A BMI body fat calculator does not measure body fat. It plugs your body mass index (BMI), age, and sex into a prediction equation built from studies that measured body fat directly. The most widely cited adult equation comes from Deurenberg et al. (1991), and its typical error is about 4 percentage points of body fat (Deurenberg et al., 1991). That makes it a rough screening estimate, useful for a first look or for tracking a trend, and unreliable as a precise number.
This guide shows the formula, works through an example by hand, explains where the estimate breaks down, and compares it with other ways of measuring body composition.
Key Takeaways
- BMI-based body fat estimates are predictions, not measurements. The standard adult equation has a typical error of about 4 percentage points (Deurenberg et al., 1991).
- BMI cannot tell fat from muscle. In one large U.S. sample, a BMI of 30 or higher missed more than half of the people whose body fat percentage met an obesity definition (Romero-Corral et al., 2008).
- Use one method, under consistent conditions, if you want to track change over time.
- Check any result against waist circumference, health history, and advice from a healthcare professional.
What BMI Measures and What It Leaves Out
BMI is body weight relative to height. The Centers for Disease Control and Prevention (CDC) uses it as a screening measure, not a diagnosis, and notes that it is not diagnostic of an individual's body fatness or health (CDC, n.d.-a; CDC, n.d.-b). For adults 20 and older, the CDC categories apply regardless of age, sex, or race (CDC, n.d.-a).
| Adult BMI category | BMI (kg/m²) |
|---|---|
| Underweight | Below 18.5 |
| Healthy weight | 18.5 to less than 25 |
| Overweight | 25 to less than 30 |
| Obesity | 30 or higher |
Source: CDC (n.d.-a).
Why BMI Is Not a Body Fat Measurement
BMI uses only height and weight, so it treats muscle, bone, water, and fat the same way. That limits how well it identifies excess body fat in individuals. In a study of U.S. adults that used bioelectrical impedance to estimate body fat, a BMI of 30 or higher was very specific (95% in men, 99% in women) but poorly sensitive (36% in men, 49% in women) for body-fat-defined obesity. The authors concluded that this cutoff misses more than half of people with excess fat, and that BMI was least accurate in the intermediate range of 25 to 29.9, in men, and in older adults (Romero-Corral et al., 2008).
The study used an estimated body fat reference rather than a laboratory measurement, so treat the exact percentages with some caution. The direction of the finding is still clear: a "normal" or "overweight" BMI does not rule out a high body fat percentage, and a high BMI does not always mean high body fat.
How to Calculate BMI
Convert height and weight to one system, then use whichever version matches your units.
- U.S. units: BMI = 703 × weight in pounds ÷ (height in inches)²
- Metric units: BMI = weight in kilograms ÷ (height in meters)²
The BMI value is the same for men and women. Sex enters the calculation only in the body fat equation below. For children and teens, the CDC uses a separate BMI-for-age method, and the adult categories above do not apply (CDC, n.d.-a).
How a BMI Body Fat Calculator Works
Deurenberg et al. (1991) analyzed 1,229 people aged 7 to 83 whose body fat had been measured by densitometry and anthropometry, and published age- and sex-specific prediction equations. For adults, the equation is:
Body fat % = 1.20 × BMI + 0.23 × age − 10.8 × sex − 5.4, with sex scored 1 for males and 0 for females (Deurenberg et al., 1991).
Put another way, the calculator subtracts 16.2 for men and 5.4 for women:
| Adult | Equation |
|---|---|
| Men | 1.20 × BMI + 0.23 × age − 16.2 |
| Women | 1.20 × BMI + 0.23 × age − 5.4 |
Source: Deurenberg et al. (1991). The paper gives a different equation for children aged 15 and younger.
What the Age and Sex Terms Do
Each additional year adds 0.23 percentage points to the estimate, so a 40-year-old and a 30-year-old with the same BMI differ by 2.3 points. The sex term lowers the estimate for men by 10.8 points relative to women with the same BMI and age (Deurenberg et al., 1991). The equation therefore builds in two patterns from its source data: body fat tends to run higher at a given BMI with age, and higher in women than in men.
How Accurate Is It?
The adult equation explained 79% of the variation in measured body fat, with a standard error of estimate of 4.1 percentage points. The authors reported that it slightly overestimated body fat in obese subjects, and that its prediction error was comparable to skinfold thickness or bioelectrical impedance estimates (Deurenberg et al., 1991). In practice, an estimate of 25% could reasonably correspond to a true value several points higher or lower.
Online calculators can also differ from one another because other published BMI-based equations use different coefficients, and tools may round differently. If you track changes, stick with one calculator.
