Body Fat %
US Navy circumference method
The US Navy method estimates body fat from three tape measurements for men — height, neck and waist — and four for women, which add the hip. It does not weigh you, and the percentage it reports does not use your body weight at all — weight enters only afterwards, to split that percentage into fat mass and lean mass.
That is the method's defining property and its central limitation in one sentence. Two people with identical neck, waist and height measurements receive the identical body fat percentage even if one of them weighs 30 lb more. This estimate is about shape, not mass.
It is a field method for tracking your own trend, not a clinical measurement. Anything with real consequences attached — a medication dose, a surgical assessment — needs a scan.
Working the default measurements
Male, 5 ft 9 in (175.26 cm), 165 lb (74.84 kg), neck 15 in (38.10 cm), waist 33 in (83.82 cm). The tool converts everything to centimetres first:
waist - neck = 45.72 cm;log10(45.72) = 1.6601;log10(175.26) = 2.2437- Denominator:
1.0324 - 0.19077(1.6601) + 0.15456(2.2437) = 1.0324 - 0.3167 + 0.3468 = 1.0625 495 / 1.0625 - 450 = 15.9%— the Fitness band for men- Fat mass:
74.84 kg x 0.159 = 11.9 kg (26.2 lb). Lean mass:138.8 lb.
Now measure the waist one centimetre (0.4 in) higher on the abdomen and record 33.4 in instead. The result moves to 16.7%. A single centimetre of tape placement is worth almost a full percentage point — more than most people's monthly progress. This is the dominant error source in the method and it is entirely under your control.
The equations this tool actually runs
Two published forms of the Navy method circulate. The direct percentage form, in inches, is the one usually quoted:
Men: BF% = 86.010 x log10(waist - neck) - 70.041 x log10(height) + 36.76
The version implemented here estimates body density first and converts it with the Siri equation, working in centimetres:
- Men:
BF% = 495 / (1.0324 - 0.19077 x log10(waist - neck) + 0.15456 x log10(height)) - 450 - Women:
BF% = 495 / (1.29579 - 0.35004 x log10(waist + hip - neck) + 0.22100 x log10(height)) - 450
For the worked example above the two men's forms agree to a tenth of a point — 15.93 against 15.89 — so it makes no practical difference which one a given calculator uses. The women's equation adds hip circumference and a different constant set; the hip field only appears once you select Female.
The constants come from Hodgdon and Beckett's 1984 work at the Naval Health Research Center, which fitted circumference measurements against underwater weighing in 602 male naval personnel, with a separate female sample, reporting a multiple correlation of 0.90.
What a tape measure cannot know
Your weight is not in the percentage. The same tape numbers give the same result at 150 lb or 185 lb. If you add muscle to your legs and back while your neck and waist hold steady, this method reports no change whatsoever — which is either a useful property or a serious blind spot depending on what you are training for.
The stated accuracy is roughly 3 to 4 percentage points of standard error against hydrostatic weighing. That is the error of a fitted equation on the population it was fitted to; individuals outside that population — very lean, very heavy, or with unusual limb-to-torso proportions — do worse than the headline figure suggests.
It was developed on military personnel, a narrow slice of adult body types by age and conditioning.
The neck measurement is small and its logarithm carries weight. Half an inch of error on the neck moves the men's result by about one percentage point, in the opposite direction to the same error on the waist. Measure just below the larynx, sloping slightly downward at the front, and be consistent about it.
Hydration, meal timing, bloating and posture all move waist circumference by a centimetre or more within a single day, which is why a single reading means little and a four-week trend means something.
Reference methods disagree with each other too. DEXA, hydrostatic weighing and air-displacement plethysmography differ by a couple of points on the same person. There is no consumer-accessible ground truth here.
The category thresholds used here
| Category | Men | Women |
|---|---|---|
| Essential fat | under 6% | under 14% |
| Athletes | 6 - 13% | 14 - 20% |
| Fitness | 14 - 17% | 21 - 24% |
| Average | 18 - 24% | 25 - 31% |
| Obese | 25% and above | 32% and above |
The women's bands sit eight to ten points higher at every level. That is essential fat associated with hormone production and reproductive function, not a handicap, and it means a body fat percentage compared across sexes says nothing on its own.
It is also worth being honest about the provenance of these bands: they are fitness-industry conventions, not a WHO or CDC standard. There is no equivalent of the BMI cut points for body fat percentage carrying the same institutional backing, so treat a category boundary as a label rather than a threshold with evidence behind it.
Use it for the direction, not the level
Measure the same way every time: same hour of the morning, before eating, tape flat against skin without compressing the tissue, three readings averaged. Weekly or fortnightly is plenty; daily readings are mostly noise from the sources listed above.
What this method genuinely does well is answer a question the scale cannot: whether a six-pound loss over two months came off your waist or off everything at once. BMI cannot distinguish those either, which is why the two calculators are complements rather than alternatives. If you are running a deficit, Calorie (TDEE) sets the intake side; this page tells you whether it is working the way you intended.
If the number has to be right rather than merely consistent, get a DEXA scan. This is not a clinical assessment and nothing here is medical advice.
The four outputs
- Body Fat %, to one decimal.
- Category, from the table above.
- Fat Mass — the percentage applied to the weight you entered.
- Lean Mass — weight minus fat mass. This includes muscle, bone, organs, connective tissue and body water. It is not "muscle mass", and a large part of a short-term change in it is water.
Because only the last two use weight, an error in the weight you type shifts the masses but leaves the percentage untouched. If you want a metabolic estimate driven by lean mass rather than scale weight, take the lean figure from here into a Katch-McArdle calculation — the TDEE page explains how far that diverges from the weight-based equations.
Frequently asked questions
Why does my weight not change the percentage?
The Navy equations use only circumferences and height. Weight is applied afterwards to split the percentage into fat and lean mass, so two people with identical tape measurements get identical percentages at very different weights.
How precise is this really?
Roughly 3 to 4 percentage points of standard error against hydrostatic weighing, and worse for body types unlike the naval personnel it was fitted to. Use it to track direction over weeks, not to certify a number.
Where exactly do I put the tape?
Neck just below the larynx, sloping slightly downward at the front. Waist at the navel for men and at the narrowest point for women. Hip at the widest point of the buttocks. Stand naturally and do not pull the tape tight - one centimetre of waist error is worth almost a full percentage point.
Why do men and women use different equations?
Fat distribution and essential fat levels differ, so the women's equation adds hip circumference and uses a different constant set. The hip field appears only after you select Female.
How often should I measure?
Weekly or fortnightly. Hydration, meals and posture move waist circumference within a single day, so daily readings mostly record noise.
Is a 'Fitness' or 'Athlete' category a health target?
No. These bands are fitness-industry conventions rather than a clinical standard, and very low body fat carries its own risks. Treat the label as a description, not a goal.
Sources and further reading
Figures and definitions on this page are drawn from the following primary sources. If you find something out of date, tell us and we will correct it.