations at the Naval Health Research Center in San Diego in 1984, in two technical reports: 84-11 for men and 84-29 for women. They measured serving Navy personnel, took hydrostatic (underwater) weighing as the reference body density, and fitted log-transformed circumference terms against it. When the Institute of Medicine reviewed the cross-validation figures in 1990, the male equation correlated with hydrostatic body fat at r = 0.89 with a standard error of measurement of 3.63 percentage points, and the female equation at r = 0.84 with 3.82 points: errors the review described as comparable to the generalised skinfold equations in common use.

The Department of Defense adopted those equations across the services, which is how a piece of 1984 naval research became the tape test used at military installations worldwide, and then (through calculators like this one) the most widely used body fat estimate outside a clinic. The coefficients you see below are the ones from the original reports, so any calculator that produces a different answer from the same measurements is either using a different form of the equation or has made a unit error.

The unit trap that makes other calculators disagree with this one

Hodgdon and Beckett's work is published in two equivalent forms, and they are calibrated on different units. The body density form (1.0324 minus 0.19077 times the log of waist minus neck, plus 0.15456 times the log of height) is defined on centimetres. The direct percentage form (86.010 times the log of waist minus neck, minus 70.041 times the log of height, plus 36.76) is defined on inches. Feed the same body through the matching pair and they agree within about two-tenths of a point. Mix them up and they do not.

The size of that mistake is easy to demonstrate. For the worked example below (a man of 180 cm with a 38 cm neck and a 92 cm waist), the correct answer is 21.2%. Push his measurements in inches through the centimetre density form and you get 14.7%. Push his centimetres through the inch-based direct form and you get 27.8%. That is a swing of thirteen percentage points across the same tape readings, and it is the reason results from different sites can be six or seven points apart with no bug in sight. This page carries centimetres through the density form and cross-checks the inch form in its test suite, so the two paths have to agree before anything ships.

How accurate it is, and who it misreads

For an ordinary adult of ordinary build, expect to land within three to four percentage points of a DEXA scan. That is good enough to tell you which half of the scale you are on, good enough to track change over months, and not good enough to argue about a decimal place. It is also better than it sounds: the reference methods disagree with each other by a couple of points, so 'true' body fat percentage is itself a slightly fuzzy quantity.

The error is not evenly spread, though, and knowing which way it leans for your build is more useful than the number itself.

• Lean, muscular men: the most documented failure case. A thick neck raises the neck term and a muscular abdominal wall keeps the waist from shrinking, so the estimate typically reads high; this page flags the caveat below 10% for men and 16% for women.
• People with narrow waists and high visceral fat: the reverse case. Fat stored inside the abdominal cavity rather than under the skin can leave the waist modest while metabolic risk is real, and the tape reads low.
• Very tall and very short adults: the height term was fitted on a military population with a limited spread, and behaves less well at the extremes.
• Pregnancy: the equation is meaningless, because abdominal circumference no longer reflects stored fat.
• Anyone under 18: growth changes body composition continuously, and adult equations have no valid interpretation. Use BMI-for-age percentiles instead.
• Post-surgical swelling, ascites or significant oedema: all inflate a circumference without adding fat.

Read the result as a range, not a point

The result panel prints your estimate with a ±3.5 point band around it, and the band is the honest statement. An estimate of 21% means 'probably somewhere between 17.5 and 24.5, most likely near 21'. Two people whose results differ by a point are, for practical purposes, indistinguishable. The same applies to your own readings across a fortnight: if the number has moved less than a point, nothing has happened that a tape measure can detect.

This matters most at band boundaries. The ACE categories shown beside your result have hard edges (18% and 25% for men, 25% and 32% for women) and a single tape reading can sit on either side of one of those lines without saying anything different about you. Bands are a vocabulary for describing ranges of numbers, not labels for people. What the number does well is direction: measured the same way each month, a series of estimates tracks real change reliably even when the absolute level carries a three-point uncertainty.

