BMI vs Body Fat Percentage — Which One Should You Trust?
BMI and body fat percentage are the two numbers people most often confuse — and the two most often used to judge each other. BMI is a ratio of weight to height. Body fat percentage is the share of your weight that is fat. They sound similar, they are often quoted together, and they can disagree wildly on the same person.
This guide settles which one to trust, and when. The short answer: neither alone. BMI is a fast population screening tool with a known blind spot (muscle); body fat percentage is a better individual measure with a known weakness (measurement error). Used together, with waist-to-height ratio as the tiebreaker, they give a picture that neither gives alone.
Use the calculator
Start with the numbers: try our BMI calculator to work out your own figures before applying anything on this page. The worked examples below use round numbers so you can follow the maths; the calculator uses your real weight, height, age and goal.
Related reading: BMI calculator, body fat calculator, Body Fat Percentage Explained, Waist-to-Height Ratio Explained, full guides section.
The formulas behind both numbers
BMI is weight (kg) divided by height (m) squared — BMI = kg ÷ m² — with WHO categories of underweight below 18.5, healthy 18.5–24.9, overweight 25–29.9 and obese at 30 or above (World Health Organization). Body fat percentage is fat mass divided by total mass, and the cheap way to estimate it is the U.S. Navy method (Hodgdon & Beckett, 1984) — for men, body fat % = 495 ÷ (1.0324 − 0.19077 × log10(waist − neck) + 0.15456 × log10(height)) − 450, and for women, 495 ÷ (1.29579 − 0.35004 × log10(waist + hip − neck) + 0.22100 × log10(height)) − 450, all measurements in centimetres, with a typical error of ±3–4 percentage points. The BMI-based estimate is the Deurenberg formula (Deurenberg et al., 1991): body fat % = 1.20 × BMI + 0.23 × age − 10.8 × sex (1 for men, 0 for women) − 5.4, with an error of about ±4–5 points. And the tiebreaker is waist-to-height ratio — waist divided by height, with 0.5 as the widely used threshold (Ashwell et al., 2012). The comparison is really a comparison of errors: BMI's error is systematic (it misreads muscle as fat, always in the same direction), while body fat percentage's error is random (measurement noise, in either direction). Systematic error is worse for decisions, because it cannot be averaged away.
The key points
BMI's blind spot is muscle. BMI cannot tell muscle from fat, so it misclassifies muscular people as overweight and lightly muscled people as healthy. A rugby player at 100 kg and 185 cm has a BMI of 29.2 — 'overweight' — with a body fat percentage of perhaps 15% and a waist-to-height ratio of 0.45. An older adult at 70 kg and 175 cm has a BMI of 22.9 — 'healthy' — with 30% body fat and a waist-to-height ratio of 0.55. BMI calls the first person unhealthy and the second healthy, and it is wrong about both. This is not a rare edge case: a 2008 study in the International Journal of Obesity (Romero-Corral et al.) found BMI misclassified a large share of people when checked against body fat percentage.
Body fat percentage's weakness is measurement error. Body fat percentage is the better concept — it is what you actually care about — but every cheap method of measuring it has a large error. The Navy method is ±3–4 points, calipers ±3–4 points in trained hands, BIA scales ±3–8 points. A Navy estimate of 24% could really be 20% or 28%, and those two readings sit in different categories. DEXA (±2–3 points) is better but expensive and not available weekly. So body fat percentage is the right number measured with the wrong tools, while BMI is the wrong number measured with perfect precision.
For populations, BMI wins. For individuals, body fat wins. This is the key distinction. For screening a thousand people, BMI is fast, cheap, standardised and good enough — its errors average out across a group. For one person making decisions about their own health, the errors do not average out, and body fat percentage plus waist-to-height ratio give a better picture. Public-health guidance uses BMI because it is practical at scale; your personal plan should use body fat percentage and waist measurements because they describe you, not a population.
Waist-to-height ratio is the tiebreaker. When BMI and body fat percentage disagree, the waist-to-height ratio usually settles it. A high BMI with a waist ratio below 0.5 and a low body fat percentage is muscle, not risk. A 'healthy' BMI with a waist ratio above 0.5 is central fat, and it carries risk that BMI misses. The ratio is also the best single predictor of cardiometabolic risk in the research summarised by Ashwell et al. (2012). If you only measure one thing beyond weight, measure your waist.
Age changes the comparison. BMI's blind spot gets worse with age, because muscle is lost and fat is gained while weight stays the same — a 70-year-old and a 30-year-old with the same BMI of 24 have very different body compositions, and the older one almost certainly has more fat. This is why the healthy body-fat ranges rise with age (Gallagher et al., 2000) while the BMI categories do not change. For older adults, body fat percentage and waist measurements are substantially more informative than BMI.
Use all three, and track the trend. The practical answer is not to choose one number but to use three: BMI for a quick check, body fat percentage (Navy method) for tracking change, and waist-to-height ratio for health risk. Record all three monthly, under the same conditions, and judge the trend over 8–12 weeks. If all three are moving in the right direction, the plan is working — whatever any single reading says on a given day.