Worked Example
Take an adult who is 5 feet 6 inches (66 inches) tall and weighs 152 pounds.
- BMI: 703 × 152 ÷ 66² = 703 × 152 ÷ 4,356 ≈ 24.5, which falls in the CDC healthy weight range (CDC, n.d.-a).
- If the person is a 30-year-old woman: 1.20 × 24.5 + 0.23 × 30 − 5.4 = 29.4 + 6.9 − 5.4 ≈ 30.9%.
- If the person is a 30-year-old man: 29.4 + 6.9 − 16.2 ≈ 20.1%.
Same height, weight, and age, yet the estimates differ by 10.8 points because of the sex term. With a typical error of about 4 points (Deurenberg et al., 1991), neither number tells you much about an individual. They are starting points for a conversation, not a result.
How to Read the Result
The American Council on Exercise (ACE) publishes a widely reproduced set of body fat categories. Different copies of the chart vary slightly at the edges, so check the version a given tool uses.
| 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% and above | 32% and above |
Source: ACE (n.d.), as reproduced by Healthline.
These categories describe typical ranges for groups. They are not individual targets or diagnostic cutoffs. ACE's own chart carries a note that it provides generic guidelines and that a physician can advise on ranges that fit you (ACE, n.d.). Essential fat is the minimum the body needs to function, so it is not a goal to aim for.
Where BMI-Based Estimates Go Wrong
- Muscle mass. BMI cannot distinguish lean mass from fat. In the large U.S. sample above, BMI correlated better with lean mass than with body fat percentage in men, and in the 25 to 29.9 range it failed to discriminate between the two in either sex (Romero-Corral et al., 2008). Muscular people and athletes can be misclassified as a result.
- Age. The diagnostic performance of BMI declined as age increased in the same study (Romero-Corral et al., 2008).
- Higher body fat. The Deurenberg equation slightly overestimated body fat in obese subjects (Deurenberg et al., 1991).
- Where fat is stored. The CDC also treats waist circumference as a screening tool, because it adds information about fat distribution that BMI lacks (CDC, n.d.-b). A calculator based on BMI alone cannot show this.
- Different populations. The equation comes from one study sample. It may fit people with a different body build or background less well.
- Children, teens, and pregnancy. Adult BMI categories do not apply to children and teens (CDC, n.d.-a). Adult equations are also not designed for pregnancy, so ask your clinician.
BMI Estimates Compared With Other Methods
| Method | What it uses | Main limitation |
|---|---|---|
| BMI-based estimate | Height, weight, age, sex | Prediction only; typical error about 4 points |
| Tape-measure (circumference) methods, including the Navy method | Height plus neck, waist (and hip for women) | Underestimated body fat versus DXA in military samples |
| Skinfold calipers | Fat thickness at set sites | Depends on measurer skill |
| Bioelectrical impedance (BIA) | Small electrical current | Sensitive to hydration, food, activity, and temperature |
| DXA scan | Low-dose X-rays | Needs a clinic or imaging facility |
Tape-Measure and Navy Methods
Military circumference equations estimate body fat from tape measurements. In a study of 1,407 Army recruits, these methods underestimated body fat relative to DXA at the start of eight weeks of basic training. They also underestimated the change in men and detected none in women, even though DXA showed body fat falling in both sexes (Foulis et al., 2023). A study of U.S. Marines found similar accuracy questions, particularly for women and for older adults (Potter et al., 2022). These studies used Army and Marine Corps equations on military populations, so they do not directly validate the Navy method for the general public. They do show that tape methods can miss real change over a short period.
Bioelectrical Impedance
An NIH consensus statement concluded that BIA can be a useful technique for healthy people, but that its values are affected by body position, hydration status, food and beverage intake, air and skin temperature, and recent physical activity. It also concluded that reliable BIA requires standardizing and controlling those variables (NIH Technology Assessment Conference Statement, 1996).
DXA
DXA is often used as the criterion method when researchers test other body fat estimates, including in the military studies above (Potter et al., 2022). It separates bone, lean, and fat tissue and can report them by body region. It is less convenient and more costly than a calculator or a home scale, and you need to book it through a clinic or facility.
Skinfolds and Laboratory Methods
Skinfold calipers measure the thickness of fat under the skin at specific sites, so results depend on the skill of the person measuring. Hydrostatic weighing and air displacement (Bod Pod) are laboratory methods that estimate body density, and they are generally harder to access.
How to Track Changes Responsibly
Use one method and repeat it under similar conditions. For BIA, that means similar timing, hydration, food intake, and recent activity, since those variables affect the reading (NIH Technology Assessment Conference Statement, 1996). With a BMI-based estimate, the only inputs that change are weight and age, so the "trend" mostly reflects your weight. That makes it a poor way to tell whether you are losing fat or gaining muscle. Pair it with a waist measurement, how your clothes fit, and your training log (CDC, n.d.-b).