What the number means

Essential fat (the minimum needed for hormones, organs and basic function) is about 2–5% in men and 10–13% in women; nobody should aim to live there. The ACE bands shown with your result give the usual interpretation ranges: athletes typically sit around 6–13% (men) or 14–20% (women), and 'fitness' ranges above that are entirely healthy places to be. Lower is not automatically better; hormonal and immune problems appear at the extremes.

Those bands are interpretation conventions rather than clinical thresholds. Unlike BMI, body fat percentage has no single internationally agreed cut-off for health risk, partly because it is so much harder to measure at population scale and partly because where the fat sits matters as much as how much there is. That is why a waist-based measure belongs alongside this one: two people at 25% body fat, one carrying it subcutaneously and one viscerally, have measurably different cardiometabolic risk profiles and an identical percentage.

If you have added your weight, the panel also converts the percentage into kilograms of fat and lean tissue, and into a fat-free mass index. Lean mass is the more actionable half of that pair: it is what you are trying to protect during a deficit, and it is what changes slowly and legibly when training is working.

Why we show a second estimate, and what agreement tells you

Every anthropometric equation is a fitted model with its own blind spot, and the blind spots do not overlap. The tape method sees your circumferences but not your mass. Relative fat mass sees only height and waist, and reads a muscular trunk as fat. Deurenberg and CUN-BAE start from BMI, so they see mass but cannot tell muscle from fat at all, and they run high for lifters and low for the sedentary and slight. Running all four on the same person and laying them side by side costs nothing and turns one opinion into four.

Where they cluster within two or three points of each other (which is the common case for an average build), you can treat the middle of that cluster as a reasonable working figure. Where they scatter by more than about six points, something in your build is breaking one of the models, and the usual culprit is muscularity: BMI-based equations call a lifter fat, the tape method calls the same lifter lean, and the truth sits nearer the tape. In that situation the honest answer is that a home estimate cannot settle it, and a DEXA or BodPod scan is the only thing that will.

There is a second, subtler use for the spread. Because each equation depends on different inputs, a large disagreement is often a measurement error rather than an unusual body: a neck taped too low, a waist taken at the narrowest point rather than the navel, or a height entered in the wrong unit. If the four numbers on this page look wild, re-measure before you re-plan your training.

What a body fat percentage is actually good for

It answers one question that weight cannot: how much of you is fat, and by subtraction how much is everything else. That makes it the natural metric for anyone whose weight is being pushed around by training: a lifter gaining 4 kg while dropping two points of body fat has had an excellent quarter that the bathroom scale describes as a failure. It is also the input to lean body mass, which sets protein targets, some drug doses and the maintenance calorie estimates that actually respect body composition.

What it is not good for is self-assessment against an ideal. There is no ideal body fat percentage, only ranges that are compatible with health, ranges that are associated with athletic performance in particular sports, and ranges at the extremes where endocrine and immune function start to suffer. A percentage is a description of tissue, and it makes a poor goal in its own right compared with the behaviours that move it.

How to track it over months

Measure once a month, not once a week. The method's noise is larger than a fortnight of genuine change, so weekly readings mostly measure your hydration and your mood. Fix the conditions (same morning routine, same tape, same landmarks, same person pulling it) and write down the raw circumferences alongside the percentage, because the raw numbers survive a change of equation and the percentage does not.

Judge by three or four readings in a row rather than by the newest one. A run of 24, 23, 23, 22 is a real downward trend; a single drop from 24 to 22 is more likely to be tape placement. And expect the pace to be modest: losing a percentage point of body fat means shedding roughly a kilogram of fat for an 80 kg adult, so half a point a month is a healthy, sustainable rate rather than a slow one.

If you want a second, independent series to compare against, a monthly waist measurement is free and needs no equation at all, and a three-site skinfold test tracks change well if the same person takes the folds each time. The point is not to find the one true method; it is to have two series that move together, because two methods agreeing on a direction is far stronger evidence than either method's absolute value.