At a glance
| Person | BMI | Body fat % (Navy) | Waist-to-height | Verdict |
|---|---|---|---|---|
| Rugby player, 100 kg, 185 cm | 29.2 (overweight) | ~15% | 0.45 | Healthy — BMI is wrong |
| Older adult, 70 kg, 175 cm | 22.9 (healthy) | ~30% | 0.55 | At risk — BMI misses it |
| Average man, 80 kg, 178 cm | 25.2 (overweight) | ~16% | 0.48 | Borderline — check the trend |
| Average woman, 65 kg, 165 cm | 23.9 (healthy) | ~26% | 0.44 | Healthy — all three agree |
| Muscular lifter, 95 kg, 180 cm | 29.3 (overweight) | ~14% | 0.47 | Healthy — BMI is wrong |
| Sedentary adult, 90 kg, 175 cm | 29.4 (overweight) | ~32% | 0.58 | At risk — all three agree |
A worked example: one man, both numbers, in full
Take a 40-year-old man, 95 kg, 180 cm, waist 88 cm, neck 42 cm, who lifts weights four times a week. His BMI is 95 ÷ 1.80² = 29.3, which the WHO classifies as overweight, one step from obese. His Navy body-fat estimate: waist − neck = 46 cm, log10(46) = 1.663, log10(180) = 2.255; density = 1.0324 − 0.19077 × 1.663 + 0.15456 × 2.255 = 1.0324 − 0.3173 + 0.3486 = 1.0637; body fat = 495 ÷ 1.0637 − 450 = 465.4 − 450 ≈ 15.4%. The Deurenberg BMI-based formula gives a very different answer: 1.20 × 29.3 + 0.23 × 40 − 10.8 × 1 − 5.4 = 35.2 + 9.2 − 10.8 − 5.4 = 28.2% — which is wrong by about 13 points, because the formula assumes an average amount of muscle for the BMI, and this man has far more than average. His waist-to-height ratio is 88 ÷ 180 = 0.49, just inside the healthy range. So the three numbers say: BMI 'overweight', Navy 'about 15% body fat', waist ratio 'healthy'. The correct conclusion is that he is lean and muscular, that BMI is misleading him, and that the Navy estimate and the waist ratio agree. If he had followed BMI alone, he might have dieted unnecessarily — losing muscle he worked hard to build — for a problem he did not have. That is the cost of trusting the wrong number.
How to use this guide
Calculate your BMI with the BMI calculator, your body fat percentage with the body-fat calculator (Navy method), and your waist-to-height ratio with a tape measure. Write all three down with the date. If they agree, the picture is clear. If they disagree, trust the combination: a high BMI with a low waist ratio and low body fat is muscle; a 'healthy' BMI with a high waist ratio is central fat that needs attention. Then set the plan from the numbers that describe you: if the waist ratio is above 0.5, a modest 300–500 kcal deficit with protein at 1.6 g/kg and resistance training is the standard approach, respecting the safety floors (1,200 kcal/day women, 1,500 kcal/day men). Re-measure all three monthly, and judge the trend over 8–12 weeks — the direction of the trend matters more than any single reading.
What this guide does NOT tell you
Neither BMI nor body fat percentage diagnoses anything: they are screening and tracking numbers, not medical tests, and they should be used alongside blood pressure, blood sugar, cholesterol and the other markers your clinician follows. The error figures here — Navy ±3–4 points, Deurenberg ±4–5 points, BIA ±3–8 points — are typical values from the published literature, and your individual error may be larger. The healthy ranges are population ranges (Gallagher et al., 2000; WHO; American Council on Exercise) that do not account for your ethnicity, body type, medical history or athletic background. This guide does not give individual diet plans or calorie targets; if you want to change your body composition, discuss targets with a clinician or registered dietitian. And if any of your numbers is very far outside the healthy range, or you have symptoms, seek medical advice rather than treating the numbers as a do-it-yourself project.
Frequently asked questions
Which is more accurate, BMI or body fat percentage?
Body fat percentage measures what you actually care about — the share of your weight that is fat — but every cheap method of measuring it has an error of ±3–4 points or more. BMI is measured with perfect precision but measures the wrong thing, because it cannot tell muscle from fat. For an individual, body fat percentage plus waist-to-height ratio is the better combination; for screening large groups, BMI is more practical.
Why does my BMI say overweight when I am fit?
Because BMI cannot see muscle. A muscular person at 95 kg and 180 cm has a BMI of 29.3 ('overweight') with a body fat percentage of perhaps 15% and a waist-to-height ratio of 0.49. BMI assumes an average amount of muscle for your height and weight, and if you have more than average, it misclassifies you. Check your waist-to-height ratio and body fat percentage — if they are healthy, BMI is misleading you.
Can you have a healthy BMI and still be unhealthy?
Yes — this is sometimes called normal-weight obesity. An older or sedentary person can have a BMI of 23 with 30% body fat and a waist-to-height ratio of 0.55, carrying a real cardiometabolic risk that BMI misses. Muscle is lost and fat is gained with age while weight stays the same, so BMI becomes less informative as you get older. Waist measurements catch what BMI misses.
What is the Deurenberg formula?
A published formula (Deurenberg et al., 1991) that estimates body fat percentage from BMI, age and sex: body fat % = 1.20 × BMI + 0.23 × age − 10.8 × sex (1 for men, 0 for women) − 5.4. It is a rough estimate with an error of about ±4–5 points, and it is unreliable for muscular people because it assumes average muscle for the BMI. The Navy tape-measure method is generally better for individuals.
Should I use BMI at all?
Yes, as a quick check and for comparison with population data — it is fast, standardised and useful at scale. But do not make personal decisions from BMI alone. Combine it with a body-fat estimate (Navy method) and a waist-to-height ratio, and judge the trend over months rather than any single reading.
What should I do if my numbers disagree?
Trust the combination. High BMI with a low waist ratio (below 0.5) and low body fat is muscle — do not diet. A 'healthy' BMI with a high waist ratio (above 0.5) is central fat — a modest deficit, more activity and resistance training are the standard approach, and a clinician can assess your overall risk. Re-measure monthly and watch the trend.
Estimates only, not medical advice. Minimum safe intakes: 1,200 kcal/day women, 1,500 kcal/day men — consult a healthcare professional before a very-low-calorie diet.