When to Talk to a Healthcare Professional
Ask a clinician if your result is hard to square with your health, if you are considering a major diet change, or if you notice sudden unexplained changes in weight. They can interpret BMI alongside blood pressure, blood tests, waist circumference, and your history, none of which a calculator sees.
Conclusion
A BMI body fat calculator is a quick, free estimate built on one prediction equation with a typical error of about 4 percentage points (Deurenberg et al., 1991). It can give a rough starting point, but BMI misses a large share of people with excess body fat and cannot separate fat from muscle (Romero-Corral et al., 2008). Use it for a first look, choose one method for tracking, and rely on a healthcare professional or a direct measurement such as DXA when the number matters for a health decision.
FAQ
How accurate is a BMI body fat estimate?
The standard adult equation has a standard error of about 4.1 percentage points and slightly overestimates body fat in obese individuals (Deurenberg et al., 1991). BMI also misses many people with excess body fat (Romero-Corral et al., 2008). It is a rough screening tool.
Is it reliable for athletes or very muscular people?
Often not. BMI cannot separate muscle from fat, and in men it correlated better with lean mass than with body fat in one large sample (Romero-Corral et al., 2008). Athletes may get better information from methods that measure tissue directly, such as DXA.
What is a healthy body fat percentage?
There is no single number. ACE's chart lists "fitness" at 14–17% for men and 21–24% for women, and "average" at 18–24% and 25–31% (ACE, n.d.). These are group categories, not personal targets. A healthcare professional can tell you what range suits your health and goals.
Why does the calculator ask for my age and sex?
The equation adds 0.23 percentage points per year of age and subtracts 10.8 points for men relative to women (Deurenberg et al., 1991). These terms reflect patterns in the data the equation was built from.
How can I calculate it by hand?
Find BMI first (703 × pounds ÷ inches², or kg ÷ m²). Then use 1.20 × BMI + 0.23 × age − 16.2 for men or 1.20 × BMI + 0.23 × age − 5.4 for women (Deurenberg et al., 1991).
Is the Navy tape method better than a BMI-based estimate?
The evidence I could verify does not settle this. Military circumference equations underestimated body fat compared with DXA and missed changes during eight weeks of training in recruits (Foulis et al., 2023). Neither approach replaces a direct measurement, and the studies did not compare the two head to head.
References
- American Council on Exercise (ACE). (n.d.). Body fat percentage categories (essential fat, athletes, fitness, average, obese). Reproduced in Healthline, Ideal body fat percentage. https://www.healthline.com/health/exercise-fitness/ideal-body-fat-percentage
- Centers for Disease Control and Prevention (CDC). (n.d.-a). Adult BMI categories. https://www.cdc.gov/bmi/adult-calculator/bmi-categories.html
- Centers for Disease Control and Prevention (CDC). (n.d.-b). Assessing your weight. https://www.cdc.gov/healthyweight/assessing/index.html
- Deurenberg, P., Weststrate, J. A., & Seidell, J. C. (1991). Body mass index as a measure of body fatness: Age- and sex-specific prediction formulas. British Journal of Nutrition, 65(2), 105–114. https://doi.org/10.1079/BJN19910073
- Foulis, S. A., Friedl, K. E., Spiering, B. A., Walker, L. A., & Guerriere, K. I. (2023). Body composition changes during 8 weeks of military training are not accurately captured by circumference-based assessments. Frontiers in Physiology, 14, 1183836. https://doi.org/10.3389/fphys.2023.1183836
- NIH Technology Assessment Conference Statement. (1996). Bioelectrical impedance analysis in body composition measurement. Nutrition, 12(11–12), 749–762. https://pubmed.ncbi.nlm.nih.gov/8974099/
- Potter, A. W., Tharion, W. J., Holden, L. D., Pazmino, A., Looney, D. P., & Friedl, K. E. (2022). Circumference-based predictions of body fat revisited: Preliminary results from a US Marine Corps body composition survey. Frontiers in Physiology, 13, 868627. https://doi.org/10.3389/fphys.2022.868627
- Romero-Corral, A., Somers, V. K., Sierra-Johnson, J., Thomas, R. J., Collazo-Clavell, M. L., Korinek, J., Allison, T. G., & Lopez-Jimenez, F. (2008). Accuracy of body mass index in diagnosing obesity in the adult general population. International Journal of Obesity, 32(6), 959–966. https://doi.org/10.1038/ijo.2008.11