How it's calculated

Worked example: a man of 180 cm, 82 kg, aged 35

1. Take the measurements: height 180 cm, neck 38 cm, waist 92 cm, measured three times each and averaged.
2. Subtract neck from waist: 92 − 38 = 54 cm. This difference, not the waist alone, is what the equation reads.
3. Log and scale that term: log10(54) = 1.7324, and 1.7324 × 0.19077 = 0.3305.
4. Log and scale the height term: log10(180) = 2.2553, and 2.2553 × 0.15456 = 0.3486.
5. Assemble the density: D = 1.0324 − 0.3305 + 0.3486 = 1.0505 g/cm³.
6. Convert with Siri: 495 ÷ 1.0505 = 471.2, minus 450 = 21.2% body fat.
7. State it as a range: 17.7% to 24.7%, which falls in the ACE 'average' band for men (18% to just under 25%).
8. Cross-check against the inch-based direct form: 86.010 × log10(21.26) − 70.041 × log10(70.87) + 36.76 = 21.3%, agreeing to within a tenth of a point.
9. Bring in the second opinions: relative fat mass gives 24.9%, Deurenberg 22.2% and CUN-BAE 23.2% at a BMI of 25.3. That is a spread of 3.7 points, which is narrow, so the cluster around 22–23% is a fair working figure.
10. Convert to tissue at 82 kg: 21.2% is 17.4 kg of fat and 64.6 kg of lean mass, a fat-free mass index of 19.9.

Where this number is used in the real world

• Military and emergency-service screening, where a tape is the only body composition tool that scales to an entire organisation.
• Weight-class sports (boxing, wrestling, rowing, martial arts) where athletes need to know how much of a cut is fat and how much would be muscle and water.
• Physique and strength training, where the useful question during a bulk or cut is what happened to lean mass, not what happened to weight.
• Clinical nutrition, where lean body mass derived from a fat percentage informs protein requirements and some drug dosing.
• Personal tracking through a deficit, as the metric that distinguishes fat loss from the scale simply moving.
• Cross-checking a smart scale or body composition monitor, whose bioimpedance reading swings with hydration and rarely states its own error band.
• Research and survey work where DEXA is unaffordable but circumferences were recorded anyway.

Frequently asked questions

Keep going

A single number rarely tells the whole story. Alongside the body fat result, the RFM calculator, the skinfold calculator, the lean body mass calculator, the FFMI calculator and the army body fat calculator each add a different angle on the same measurements. For the reasoning behind the numbers, read Navy formula accuracy, Body fat methods and Track without DEXA.

Sources

1. Hodgdon JA, Beckett MB. Prediction of percent body fat for U.S. Navy men and women from body circumferences and height. Naval Health Research Center Reports 84-11 and 84-29, 1984. apps.dtic.mil/sti/citations/ADA143890
2. Institute of Medicine (US) Committee on Military Nutrition Research. Body Composition and Physical Performance: Applications for the Military Services. National Academies Press, 1990: cross-validation of the Navy circumference equations. www.ncbi.nlm.nih.gov/books/NBK235939/
3. Brozek J, Grande F, Anderson JT, Keys A. Densitometric analysis of body composition: revision of some quantitative assumptions. Ann N Y Acad Sci 1963;110:113–40. doi.org/10.1111/j.1749-6632.1963.tb17079.x
4. Deurenberg P, Weststrate JA, Seidell JC. Body mass index as a measure of body fatness: age- and sex-specific prediction formulas. Br J Nutr 1991;65:105–14. doi.org/10.1079/bjn19910073
5. Gómez-Ambrosi J, et al. Clinical usefulness of a new equation for estimating body fat. Diabetes Care 2012;35:383–8. doi.org/10.2337/dc11-1334
6. Woolcott OO, Bergman RN. Relative fat mass (RFM) as a new estimator of whole-body fat percentage. Sci Rep 2018;8:10980. doi.org/10.1038/s41598-018-29362-1