Body Fat Percentage Methods Compared: Navy, Skinfold, RFM, and Boer

Gold-standard body fat measurement requires DEXA scanning, underwater weighing, or Bod Pod testing — methods that are accurate but expensive and inaccessible. The four field methods covered here — Navy, skinfold calipers, RFM, and Boer — use body measurements that anyone can take with a tape measure or calipers. Each has a different formula, different assumptions, and a different accuracy profile. Our body fat calculator runs all the major methods so you can compare results side by side.

Method 1: U.S. Navy Circumference Method

Developed by the Naval Health Research Center in 1984 (Hodgdon and Beckett), the Navy method uses neck and waist circumference (and hip for women). It was designed for fitness screening at scale — no equipment beyond a tape measure.

Formulas

Men: % BF = 86.010 × log10(abdomen − neck) − 70.041 × log10(height) + 36.76

Women: % BF = 163.205 × log10(waist + hip − neck) − 97.684 × log10(height) − 78.387

(All measurements in centimeters; height in centimeters.)

Accuracy and Limitations

Independent validations report a standard error of estimate (SEE) of roughly ±3-4 percentage points against DEXA or hydrostatic weighing in general populations. The method performs best for people with typical fat distribution patterns. It tends to underestimate body fat in people who carry fat primarily in non-measured areas, and can be sensitive to where exactly on the abdomen the measurement is taken. Measurement site standardization is essential for reproducibility.

Best For

Quick screening. The U.S. military uses variants of this method as a fitness standard. It is the most reproducible tape-measure method when protocol is followed carefully.

Method 2: Skinfold Calipers (Durnin-Womersley / Jackson-Pollock)

Skinfold methods measure subcutaneous fat at specific anatomical sites using spring-loaded calipers, then plug those measurements into a density equation that converts to body fat percentage. The most widely used protocols are:

• Durnin-Womersley (1974): 4-site protocol (biceps, triceps, subscapular, suprailiac) — validated across ages and both sexes
• Jackson-Pollock 3-site: Different sites for men (chest, thigh, abdominal) and women (triceps, thigh, suprailiac) — faster but slightly less comprehensive
• Jackson-Pollock 7-site: Most sites, highest accuracy among caliper methods

How Skinfold → Body Fat

Skinfold thickness at multiple sites is summed, then converted to body density via an age-specific regression equation. Body fat percentage is then calculated from density using the Siri equation: % BF = (4.95 / density − 4.50) × 100.

Accuracy and Limitations

In trained hands, the Jackson-Pollock 7-site protocol has SEE of approximately ±3.5% against hydrostatic weighing. The critical limitation is tester skill — inter-tester variability can be 5% or more when the same person is measured by two different testers. Skinfold methods also assume a specific subcutaneous-to-total-fat ratio that may not hold for all ethnicities and age groups. Calipers systematically underestimate body fat in obese individuals where skinfold thickness exceeds caliper range.

Best For

Research settings and trained personal trainers who measure the same clients repeatedly with the same protocol. Our skinfold calculator supports the Durnin-Womersley and Jackson-Pollock protocols.

Method 3: Relative Fat Mass (RFM)

RFM was proposed by Woolcott and Bergman (Scientific Reports, 2018) as a simple improvement on BMI for predicting whole-body fat percentage. It requires only height and waist circumference.

Formulas

Men: RFM = 64 − (20 × height / waist circumference)

Women: RFM = 76 − (20 × height / waist circumference)

(Both in the same units — meters or centimeters, as long as consistent.)

Accuracy and Limitations

Woolcott and Bergman's validation against DXA across a large NHANES sample found RFM outperformed BMI as a fat mass estimator and had similar or better accuracy to some multi-site anthropometric methods. SEE was approximately ±7-8 percentage points in the full population — wider than skinfold methods in trained hands, but comparable to casual tape-measure approaches. RFM does not differentiate subcutaneous from visceral fat, and its waist-to-height ratio basis makes it sensitive to the same truncal fat distribution assumptions as waist-based methods.

Best For

Quick population screening where simplicity is paramount, or as a complement to BMI. Our RFM calculator makes this a two-measurement estimate.

Method 4: Boer Formula (Lean Body Mass)

The Boer equation (Metabolism, 1984) estimates lean body mass (LBM) directly from height and weight, and body fat is derived as: % BF = (total mass − LBM) / total mass × 100.

Formulas

Men: LBM = (0.407 × weight in kg) + (0.267 × height in cm) − 19.2

Women: LBM = (0.252 × weight in kg) + (0.473 × height in cm) − 48.3

Accuracy and Limitations

Because Boer uses only weight and height — the same inputs as BMI — it cannot capture fat distribution differences between individuals of the same weight and height. It is less accurate than methods that measure circumferences or skinfolds, but is sometimes useful as a baseline LBM estimate when you have accurate body fat from another method and want to derive LBM (or vice versa). It tends to perform best for average-build adults and degrades at extremes of leanness or obesity.

Best For

Estimating lean body mass when a body fat estimate from another method is unavailable, or as one data point in a multi-method comparison. Used internally in some calculators that derive LBM for protein target recommendations.

Head-to-Head Comparison

Method	Equipment	Sites	Typical SEE vs. DEXA	Skill Required
Navy	Tape measure	Neck, waist (+ hip for women)	±3-4%	Low
Skinfold (7-site)	Calipers	7 sites	±3.5% (trained)	High
RFM	Tape measure	Height, waist	±7-8%	Very low
Boer	Scale, tape measure	Weight, height	±8-10%	Very low

Which Method Should You Use?

For most people without professional guidance, the Navy method offers the best balance of accuracy and accessibility. Measure carefully, use the same protocol each time (same time of day, same landmark points), and track the trend rather than the absolute number.

If you work with a trainer who is experienced with calipers and consistent in their technique, skinfold testing gives you the most granular data over time.

RFM is a reasonable quick check and better than BMI alone at estimating fat percentage for population screening.

All four methods are available in our body fat calculator so you can see where different methods agree and diverge for your measurements.

Frequently Asked Questions

Which body fat method is most accurate without a DEXA scanner?

The Navy circumference method and Durnin-Womersley skinfold (when performed by a trained tester) are roughly equivalent in accuracy at ±3-4% against DEXA. The skinfold method can outperform Navy in skilled hands; it degrades significantly in unskilled hands.

Can I trust my gym's body fat reading from an InBody or BIA scale?

Bioelectrical impedance (BIA) scales and devices vary widely. High-quality clinical BIA can achieve ±3-4% accuracy. Consumer BIA scales are often less accurate and highly sensitive to hydration status. Measure at the same time of day under the same hydration and food conditions for any meaningful trend tracking.

Why do different methods give different results for the same person?

Each method measures a different proxy (circumferences, subcutaneous fat thickness, weight:height ratio) and uses a different regression equation to estimate total fat. Systematic biases between methods are normal. Choose one method and use it consistently.

Is the Navy formula accurate for muscular athletes?

Not reliably. A muscular person with a large neck and relatively small waist can score lower body fat than they actually have, because the formula interprets a large neck as lean mass. Athletes with unusual proportions are better served by skinfold testing or DEXA.

How often should I measure body fat?

Monthly is sufficient for most people tracking progress. More frequent measurement introduces noise from day-to-day hydration and measurement variability. Focus on the trend across 3-4 months rather than single readings.