Explore fat-free mass index percentiles for men, women and athletes, place your own FFMI on the distribution, and understand what published reference data can — and cannot — tell you about muscularity.
See where a standard FFMI value falls within published male or female reference distributions instead of relying on unsourced labels such as “average” or “elite.”
Switch between a healthy-weight U.S. adult reference from NHANES III and percentile data reported for university club-sport athletes.
The distribution comparison uses FFMI = fat-free mass ÷ height². It does not silently mix in the separate height-normalized Kouri adjustment.
DEXA, BIA and estimated body-fat methods can produce different fat-free mass values, so the page explains why testing method matters when comparing percentiles.
Enter a standard FFMI value and choose the comparison group. The tool interpolates between the published percentile points shown in the selected dataset and marks your score on the curve.
BMI 18.5–30; ages 25–69 in the published standard table
The percentile estimator is interpolation between published reference points; it is not a new population model. Scores beyond the published athlete 10th–90th percentile range are reported as below P10 or above P90 rather than extrapolated as exact percentiles.
A percentile tells you the share of a reference distribution at or below a given FFMI value. P90, for example, is near the high end of that specific reference population.
Male and female FFMI distributions are displayed separately because published datasets show clearly different central values and upper tails.
Athletic populations can sit higher than general-population references, so the page includes a dedicated club-sport athlete comparison instead of treating one chart as universal.
FFMI depends on estimated fat-free mass. Different devices and equations can move the same person to a different apparent percentile.
The bars below summarize selected published percentile values. They make the shift between a general healthy-weight adult reference and a university athlete cohort easier to see.
Selected median and upper-decile values from the two reference sets.
The athlete cohort also shows a clear upward shift relative to the adult reference.
FFMI distribution charts are useful because a single fat-free mass index number has limited meaning without context. A standard FFMI of 20 kg/m² can be above the middle of one population, closer to typical in another, and still tell you nothing by itself about sport performance, health status, training history or whether someone is natural or enhanced. The purpose of this page is to make that context visible.
The primary adult distribution here comes from a U.S. NHANES III analysis that created sex-specific FFMI percentile tables from bioelectrical impedance data. The authors found major sex differences, clear adolescent growth effects, and comparatively stable adult percentiles after restricting the sample to BMI 18.5–30. The published standard table was presented for ages 25–69. For athlete context, this page also uses percentile values reported in a multicomponent body-composition study of university club-sport athletes.
A distribution shows how values are spread across a group. In an FFMI chart, the horizontal direction usually represents percentile rank while the vertical direction represents FFMI in kg/m². The 50th percentile is the median: half the reference group falls below that value and half above it. The 90th percentile is higher than about 90% of the reference distribution, while the 10th percentile is lower than about 90% of it.
The NHANES III standard table reports about 19.16 kg/m² at the 50th percentile for men in the restricted adult reference.
The corresponding female 50th percentile is about 15.96 kg/m², illustrating why sex-specific interpretation matters.
The university club-sport cohort reported median FFMI values of 21.0 for men and 17.6 for women.
Percentiles are not grades. A higher percentile is not automatically healthier, better or more athletic. In clinical nutrition, very low FFMI can matter because low fat-free mass may signal nutritional depletion. In sport, body size and lean mass needs vary dramatically by discipline. A lightweight endurance athlete and a heavyweight strength athlete should not chase the same FFMI target.
The following selected values come from the published standard FFMI table for men and women aged 25–69 with BMI 18.5–30 kg/m² across the included U.S. race/ethnicity groups. The original publication provides values for every percentile from 1 through 99; the streamlined table below highlights commonly used points.
| Percentile | Male FFMI | Female FFMI | Interpretive position |
|---|---|---|---|
| P1 | 15.278 | 13.544 | Very low tail of reference |
| P5 | 16.168 | 14.123 | Lower 5% |
| P10 | 16.807 | 14.444 | Lower decile |
| P25 | 17.891 | 15.110 | First quartile |
| P50 | 19.160 | 15.962 | Median |
| P75 | 20.378 | 16.868 | Third quartile |
| P90 | 21.428 | 17.663 | Upper decile |
| P95 | 22.013 | 18.162 | Upper 5% |
| P99 | 23.492 | 19.017 | Very high tail of reference |
It is not a universal worldwide standard. It is based on U.S. NHANES III data, BIA-derived fat-free mass, specific race/ethnicity groups, and a BMI-restricted adult sample for the standard table. Comparisons become less direct when your population, body-composition method or body-size range differs.
Athletes are an important separate reference because training participation and sport selection can shift the distribution of fat-free mass upward. A multicomponent study of university club-sport athletes reported FFMI percentiles from P10 through P90 for both men and women. This is not a universal “athlete standard,” but it gives a useful sport-participating comparison group.
| Percentile | Men athletes | Women athletes |
|---|---|---|
| P10 | 18.4 | 15.7 |
| P20 | 19.1 | 16.4 |
| P30 | 19.7 | 16.9 |
| P40 | 20.2 | 17.3 |
| P50 | 21.0 | 17.6 |
| P60 | 21.4 | 18.2 |
| P70 | 22.0 | 18.6 |
| P80 | 22.7 | 19.2 |
| P90 | 23.8 | 19.8 |
Notice the overlap. A high value in the general population may be closer to the middle of an athlete population. This is exactly why phrases such as “good FFMI” or “elite FFMI” are incomplete unless they name the comparison group. A percentile should answer compared with whom?
