BMI for athletes: why muscle pushes the number up and what to measure instead

A drug-free powerlifter who stands 5'9" and weighs 190 lb at 12% body fat has a BMI of 28.1, which the chart calls Overweight. A rugby forward at 6'1" and 235 lb with 20% body fat reaches 31.0, which it calls Obese. Neither figure is a calculation error. Both are what BMI for athletes routinely looks like, because the formula sees a scale reading and a height and nothing else. The BMI calculator reports exactly those two values.

The useful question is how much of an athlete's BMI is muscle and how much is fat. One line of algebra answers it, and a 2025 study of 622 athletes shows how often the standard labels get it wrong.

Why BMI for athletes runs high

BMI is weight in kilograms divided by the square of height in metres:

Nothing in that fraction knows what the kilograms are made of, so muscle affects BMI exactly as much as fat does, kilogram for kilogram. The scale weighs both the same way.

The popular line that muscle weighs more than fat is loose wording: a kilogram of either weighs a kilogram. What differs is density. The two-compartment model behind underwater weighing (Siri, 1961) treats fat as 0.900 g/cm³ and fat-free mass as 1.100 g/cm³, so a kilogram of lean tissue takes up about 18% less room. An athlete can carry several extra kilograms without looking bigger, and every one of them lands in the numerator.

The effect is largest in sports that reward mass per unit of height. BMI for rugby players, throwers, heavyweight lifters and American football linemen runs high for this reason, and BMI for bodybuilders follows the same arithmetic. Endurance athletes sit at the other end. A marathon runner at 5'10" and 145 lb has a BMI of 20.8, and the formula describes that build well because it carries little muscle beyond the population average the cut-offs were drawn from.

Splitting BMI into a muscle part and a fat part

Every body weight is fat mass plus fat-free mass, so BMI splits exactly into two indices with the same units:

FFMI is the fat free mass index: fat-free mass (muscle, bone, organs and water) over height squared. FMI is the fat mass index, the same thing for fat. If f is body fat as a fraction, the split needs nothing more than the BMI you already have:

That identity is the muscle problem in one line. Two people with the same BMI can split it very differently, and the fat term is the part BMI is meant to screen for. The chart never sees the split, so it reads a high FFMI and a high FMI as the same thing.

How to calculate BMI for athletes, step by step

There is no separate formula for athletes; the arithmetic is identical. What changes is a second step, where a body fat estimate turns one number into two.

1. Calculate BMI as usual. The powerlifter is 86.2 kg and 1.753 m tall, which gives 28.1.
2. Estimate body fat with whatever method you have: a tape measure, skinfold calipers, bioelectrical impedance or DXA. Say it reads 12%.
3. Split the BMI with the formulas above.

Now put a sedentary man of the same height and weight next to him, at 32% body fat. His BMI is the same 28.1 with the same Overweight label, but he splits into an FFMI of 19.1 and an FMI of 9.0: more than two and a half times the fat per square metre of height. The scale cannot tell the lean 190 lb lifter from an overweight man of the same height. The split can.

The table runs the same steps for six builds. Body fat values are assumed for illustration; BMI comes from the calculator's own formula and rounding, and FFMI and FMI from the rounded BMI.

Read down the FMI column and the labels stop making sense. The Obese-labeled rugby forward carries less fat per square metre of height (6.2) than the office worker whose BMI reads Normal weight (7.6). The powerlifter and the sedentary man share a BMI to the decimal, yet the sedentary man's fat mass index is 2.6 times higher.

Athlete BMI chart: how much BMI can muscle explain?

Muscle cannot push BMI up without limit, and that limit is the most useful line an athlete BMI chart can draw. Kouri and colleagues (Clinical Journal of Sport Medicine, 1995) calculated FFMI for 157 male athletes, 83 who had used anabolic steroids and 74 who had not. Among the non-users, FFMI normalized to a height of 1.80 m topped out at about 25; many of the users went well past it. Treat 25 as the edge of one sample rather than a law of biology, but combine it with the split above and it caps how much BMI muscle can account for:

Two practical rules fall out of the table. First, a man whose BMI sits above the line for his measured body fat is carrying weight that muscle alone does not explain. A lifter who reports 5'9", 220 lb and 14% body fat has a BMI of 32.5 and would need an FFMI near 28, a level only the steroid users in Kouri's sample reached. For a drug-free athlete the likelier explanation is a body fat estimate that reads too low. Second, a BMI under the line is at least consistent with a lean, muscular build, which is the point where a second measurement is worth taking before any label is accepted.

The ceiling comes from men only. Kouri's study included no women, so there is no equivalent line for female athletes, and the 1.80 m normalization means the cap shifts slightly for shorter and taller men.

Tracking a training block: same BMI change, different bodies

BMI is often used to track change over time, and here the blind spot matters as much as in a single reading. Take a novice lifter at 5'10" (177.8 cm) and 75 kg with 15% body fat: BMI 23.7, FFMI 20.2, FMI 3.6. Over a training block he gains 4 kg and reaches 79 kg. His BMI becomes 25.0 and the label flips to Overweight, whichever way those 4 kg went.

The two outcomes are a successful bulk and a poor one, and BMI prints 25.0 for both. The FMI column separates them: a rise of 0.3 against a rise of 0.9. Note also that 79 kg at this height is a raw BMI of 24.99, about 30 grams short of a true 25, and the calculator's one-decimal display rounds it up to 25.0, so the label change rests on a rounding step.

