Body Fat Calculator

What the tool measures, and what it does not

Four numbers go in: height, neck, waist, and for women the hip. Nothing else. No age, no weight, no scale, no calipers.

That has a consequence worth stating plainly, because pages like this one often imply otherwise. Without a body weight there is no way to convert a percentage into pounds of fat or pounds of lean mass, so this page reports neither and cannot. What it produces is one percentage and the band that percentage falls in.

It is a statement about four tape readings, not about you. It takes no view on health, fitness, appearance or what any number ought to be.

The US Navy circumference method, by hand

The men's equation, with every length in inches:

86.010 × log₁₀(waist − neck) − 70.041 × log₁₀(height) + 36.76

Take a 5 ft 10 in man with a 15 in neck and a 34 in waist at the navel.

1. 34 − 15 = 19.0, and log₁₀(19.0) = 1.278754.
2. 86.010 × 1.278754 = 109.98.
3. log₁₀(70) = 1.845098, and 70.041 × 1.845098 = 129.23.
4. 109.98 − 129.23 + 36.76 = 17.51, so 17.5%.

The women's equation is 163.205 × log₁₀(waist + hip − neck) − 97.684 × log₁₀(height) − 78.387. For a 5 ft 6 in woman with a 13 in neck, 27 in waist and 37 in hip: 27 + 37 − 13 = 51.0, log₁₀(51) = 1.707570, 163.205 × 1.707570 = 278.68, 97.684 × log₁₀(66) = 177.74, and 278.68 − 177.74 − 78.387 = 22.56, so 22.6%.

Both reproduce the worked examples above exactly, which is the point of showing the arithmetic rather than describing it.

The other published form of the same equation, and the unit trap

You will find the same method written a second way, as a body density followed by a conversion: 495 ÷ (1.0324 − 0.19077 × log₁₀(waist − neck) + 0.15456 × log₁₀(height)) − 450. That version is the centimetre form, and the distinction is not cosmetic.

Fed centimetres it agrees with the inches form on this page to about a tenth of a percentage point. Fed inches it returns 11.1% for the same man who measures 17.5%, and for one of the worked female bodies above it returns a negative number. If you are checking this page against a formula you found elsewhere, check first which units it was written for.

Where these bands come from

From this page. The cut-offs of 6, 14, 18 and 25 percent for men, and 14, 21, 25 and 32 for women, are this tool's own bands and are not a clinical standard, a medical standard or an industry standard.

That is worth saying because the page used to credit the American Council on Exercise for the table without linking anything. Six ACE-owned addresses were tried on 21 August 2026. One carried any figures at all, and it carried two of the ten.

The commonly circulated essential-fat figures of 2 to 5% for men and 10 to 13% for women differ from this tool's floors of 6 and 14. That difference is reported rather than resolved: neither set could be read on a page that would open, so the page states its own and shows you the other.

How to measure so the number means something

• Waist, men: at the navel. This is the abdominal circumference the male equation was fitted on, and it is usually not the narrowest part of the torso.
• Waist, women: at the narrowest part of the abdomen. The two equations genuinely want different landmarks here.
• Neck: just below the larynx, tape sloping slightly downward at the front.
• Hip, women only: at the widest point of the hips and buttocks.

The cost of getting the waist landmark wrong is larger than it looks. On the 5 ft 10 in body above, a 34 in navel reading gives 17.5% and a 32 in narrowest reading gives 13.4%. That is 4.2 percentage points and two different band names from the same tape on the same body. An earlier version of this page gave the narrowest-point instruction to every reader, which was the female protocol handed to everyone.

Edge cases and what the tool refuses

• A waist not larger than the neck. No result. The logarithm has no real value there, so the page withholds rather than printing something.
• A result at exactly 2 or 75. The output is clamped to that window. A figure sitting on either edge is the edge of the window, not a reading.
• Switching between inches and centimetres. The two systems hold separate values, and the figures shipped as defaults are not exactly the same body. Enter converted values rather than rounded ones if you want them to agree.
• A figure that sits between two printed ranges. It cannot. The printed ranges are the code's own boundaries, stated as under-6, under-14, under-18 and under-25 rather than as 6 to 13 and 14 to 17, which used to leave a gap between the table and the label.
• Day-to-day variation. Hydration, meal timing and where the tape sits all move the waist reading, and the waist is the input the result is most sensitive to. A difference between two readings on the same tape is more informative than the absolute figure.

What was corrected on this page

Recorded rather than quietly fixed, because a reader who used the old page deserves to know which part was wrong.

• The measuring instructions told everyone to use the narrowest point of the torso. For men the equation wants the navel, and the difference is over four percentage points.
• The band table appeared in four places with three different versions of itself, including two different lowest-band floors and a top band called Obese in one place and Above average in another.
• The page described inputs the tool has never had: age, weight, and a breakdown of fat mass against lean mass, none of which it can produce from four tape readings. The page description also advertised a BMI-based estimate that does not exist here.
• The bands were credited to a named organisation with no link and no date. Most of the table could not be found on any page that organisation serves.

Nothing you type here leaves the page

Your height and tape measurements are processed by this page in your browser. They are not sent to a server, not stored after you close the tab, and not visible to anyone else. There is no account and nothing to sign up for.

Sources, and what could not be verified

• The equation. Hodgdon and Beckett, 1984, Naval Health Research Center. The coefficients used here match the published inches form digit for digit. The primary report itself could not be retrieved: the Defense Technical Information Center record answered 403 on 21 August 2026, so the equation is used on its coefficients rather than cited to a document nobody here has read.
• The two verified figures. An American Council on Exercise press-room page gives an average nonathlete range of 18 to 24 percent for men and 25 to 31 percent for women, and describes below 6 percent for men and below 14 percent for women as figures that may be considered dangerously low. Read and quoted on 2026-08-21.
• What did not verify. Five further ACE addresses returned an index, a nav shell or an unrelated article rather than a classification table. The Gallagher 2000 paper often cited for population ranges answered 404 at its journal and 403 at a mirror. Those figures are absent from this page rather than repeated from memory.

Scope. This page puts four tape measurements through a fitted equation and reports which of five named regions the answer falls in. It is not a medical tool, not a fitness assessment, and not a measurement of body composition. It reports nothing about health and takes no view on what any figure means. It has not been reviewed by a clinician and does not claim to have been.