The number 25 frequently appears in discussions of FFMI in bodybuilding and strength training. It is sometimes presented as a biological boundary: below it, a physique would be “natural”; above it, the use of performance-enhancing drugs would be certain.
That interpretation is far more categorical than the scientific study from which the number originates.
To understand what an FFMI of 25 actually means, we need to return to the paper by Kouri, Pope, Katz and Oliva, published in 1995 in the Clinical Journal of Sport Medicine.
First, what is FFMI?
FFMI relates fat-free mass to height:
FFMI = fat-free mass in kg / height in m²
Fat-free or lean mass is not synonymous with muscle mass. It includes all non-fat tissues and also depends on the method used to estimate body composition.
If these concepts are not yet clear, start with our FFMI and FMI guide and our explanation of lean body mass.
What did the 1995 study investigate?
The authors studied 157 male athletes, including 83 anabolic-androgenic steroid users and 74 non-users according to the groups described in the study.
They calculated FFMI and then applied a correction to normalize results to a reference height of 1.80 m.
In this sample, normalized FFMI values among the reported non-users reached an upper range close to 25.0, while several steroid users exceeded this level and some exceeded 30.
The authors also estimated FFMI for 20 Mr. America winners from the pre-steroid era and reported an average of approximately 25.4.
This is essentially where the famous “FFMI 25” comes from.
Why do people talk about “normalized” FFMI?
The authors did not use only raw FFMI. They applied a height-related correction to compare values around a stature of 1.80 m.
There is a small difference in convention that is worth documenting. The PubMed abstract of Kouri et al. (1995) reports a correction of 6.3 × (1.80 − height in metres). Later publications, however, present the equation attributed to Kouri with a coefficient of 6.1, which has become a commonly used variant.
Bethemesh uses 6.1:
normalized FFMI = FFMI + 6.1 × (1.80 − height in metres)
The difference between 6.1 and 6.3 is very small at typical heights. At 1.70 m, it represents 0.02 FFMI points; at 1.90 m, the absolute difference is also 0.02. This coexistence can nevertheless explain why two calculators display slightly different values from the same inputs.
Raw FFMI and normalized FFMI should therefore not be confused when discussing the historical threshold of 25, and the convention used should be stated.
The FFMI and FMI calculator displays raw FFMI separately from its normalized value using the 6.1 coefficient.
Does an FFMI above 25 prove steroid use?
No.
The study does not provide a test that can determine with certainty whether an individual uses steroids. The authors themselves described their findings as preliminary and discussed FFMI as a possible initial indicator in certain contexts.
Turning an observation from a specific sample into a universal rule — “FFMI > 25 = steroid use” — therefore goes beyond what the publication establishes.
An FFMI value is not proof of either use or non-use of a substance.
Why isn’t 25 a universal genetic limit?
Several reasons call for caution.
The sample was specific
The study involved male athletes. The same threshold cannot automatically be extended to every population, age group or to women.
FFMI depends on estimated fat-free mass
An error in body-fat percentage changes calculated fat-free mass and therefore FFMI. Different body-composition methods can produce different estimates.
Read our comparison of body-fat measurement methods to understand this source of uncertainty.
Fat-free mass is not just muscle
FFMI is not a “pure muscle mass index.” Water and other non-fat tissues contribute to fat-free mass.
An observed distribution is not an absolute physiological boundary
The fact that a studied group falls mostly or entirely below a given value does not prove that no individual could naturally exceed it.
The threshold can be useful for understanding the study; it should not be turned into a biological law.
Can FFMI be used to track progress?
Yes, as long as its limitations are understood.
For personal tracking, FFMI can provide a way to relate changes in estimated fat-free mass to height. In an adult whose height is stable, changes in FFMI mainly reflect changes in estimated fat-free mass.
It is therefore important to use a reasonably consistent body-composition method and avoid overinterpreting small variations.
The lean body mass calculator can compare several anthropometric estimates, while the FFMI and FMI calculator calculates the indices directly from body-fat percentage.
Why is the threshold still so popular?
The number 25 has several qualities that make it easy to spread: it is simple, memorable and appears to provide a clear answer to a complex question.
But body composition, genetics, training and substance use cannot reliably be reduced to a single threshold.
The most defensible use of the value is therefore historical and contextual: 25 describes a notable result from a specific study, not a boundary capable of classifying every individual with certainty.
Key takeaways
The FFMI value of 25 comes primarily from a 1995 study of male athletes. In that sample, the reported non-users did not exceed a normalized FFMI of about 25. The authors nevertheless considered their findings preliminary.
An FFMI above 25:
- does not by itself prove steroid use;
- is not a demonstrated universal genetic limit;
- depends on the prior estimate of body composition;
- should be compared using the same FFMI definition, particularly the height-normalization convention.
To put this number in the wider context of body-composition indices, read FFMI and FMI: understanding fat-free mass and fat mass indices. For the broader picture, return to Body composition: understanding what your weight doesn’t tell you.
Main source
- Kouri E.M., Pope H.G. Jr, Katz D.L., Oliva P., Fat-free mass index in users and nonusers of anabolic-androgenic steroids, Clinical Journal of Sport Medicine, 1995;5(4):223-228. PMID: 7496846.