The percentile distributions on this page use standard FFMI. If you have weight, body-fat percentage and height, first estimate fat-free mass and then divide it by height squared.
Example: 80 kg at 15% body fat gives an estimated FFM of 68 kg. At 1.80 m tall, standard FFMI = 68 ÷ 1.80² ≈ 20.99 kg/m².
A separate formula from the 1995 Kouri athlete study adds a small height adjustment: normalized FFMI = FFMI + 6.3 × (1.80 − height in meters). That normalized score is often used in bodybuilding discussions. It should not be plugged directly into the NHANES or athlete percentile chart here because those references are based on standard FFMI.
If you need to calculate your score first, use the FFMI Calculator, then return here for the distribution comparison. For age-specific interpretation, see the Age-Adjusted FFMI Norms guide.
FFMI is only as stable as the fat-free mass estimate used to calculate it. DEXA estimates lean soft tissue and bone mineral content using X-ray attenuation. BIA estimates body composition from electrical impedance combined with prediction equations. Skinfold and circumference methods estimate body fat through regression equations. Consumer scales may use proprietary algorithms. These approaches can disagree even when body weight is identical.
For progress tracking, repeat the same device or estimation method rather than switching between DEXA, BIA and formulas.
Hydration, meals, recent exercise and glycogen can influence some measurements, especially impedance-based estimates.
A BIA-derived reference is most directly comparable with similarly derived values, though device equations can still differ.
Small one-off changes can reflect measurement variability rather than real tissue gain or loss.
The number 25 is widely repeated online, but it needs careful context. In 1995, Kouri and colleagues studied 157 male athletes, including 74 nonusers and 83 users of anabolic-androgenic steroids. The study used a height-normalized FFMI and reported that the nonuser group extended to a clear upper point around 25, while many steroid users exceeded 25 and some exceeded 30. The authors explicitly called the findings preliminary.
It came from a specific small male athlete sample and a normalized FFMI formula. Measurement error, body-fat estimation, height adjustment, genetics, sport selection and population differences all matter. A score over 25 cannot prove anabolic-drug use, and a score under 25 cannot prove natural status.
For that reason, the FFMI distribution charts on this page do not label 25 as a hard biological ceiling. Instead, they show published percentiles and explain which dataset each curve comes from. If your goal is to understand realistic progress rather than police “natural limits,” the Muscle Gain Projection and Client FFMI Assessment pages are better companions.
For an individual lifter, percentile rank is most useful as descriptive context. It can tell you whether your measured fat-free mass is relatively low, middle or high for a reference group. It can also help track a large change when the same testing method is used consistently. It is less useful as a target by itself.
Record your standard FFMI alongside body weight, body-fat method, waist measurement, key lifts and photos. A rising FFMI can indicate increasing fat-free mass, but one noisy body-fat reading can also move the number. Look for repeated trends.
Use sport-specific norms where available. A general athlete median still hides major differences between endurance, combat, aesthetic, team, throwing, strength and power sports.
Low FFMI is used in nutritional assessment and research, but clinical interpretation requires validated cutoffs, disease context and professional judgment. An online percentile chart is not a diagnosis.
BMI and FFMI both divide a mass measure by height squared, but they describe different components. BMI uses total body weight. FFMI uses fat-free mass. Two people with identical BMI can have different FFMIs if one carries more fat and the other more lean tissue. Conversely, a high FFMI does not automatically mean a low body-fat percentage; someone can have both substantial lean mass and substantial fat mass.
For a fuller profile, combine FFMI with body composition analysis, fat mass index where available, waist measures and performance data. That multi-metric approach is more informative than trying to turn one number into a complete physique rating.
Comparing height-normalized Kouri FFMI against a standard FFMI percentile table can shift your apparent position.
A general adult percentile and an athlete percentile answer different comparison questions.
A displayed percentile is not more precise than the body-composition estimate that produced the FFMI.
Another frequent mistake is assuming that a percentile is permanent. Changes in training, body weight, illness, aging and measurement conditions can all move FFMI. If you are monitoring a client, keep the test protocol consistent and consider using the Client FFMI Assessment workflow to document the measurement method and context.
Kudsk and colleagues created sex-specific FFMI percentile tables from NHANES III bioelectrical impedance data. The published standard table provides 1st–99th percentiles for men and women aged 25–69 with BMI 18.5–30 kg/m². Read the full-text article at NIH/PMC.
A multicomponent body-composition study reported FFMI percentiles from P10 to P90 for men and women in university club sports, providing a useful athletic comparison distribution. Read the full-text article at NIH/PMC.
Schutz, Kyle and Pichard published FFMI and FMI percentiles in 5,635 apparently healthy adults, reporting young-adult median FFMI values of 18.9 in males and 15.4 in females. View the PubMed record.
The 1995 athlete study introduced the frequently cited height-normalized FFMI discussion around 25 in nonusers. The paper described its screening conclusions as preliminary. View the PubMed record.
Calculate standard and normalized FFMI from height, weight and body-fat percentage.
Calculate FFMIExplore FFMI records and comparison data across athlete and physique examples.
Explore DatabaseUnderstand how age and reference population affect FFMI interpretation.
View Age NormsUse a structured FFMI workflow for coaching and client body-composition reviews.
Assess ClientCommon questions about FFMI percentiles, athlete distributions, measurement methods and the popular FFMI 25 claim.