What a 2025 DXA study found

A recent test of the standard cut-offs in athletes is Milanese and colleagues, "Revising BMI cut-off points for overweight and obesity in male athletes", published in Nutrients in 2025. They studied 622 male athletes in Italy with a mean age of 25.7 from a range of sports. Each athlete also had a DEXA scan. Body fat percentage from that scan was the reference, and every athlete was classified both ways:

By BMI, 27.5% of the group was overweight or obese; by measured fat, 3.9% was. In this sample BMI overestimates body fat status for roughly one athlete in four. The authors then searched for the BMI values that best matched the DXA classification and proposed a BMI cut off for athletes of 28.2 for overweight and 33.7 for obesity. For a man 1.80 m tall those lines fall at 91.4 kg and 109.2 kg, against 81.0 kg and 97.2 kg under the standard 25 and 30.

Those numbers are a proposal from one population: young adult men in Italy who were training when they were measured. No guideline has adopted them, they say nothing about women, and they describe a current athlete, not someone who trained ten years ago.

One formula, three sets of cut-offs

Cut-offs for the same formula already move in both directions. The WHO expert consultation on Asian populations (The Lancet, 2004) proposed lower action points because diabetes and heart disease risk rises at a lower BMI in many Asian groups. The athlete proposal goes higher because muscle raises BMI without adding that risk.

An Asian male rugby player sits under both adjustments at once, pulled down by one and up by the other. The number itself is identical in every row; only the reading changes, which is the strongest argument for adding a measurement that sees composition rather than arguing over where the lines go.

Why rescaling the formula does not help

BMI Prime divides BMI by 25, so the powerlifter's 28.1 becomes 1.12. The Ponderal Index cubes height instead of squaring it and gives him 16.0 kg/m³. Both are built from the same weight and height, so extra muscle raises them in the same proportion it raises BMI. Changing the exponent on height changes how an index treats tall and short people; it adds no way to see what the weight is made of.

FFMI vs BMI, and the measurements that see muscle

For an athlete, BMI vs body fat percentage is not an either-or choice. BMI is free and repeatable, and the body fat estimate is what makes the FFMI split possible. In an FFMI vs BMI comparison, FFMI wins on separating muscle from fat but loses on precision, because it is only as good as the body fat number it starts from. The main options are a tape measure, skinfold calipers, bioelectrical impedance and a DXA scan, the usual research reference; see body fat testing methods compared for how each one works and where it goes wrong.

The body fat calculator runs the U.S. Navy tape method from neck, waist and, for women, hip measurements. The lean body mass calculator estimates fat-free mass from height and weight with the Boer, James and Hume formulas, which, like BMI, assume an average build, so a measured body fat figure beats it for a muscular athlete.

The cheapest check needs no body fat estimate at all. NICE guideline NG246 (2025) recommends measuring waist-to-height ratio alongside BMI, with 0.4 to 0.49 as healthy, 0.5 to 0.59 as increased central adiposity and 0.6 or more as high. The guideline says those bands can be used for adults with a BMI under 35, including those with high muscle mass, and advises interpreting BMI with caution in that group. The rugby forward, at 185.4 cm with an 86 cm waist, scores 86 ÷ 185.4 = 0.46, inside the healthy band while his BMI reads Obese.

For a team, a workable record is BMI, waist-to-height ratio and one body fat method, used the same way at the same time of day on every visit.

Teenage athletes: the percentile scale has the same blind spot

Players under 20 are not judged on the adult bands. For ages 2 to 19 the CDC compares a BMI with other children of the same age in months and sex, and the BMI calculator switches to that percentile automatically. The percentile is still built from weight and height, so it inherits the muscle problem, and for a heavy teenager it can be harsher than the adult scale.

A 16-year-old prop at 5'10" and 200 lb has a BMI of 28.7. For a boy of exactly 16 that is above the 95th-percentile line of 27.5 (the calculator places it at the 96.5th percentile on the CDC 2000 curves), so the CDC category is obesity. Four years later the same 28.7 would read Overweight on the adult scale. At 185 lb his BMI is 26.5, the 93.1st percentile, and the category drops to overweight. Neither label says anything about how much of him is muscle, so the same waist and body fat checks apply, and a pediatrician or team doctor reads the growth trend over several seasons rather than one number.

Normal weight obesity: the same blind spot, reversed

The error runs the other way too. The office worker in the first table, 5'5" and 130 lb, has a BMI of 21.6 and a clean Normal weight label, yet at 35% body fat her FMI of 7.6 is higher than the rugby forward's. Normal weight obesity is the name for that pattern: a normal BMI with a high body fat percentage. BMI misses it for the same reason it misreads an athlete; the formula never sees composition.

Athletes are not immune. Someone who stops training can move toward it without the scale changing much, if the muscle they lose is replaced by fat. Their BMI stays put while their FFMI falls and their FMI climbs, which is why a retired athlete's unchanged BMI says little about how their body has changed.

A short BMI history: from the Quetelet index to 1998

Adolphe Quetelet, a Belgian astronomer and statistician, noted in his 1835 book on the "average man" that adult weight rises roughly with the square of height. The ratio was called the Quetelet index for more than a century, until Ancel Keys and colleagues compared several weight-for-height indices in the Journal of Chronic Diseases in 1972. They found this one tracked body fat best across large groups of men, named it body mass index, and presented it as a tool for comparing populations rather than judging individuals.

The cut-offs came later. In June 1998 the U.S. National Heart, Lung, and Blood Institute adopted 25 as the overweight line, down from 27.8 for men and 27.3 for women, bringing U.S. guidance into line with the WHO classification and moving about 29 million Americans into the overweight group overnight. None of those steps was designed around athletes, which is why the chart still treats a rugby forward like everyone else.

Run your own height and weight through the BMI calculator, then split the result with a body fat estimate using the formulas above. If your FFMI is high, your FMI is low and your waist is under half your height, the BMI label is describing your muscle.