Case 1
- Height
- 180cm
- Body fat
- 15%
- Weight
- 80kg
- Figure
- Height-adjusted (Kouri)
21.0kg/m²
Open with these valuesFFMI Calculator
21.0kg/m²
Result: 21.0 kg/m²FFMI is fat-free mass over height squared: weight × (1 − body fat %), divided by height in metres squared. Kouri's height adjustment adds 6.3 × (1.80 − height in m). Among the 74 non-users he measured, adjusted values reached 25.0 and stopped — a ceiling for that group, not a test for one person.
Worked examples
Case 2
- Height
- 175cm
- Body fat
- 10%
- Weight
- 75kg
- Figure
- Raw FFMI
22.0kg/m²
Open with these valuesCase 3
- Height
- 190cm
- Body fat
- 20%
- Weight
- 95kg
- Figure
- Height-adjusted (Kouri)
20.4kg/m²
Open with these valuesHow it's calculated
FFMI = kg × (1 − BF%) / m² + 6.3 × (1.80 − m)
- StepMultiply your weight by (1 − body fat percentage) to get fat-free mass.
- StepConvert your height to metres and square it.
- StepDivide fat-free mass by that square — this is raw FFMI in kg/m².
- ResultFor the adjusted figure add 6.3 × (1.80 − your height in metres).
What this number means
Fat-free mass index asks how much lean tissue a body carries for its height, which is the question BMI cannot answer because BMI counts fat and muscle alike. Strip the fat out of body weight, divide what is left by height in metres squared, and you get a figure in the same kg/m² units as BMI but describing only the lean part. Kouri and colleagues added one more step in 1995. Taller people carry more lean mass at the same build, so they normalised every value to a 1.80 m man by adding 6.3 × (1.80 − height in metres) — a small correction that lets people of different heights be compared. In their sample of 157 male athletes, the 74 who had never used anabolic steroids reached an adjusted FFMI of 25.0 and no higher, while many users went past it and some past 30. That single number has been carried a long way from what it can bear. It came from 74 men, measured with skinfold calipers, in one study the authors themselves called preliminary, and it describes where a group stopped rather than where any individual must. The larger practical problem is the input. Body fat percentage is the least reliable number most people have about themselves: a caliper reading a few points out moves FFMI by roughly a full point, which is the whole distance between two of the athlete groups listed below.
The normalized FFMI values of athletes who had not used steroids extended up to a well-defined limit of 25.0. Similarly, a sample of 20 Mr. America winners from the presteroid era (1939-1959), for whom we estimated the normalized FFMI, had a mean FFMI of 25.4.
25.0 is not a drug test
It marks where 74 non-users in one 1995 study stopped, not a limit any individual cannot pass. The authors called their findings preliminary and proposed FFMI only as an initial screening measure.
The body fat figure decides the answer
Everything here rests on a percentage that is hard to measure. Three points of error in body fat shifts FFMI by about one point — more than the gap between many of the groups in the table.
Why 6.3 and not 6.1
Almost every FFMI calculator uses 6.1. The Kouri abstract says 6.3, so that is what this page uses. Between 1.68 m and 1.93 m the two differ by at most 0.024.
Where the comparison figures come from
The athlete table is Table 3 of a 2026 scoping review in the Journal of Functional Morphology and Kinesiology. Those are raw FFMI values; at 1.80 m raw and adjusted are the same number.
Commonly misread
An FFMI above 25 means someone used steroids.
It means they are past where 74 men in one small 1995 study topped out. The paper proposed the figure as a screening prompt and called the findings preliminary.
FFMI is a more accurate version of BMI.
It answers a different question, and it needs a number BMI does not: body fat percentage. BMI takes two measurements you can trust; FFMI takes three, and the third is an estimate.
Raw and adjusted FFMI are interchangeable.
They differ by 6.3 × (1.80 − height). At 1.70 m the adjustment adds 0.63, which is enough to move a result across a comparison you might care about.
Reference table
| Athlete group | Raw FFMI (kg/m²) |
|---|---|
| Flat jockeys, England national | 17.5 |
| Judo, Portugal national | 20.7 |
| Soccer, Argentina first division | 21.5 |
| Latin dance, world top 6 | 22.1 |
| Water polo, 2003 World Championship top 4 | 22.3 |
| Australian football, Australia | 22.3 |
| American football, NFL safety | 24.0 |
| American football, NFL defensive back | 25.5 |
| Rugby forward, Australia national | 26.3 |
| Sumo, Japan national | 26.9 |
| American football, NFL running back | 27.6 |
| American football, NFL offensive lineman | 28.2 |
Questions
How do you calculate FFMI?
Multiply body weight by one minus your body fat fraction to get fat-free mass, then divide by height in metres squared. The height-adjusted version adds 6.3 × (1.80 − height in metres).
What is a good FFMI?
There is no threshold with evidence behind it, which is why this page shows published athlete figures instead of bands. Elite groups in a 2026 review run from 17.5 for flat jockeys to 28.2 for NFL offensive linemen.
Is an FFMI of 25 the natural limit?
That figure comes from 74 steroid-free athletes in a 1995 study whose adjusted FFMI went no higher. It describes where one small group stopped, and the authors themselves called the finding preliminary.
What is normalised FFMI?
It is FFMI adjusted to the height of a 1.80 m person, so people of different heights can be compared. Kouri's correction adds 6.3 × (1.80 − height in metres); at exactly 1.80 m nothing changes.
Does FFMI work for women?
The arithmetic is the same, but Kouri's sample was 157 men and the 25.0 ceiling describes them only. Published female athlete figures sit lower — around 16.6 for USA national artistic gymnasts.
How accurate is FFMI?
Only as accurate as the body fat percentage you feed it. Calipers, scales and DXA can disagree by several points, and each point of body fat moves FFMI by roughly a third of a point.
Sources and last check
Information, not medical advice.