Body fat calculator
The tape method is only as good as the tape work. Measure first thing in the morning before eating, and keep the tape level and snug without compressing. The waist is the one measurement that is not the same for everyone: the male equation was fitted on the abdomen at the navel, the female one on the natural waist at its narrowest point, and swapping them moves the answer by a couple of percentage points. Repeatability matters more than absolute accuracy here: the equation carries a standard error of roughly 3.5 percentage points at best, and independent testing has found more, but a consistent technique still tracks change well.
The category row uses the American Council on Exercise ladders, which differ by sex. Women: essential below 14%, athletic 14 to 20, fitness 21 to 24, average 25 to 31, above average 32 and up. Men: essential below 6%, athletic 6 to 13, fitness 14 to 17, average 18 to 24, above average 25 and up.
A tape-measure estimate from an equation fitted on US Navy personnel in 1984. Its own standard error is around 3.5 percentage points, and independent validation against multi-compartment models has found total errors above 5. It is far better at tracking change over time than at giving an exact figure, and it is not a body composition measurement.
This is the US Navy circumference method: body fat estimated from height, neck and waist for men, with hips added for women. It is a regression published in 1984, not a measurement of your body, and the figure it returns carries an error band of several percentage points. What it needs is a tape measure and nothing else.
How to measure body fat with a tape
The equations come from two Naval Health Research Center reports by James Hodgdon and Mary Beth Beckett, published in 1984: report 84-11 for men and 84-29 for women. They were fitted on serving US Navy personnel, with body density from hydrostatic weighing as the criterion. That derivation population is the first thing to know about the output. Navy personnel of the early 1980s were young, screened for fitness and narrow in build compared with the general adult population, and a regression is only as general as the people it was fitted on.
The second thing is the size of the error. A National Academies review of military body composition standards puts the standard error of the Navy circumference equations at approximately 3.5 percentage points of body fat. Independent validation has been less kind. Moon and colleagues compared four military circumference equations against a Siri three-compartment model in 31 college-age men and found the Navy and Air Force equation produced a total error of 5.2 percentage points, which they classified as unacceptable for estimating an individual. Read the number this panel gives as a band several points wide, centred somewhere near the truth.
The mechanism explains where it fails. The male equation runs on waist minus neck, and the female equation on waist plus hip minus neck, each scaled against height. Neck stands in for frame and lean tissue. So a heavy lifter with a thick neck reads leaner than they are, and someone with a naturally slight neck reads fatter, with no way for the equation to tell the two apart. The errors are consistent for a given person, though, which is why the method is far better at tracking a trend across weeks than at settling a single number. Two measurements a month apart, taken the same way, are the useful output.
The Boer lean mass row is on the panel as a second opinion that ignores the tape entirely: it estimates lean mass from height and weight alone. Where it disagrees sharply with the lean mass derived from your circumferences, one of the two is being misled by your build, and that disagreement is more informative than either figure by itself.
What people use it for
- Tracking composition change during a training block
- Getting a fat estimate that BMI cannot give at all
- Estimating lean mass for a nutrition target
- Comparing this month's tape against last month's
- Sanity-checking a smart scale reading
Questions
Its own standard error is about 3.5 percentage points. One independent study against a three-compartment model measured a total error of 5.2 points in college-age men and judged that unacceptable for individual estimation. Consistent technique still makes it useful for tracking change.
No. DEXA measures tissue directly; this predicts a body density from three tape measurements and converts it. They will not agree, and where they differ the scan is the better number.
Serving US Navy personnel, measured against hydrostatic weighing, in two 1984 reports. That population was young, fit and narrower in build than adults generally, so the further you sit from it, the looser the estimate.
The category row here follows the American Council on Exercise ladders. For men: 6 to 13% athletic, 14 to 17% fitness, 18 to 24% average, 25% and above the top band. For women each figure sits about eight points higher: 14 to 20% athletic, 21 to 24% fitness, 25 to 31% average, 32% and above the top band. Below the athletic band is essential fat, roughly 2 to 5% for men and 10 to 13% for women. These are conventions rather than clinical thresholds.
Scales use bioelectrical impedance, which changes with hydration, food and time of day. Neither method is a reference standard. Both are usable for trends and neither settles an argument.
Women do; the female equation uses waist plus hip minus neck. The male equation has no hip term, so the field is ignored when male is selected.
Because it is the equation's only proxy for frame size and lean tissue. A centimetre on the neck moves the result noticeably, which is both how the method works and where it goes wrong for unusual builds.
Morning, before eating, after using the bathroom, at the same time of day each time. Waist circumference varies by a centimetre or more across a normal day.
No. Take a second one a few weeks later with identical technique and act on the direction of travel. A single figure sits inside an error band wide enough to cover two whole categories.