Estimate body fat from circumferences. Metric or imperial.
. US Navy circumference method. Accuracy ±3–4 % vs DEXA — useful for trends, not clinical diagnosis.
The number on the bathroom scale tells you nothing about whether the kilos you carry are muscle, fat, water, or bone. Two adults of the same height and weight can have wildly different body compositions: an athlete with twelve percent body fat and a sedentary office worker with twenty-eight percent will weigh the same on the scale, but their cardiovascular risk, energy needs, and what their bodies look like are essentially different. Body composition matters for medical risk stratification (visceral fat is a stronger predictor of metabolic syndrome than weight), for training adjustments (cutting calories at twelve percent body fat is risky; cutting at thirty percent is overdue), and for tracking progress when the scale stalls but the mirror is improving. The gold standards — DEXA scans and hydrostatic weighing — give clinical accuracy but cost money and require a specialised facility. The US Navy circumference method, developed by Hodgdon and Beckett at the Naval Health Research Center in 1984, is the most widely-used field method: it uses three (men) or four (women) tape-measure circumferences plus height to estimate body fat with an error around three to four percent versus DEXA. That accuracy is enough to track personal trends and compare across people; this calculator implements it.
The US Navy method uses log-linear regression equations fit on a sample of 274 men and 214 women. For men: BF % = 495 / (1.0324 − 0.19077 × log₁₀(waist − neck) + 0.15456 × log₁₀(height)) − 450. For women: BF % = 495 / (1.29579 − 0.35004 × log₁₀(waist + hip − neck) + 0.22100 × log₁₀(height)) − 450. All measurements are in centimetres in the formula; the calculator handles inch-to-centimetre conversion internally so the user picks the units they have a tape measure for. The waist is measured at the navel for men and at the narrowest point above the navel for women. The neck is measured just below the larynx with the tape sloping slightly downward to the front. The hip (women only) is measured at the widest point of the buttocks with the feet together. The two formulas reflect different fat distribution patterns: men accumulate more visceral fat and women accumulate more gluteofemoral fat, so the women's formula adds the hip measurement to capture that difference. The log₁₀ terms make the model nonlinear in circumferences — a one-centimetre change in waist matters more at small waists than at large ones.
Four to five inputs depending on sex: a sex selector, a unit selector (cm or inches), height, neck circumference, waist circumference, and hip circumference (only used for the female formula; ignored when sex = male). The defaults represent a 178 cm man with average athletic build (waist 86, neck 38). The result panel shows the body fat percentage as the headline KPI and a category label. The category bands match the American Council on Exercise classifications: for men, Essential (< 6 %), Athletes (6–13 %), Fitness (14–17 %), Acceptable (18–24 %), Obese (≥ 25 %). For women, Essential (< 14 %), Athletes (14–20 %), Fitness (21–24 %), Acceptable (25–31 %), Obese (≥ 32 %). The female bands are higher because women have more essential fat (around 12 % minimum, versus 3 % for men) for reproductive function.
A 178 cm man with a 86 cm waist and 38 cm neck: waist − neck = 48 cm. log₁₀(48) ≈ 1.681; log₁₀(178) ≈ 2.250. Numerator = 495. Denominator = 1.0324 − 0.19077 × 1.681 + 0.15456 × 2.250 = 1.0324 − 0.3207 + 0.3478 = 1.0595. BF % = 495 / 1.0595 − 450 ≈ 467.2 − 450 = 17.2 %. Category: Acceptable. A 165 cm woman with waist 70, hip 95, neck 32: waist + hip − neck = 133. log₁₀(133) ≈ 2.124; log₁₀(165) ≈ 2.217. Denominator = 1.29579 − 0.35004 × 2.124 + 0.22100 × 2.217 = 1.29579 − 0.7434 + 0.4900 = 1.0424. BF % = 495 / 1.0424 − 450 ≈ 474.9 − 450 = 24.9 %. Category: Acceptable / Fitness boundary. Now lose 4 cm of waist (a typical six-month outcome of disciplined training and a small caloric deficit): waist − neck for the man becomes 44, log₁₀(44) ≈ 1.643. Denominator = 1.0324 − 0.3134 + 0.3478 = 1.0668. BF % = 495 / 1.0668 − 450 ≈ 14.0 %. The 4 cm change in waist drops body fat by about 3 percentage points — the classic "recomp" experience.
First, taking measurements at the wrong site. Waist at the navel and neck below the larynx are the canonical Navy sites; measuring waist at the natural waist (above the navel) gives lower numbers and breaks comparability with the regression. Second, measuring after meals, after training, or right after waking up. Stomach contents, post-workout swelling, and overnight dehydration all shift waist by 1–2 cm, which moves the BF estimate by a percentage point. Best practice is to measure first thing in the morning, post-bathroom, in the same room temperature, every time. Third, expecting clinical accuracy. The Navy formula was fit on a young, fit cohort and is least accurate at the extremes (very lean or very obese subjects). For both ends, DEXA is the right tool. Fourth, mixing units. Don't enter height in cm and circumferences in inches; the calculator handles unit conversion only when one selector covers everything. Fifth, optimising for a single number rather than a trend. A 1 % swing week-to-week is noise; a 2 % swing month-to-month is signal. Track over a quarter, not a week.
Body composition methods sit on a spectrum from cheap and rough to expensive and precise. Skinfold calipers (Jackson-Pollock 3- or 7-site equations) give an error of around 3–5 % in trained hands. Bioelectrical impedance (smart scales like Withings, Renpho) is convenient but error-prone, ranging from 4 to 8 % depending on hydration. DEXA scans give the clinical reference at around 1–2 % error and also distinguish visceral from subcutaneous fat. Hydrostatic weighing (underwater) is similarly accurate but logistically painful. BodPod (air displacement) is comparable to hydrostatic. The US Navy method's value is its zero cost and ubiquity — anyone with a tape measure can do it, the regression is documented, and the error is acceptable for personal tracking. For population-level health, the waist-to-height ratio (waist ÷ height; keep below 0.5) is an even simpler proxy that correlates well with metabolic risk without involving body fat at all. For athletes, lean body mass in kilograms (LBM = weight × (1 − BF/100)) is often more useful than the percentage, because it tracks the muscle that training is meant to build; a 2 kg increase in LBM at constant body weight implies a 2 kg drop in fat mass.