Estimate your body fat percentage using the U.S. Navy method — from simple body measurements.
Your Results
Body fat estimates from circumference measurements have a margin of error of ±3–4%. For precise measurements, consider DEXA scan or hydrostatic weighing.
Body fat ranges differ between males and females. Women naturally carry more essential fat due to hormonal and reproductive functions.
| Category | Men | Women |
|---|---|---|
| Essential Fat | 2–5% | 10–13% |
| Athletes | 6–13% | 14–20% |
| Fitness | 14–17% | 21–24% |
| Average | 18–24% | 25–31% |
| Obese | 25%+ | 32%+ |
Accurate measurements are critical. Use a flexible tape measure and follow these steps:
This calculator uses two different approaches to estimate body fat percentage.
The formula never sees fat. It sees girth. For men it works from the difference between waist and neck, for women from waist plus hip minus neck, and in both cases it sets that against height on a logarithmic scale. Hodgdon and Beckett built it at the Naval Health Research Center in 1984, calibrated against hydrostatic weighing, so the Navy could screen thousands of sailors with a tape measure instead of a water tank.
That design explains the strengths and the blind spots together. Abdominal girth tracks abdominal fat closely across a population, and the neck term subtracts a frame proxy that barely moves as body fat changes. What a tape cannot report is what sits underneath it. A 90 cm waist on a heavily trained abdomen and a 90 cm waist on an untrained one return the same percentage.
The method also carries the population it grew out of: young, screened, largely male military personnel. Applied far outside that group — very lean athletes, adults in their sixties, women who carry most of their fat on hips and thighs — the estimate drifts, and it drifts silently. Nothing in the output tells you which way.
Measurement error dominates this calculator. Because the formula runs on a logarithm, the sensitivity is easy to state. For a man whose waist minus neck comes to about 52 cm, one centimetre of waist error moves the result by roughly 0.7 percentage points. For a woman whose waist plus hip minus neck lands near 141 cm, one centimetre is worth about 0.5 points.
Now count the ways a centimetre goes missing. Abdominal girth changes through the day with food and fluid in the gut. Holding air in versus breathing out normally is easily 2 cm. Tape tension that dents the skin against tape that merely meets it costs another. Measuring over a t-shirt costs one more. Stack three of those and the reading moves two full points while your body has not changed at all.
Every method estimates body fat from something else it can measure directly, and each one answers to a different reference standard. That is why they disagree, and why comparing across methods tells you almost nothing.
| Method | What it physically measures | Where it fits |
|---|---|---|
| DEXA | X-ray attenuation, split into fat, lean and bone by body region | Research and clinics; the usual reference standard |
| Hydrostatic weighing | Body density, from weighing you underwater | The standard Hodgdon and Beckett calibrated the Navy formula against |
| Skinfold calipers | Subcutaneous fat thickness at 3 or 7 sites (Jackson and Pollock, 1978 and 1980) | Gyms and labs; the result depends heavily on the tester |
| Navy circumference | Waist, neck, hip and height (Hodgdon and Beckett, 1984) | Free, repeatable at home, and what this page runs |
| Bioimpedance scale | Electrical resistance, which it converts to total body water | Daily use at home; the noisiest of the group |
The scale in your bathroom does not measure fat either. It passes a small current through you, measures the resistance, estimates total body water from that, and works back to lean mass using the assumption that lean tissue is about 73% water. Every link in that chain moves with hydration. Weigh in after a salty meal, after training, after coffee, or at a different point in the menstrual cycle, and the water estimate shifts, and the displayed body fat shifts with it.
Foot-to-foot scales carry a second problem. The current takes the short path, up one leg and down the other, so the trunk contributes far less to the reading than it does to your body. A scale can therefore sit several points away from DEXA and stay wrong in the same direction for months, which makes its trend more usable than its absolute number. A scale and a tape read on the same morning can differ by five points with neither one broken.
The categories in the table above come from the American Council on Exercise. They are population descriptions rather than medical thresholds, and they carry one omission worth knowing: no adjustment for age.
Body fat at a fixed BMI rises as the years pass, because muscle mass falls. This calculator's second method puts a number on it. The Deurenberg equation (1991) adds 0.23 percentage points of estimated body fat for every year of age at the same BMI and sex, so between 25 and 55 it adds 6.9 points. A 55-year-old man at 22% and a 25-year-old man at 22% land in the same row of the table above and in very different percentiles of their own age groups. ACSM publishes age-banded percentile norms in its Guidelines for Exercise Testing and Prescription if you want to place yourself precisely.
Athletic ranges are seasonal. Physique competitors touch the bottom of the athlete band for a few weeks around a contest and come back up afterwards, while endurance and team-sport athletes usually live in the fitness band year-round. Dropping under the floor in the first row of that table is not a state anyone holds, and in women it goes together with menstrual disruption and thinner bones.
BMI takes your weight and divides it by height squared. There is no term in it for what that weight consists of, which is why a 95 kg rugby player and a 95 kg office worker of the same height receive the same number and nothing useful from it. Run the BMI calculator beside this one and the gap between the two results is itself informative.
Body fat percentage also explains the frustrating middle of a diet. Someone losing fat and adding muscle at the same time can hold a flat weight for six weeks while body fat falls three points. The scale calls that a plateau. It is not one, and body recomposition covers how to train and eat through it. Water retention produces the same illusion on a shorter timescale, which why the scale lies takes apart.
The opposite pattern matters just as much: a normal BMI carried with little muscle and a high fat percentage. It is invisible to BMI by construction and obvious the moment you put a tape round your waist. Where the fat sits is a separate question again, and the waist-to-hip ratio calculator answers that one.
One measurement is a data point with error bars wider than most people assume. Four measurements taken the same way are a slope, and a slope is the thing that tells you whether the diet works.
The reason this works is that the errors repeat. Your tape tension, your landmarks and your morning routine bias every reading in roughly the same direction by roughly the same amount, so they largely cancel when you compare March against June. Change the routine and you lose that. The raw centimetres carry their own signal too: if waist has dropped 3 cm while neck and weight held steady, the fat came off the abdomen, and the tape told you that before any formula did.
Expect the numbers to move slowly. An 80 kg man at 20% body fat who loses a kilogram of pure fat over a month reads 19.0% afterwards. A single point of change, against a method whose error bars are several points wide, is exactly why one reading proves nothing and six monthly readings in a row prove a great deal.