Health Guide
How Accurate Is the Navy Body Fat Method?
About ±3–4% against a DEXA scan, better for some bodies than others. How it compares to calipers, smart scales, and lab methods.
Last updated: August 27, 2026
The honest answer is that the Navy method is a decent estimate, not a measurement — and knowing the size and shape of its error makes it considerably more useful.
The headline number
Studies comparing the Navy circumference method against DEXA scanning — the practical reference standard for body composition — put it within roughly ±3–4 percentage points for most people.
Put concretely: an estimate of 18% most likely reflects a true value somewhere around 14–22%. That's wide enough that you shouldn't treat a single reading as precise, and narrow enough to be genuinely useful for tracking change.
This is why our Body Fat Calculator carries a moderate confidence label rather than high. The rating describes the method honestly rather than flattering it.
Where the error comes from
Three separate sources, worth separating because you control one of them completely:
- Measurement error. Tape placement, posture, and time of day. Entirely within your control — see how to measure yourself accurately, where a 2 cm waist difference shifts the result by about 1.5 points.
- Fat distribution. The formula assumes your fat is distributed in a fairly typical pattern. People who store proportionally more fat in their limbs or less around the waist get less accurate estimates.
- The original study population. The equation was developed on U.S. Navy personnel in the 1980s — fitter and younger on average than the general population. It travels reasonably well beyond that group, but not perfectly.
Who it's least accurate for
The method works best for adults with reasonably typical builds. Expect wider error if you are:
- Very lean or very heavy. Accuracy tends to degrade at both extremes of the range.
- Very muscular. A thick neck can skew the waist-minus-neck term in ways the equation didn't anticipate.
- Pregnant. Body composition formulas don't apply during pregnancy.
- A child or adolescent. These equations are for adults; growth requires age- and sex-specific approaches.
How it compares to the alternatives
| Method | Typical accuracy | Practicality |
|---|---|---|
| DEXA scan | Reference standard | Clinic visit, costly |
| Hydrostatic weighing | Very good | Specialist facility |
| Skinfold calipers | Good with a skilled tester | Highly technique-dependent |
| Navy circumference | About ±3–4% | Tape measure, two minutes |
| Bioelectrical scales | Highly variable | Convenient, easily skewed |
The Navy method's case isn't that it's the most accurate — it plainly isn't. It's that it's free, repeatable, and requires nothing but a tape measure, while landing much closer than a bathroom smart scale, whose readings shift with hydration, recent meals, and skin temperature.
Skinfold calipers can beat it, but only with a practised tester; self-administered caliper readings are often worse.
Why we don't show a category band
Most body fat calculators sort your result into "athlete", "fitness", "average", and similar. We don't, and it's a deliberate choice.
Those bands differ between organisations, and we haven't identified a citable source we're confident standing behind. Inventing plausible cutoffs, or copying another site's without knowing their provenance, would undercut the entire point of showing our sources. So we show the percentage and its uncertainty. If we find authoritative cutoffs, they'll ship with the citation attached and a version bump recorded in the change log.
Getting the most out of it
Given all the above, the method rewards a particular way of using it:
- Track the trend, not the reading. A series over months is far more informative than any single result.
- Keep conditions identical. Consistency turns much of the error into a constant offset that cancels out when comparing.
- Cross-check with a second measure. BMI and waist-to-height ratio come from numbers you already have — see body fat vs BMI.
- Don't chase decimals. The method cannot resolve a 0.3% change. Treat movements smaller than a couple of points as noise.
The bottom line
Expect ±3–4% against a DEXA scan, more if your build is unusual or your tape work is inconsistent. That makes it a poor tool for a one-off verdict and a good one for tracking direction over time — which is what most people actually need.
See the formula, its 1984 source, and its confidence rationale on the Body Fat Calculator.
Keep reading