Research Review: FFMI Trends 1970–2026

Executive Summary: What Do FFMI Trends From 1970 to 2026 Actually Show?

Fat-Free Mass Index (FFMI) divides fat-free mass by height squared to express muscularity relative to stature. It is often discussed as if scientists have tracked FFMI continuously since the golden era of bodybuilding. They have not. The strongest historical evidence is much more fragmented.

In the 1970s and 1980s, sports scientists commonly reported body mass, skinfolds, body density, percent body fat and absolute fat-free mass. Those variables can inform historical context, but converting them into modern FFMI values after the fact would require compatible individual-level height and body-composition data—and even then, the measurement methods would differ from today's DXA, Bod Pod and multicomponent approaches.

What Is FFMI? Formula, Normalization and Why Height Matters

The basic equation is simple: fat-free mass in kilograms divided by height in meters squared. Fat-free mass includes skeletal muscle, bone, organs, water and other non-fat tissue, so FFMI is not a direct measurement of muscle tissue alone.

If body fat is estimated rather than directly measured, fat-free mass is commonly estimated as body weight × (1 − body-fat fraction). This means every error in body-fat assessment flows into FFMI. A hydration shift, depleted glycogen, contest-prep dehydration or device-specific bias can change the calculated result without representing a true change in skeletal muscle.

The 1970s: A Pre-FFMI Era of Body Composition and Doping History

The 1970s are important to a research review because they set the environment in which modern muscularity discussions developed. Anabolic-androgenic steroid use had already spread through elite athletics and bodybuilding in preceding decades, and anabolic steroids were placed on the IOC prohibited list in 1975. Yet FFMI was not the standard metric being tracked.

Researchers relied heavily on anthropometry, hydrostatic weighing, skinfold equations and absolute fat-free mass. These methods were useful for their time, but they create a major problem for anyone attempting to draw a smooth FFMI trend line from 1970 onward: a value derived from a 1970s skinfold equation should not automatically be treated as equivalent to a modern 4-component-model estimate.

The 1980s and Early 1990s: Better Body-Composition Context, Still No Standard FFMI Surveillance

During the 1980s and early 1990s, sports body-composition research expanded across disciplines. Scientists compared body density, skinfold thickness, percent body fat and fat-free mass among sport participants. This established an important principle that later FFMI work would confirm: athlete body composition differs substantially by sport and performance demands.

However, the same warning applies. These papers were not one harmonized FFMI database. They used varying populations, equipment, equations and sampling strategies. For historical trend analysis, they should be treated as contextual evidence rather than converted into a precise decade average.

1995: The Kouri Study and the Origin of the Famous FFMI 25 Threshold

The modern FFMI conversation changed with Kouri, Pope, Katz and Oliva's 1995 study of 157 male athletes. The group included 83 anabolic-androgenic steroid users and 74 reported nonusers. Researchers calculated FFMI and applied a height normalization to a 1.80 m reference.

The best-known result was that normalized FFMI values among reported nonusers extended up to a fairly clear boundary around 25 in that sample. The paper also examined historical pre-steroid-era Mr. America winners, which helped popularize the idea that FFMI near 25 might represent a rough natural upper boundary.

The 2000s: FFMI Spreads, but the Evidence Base Remains Heterogeneous

Through the 2000s, FFMI became more familiar in sports nutrition, physique coaching and online natural-bodybuilding discussions. At the same time, body-composition technology became more accessible. DXA, air-displacement plethysmography and bioimpedance systems were increasingly used in athletes, although there was still no unified FFMI surveillance program.

This period is also where internet culture began turning the 25 value into a binary “natty or not” rule. That interpretation was much stronger than the original evidence justified. A single threshold cannot account for sport, sex, position, frame, measurement device or statistical tails.

The 2010s: Sport-Specific Data Challenge the Universal 25 Ceiling

The 2010s produced some of the most important evidence for reinterpreting FFMI. A 2017 study of 235 NCAA Division I and II American football players measured body composition by DXA. Mean height-adjusted FFMI was 23.7 ± 2.1, but 26.4% of the players exceeded 25. The 97.5th percentile was 28.1, and six linemen even exceeded that level, with a maximum observed value of 31.7.

That result does not tell us the doping status of every athlete, and it should not be used to do so. What it demonstrates is that a one-size-fits-all 25 ceiling is a poor description of certain large, highly trained collision-sport populations.

In 2019, a study of 209 male collegiate athletes from 10 sports reported an overall adjusted FFMI of 22.8 ± 2.8. Football averaged 24.28 ± 2.39, while water polo averaged 20.68 ± 3.56. The estimated overall 97.5th percentile was 28.32, with sport-specific upper estimates differing again. This strengthened the argument for sport-specific reference values rather than universal cutoffs.

2020–2026: Larger Samples, Women, Natural Physique Athletes and Multicomponent Models

The 2020s expanded FFMI research in several important directions. A 2024 NCAA study included 1,961 athletes—596 men across 10 sports and 1,365 women across eight sports. Across sports, men averaged 21.5 ± 1.9 and women 17.9 ± 1.8. Men's throwers had the highest reported mean at 25.7, whereas men's volleyball athletes were lowest at 19.9. Among women, basketball athletes were highest at 18.9 and rowers lowest at 16.9.

A separate 2024 natural-physique study reported higher FFMI in professional than amateur natural physique athletes (23.83 ± 0.90 versus 22.80 ± 0.22). Rather than validating a hard 25 rule, the authors argued that muscularity boundaries in natural physique sport deserve reconsideration.

For the 2025 volume year, a four-component body-composition study of 225 university club-sport athletes found male FFMI values from 15.6 to 26.8 after removing a 30.0 outlier, and female values from 14.1 to 22.6. Powerlifting and rugby were among the sports with the highest FFMI. The importance of this study is not simply the top number—it is the use of a multicomponent model that reduces some assumptions inherent in simpler methods.

By August 2026, the literature increasingly treats FFMI as a contextual athlete-monitoring metric. The major advance is richer normative data and better measurement—not proof of a universal historical rise in human muscularity.

Is FFMI 25 Still a Useful Natural Limit?

It remains useful as a historical reference point, especially when discussing the 1995 paper. It is much less useful as a biological law. Later collegiate sport studies show that substantial numbers of athletes in some sports exceed it, and the largest values often cluster in positions where extreme body mass and fat-free mass are performance-relevant.

Using FFMI 25 as a doping accusation threshold creates two errors: false positives, where a drug-free person is labeled enhanced because their measured FFMI is high; and false negatives, where a drug-using person is assumed natural because their FFMI is below 25. Neither inference is scientifically defensible from FFMI alone.

Why FFMI Differs by Sex, Sport and Position

The 2024 NCAA dataset makes this especially clear. Average FFMI differed between men and women and varied substantially across sports. Within a sport, position can matter just as much. Football linemen, for example, face different performance demands from defensive backs. Throwers have different demands from distance runners.

Measurement Methods Can Create Apparent “Trends” That Are Not Biological Trends

FFMI is only as good as the fat-free mass estimate beneath it. Two-compartment models assume fixed characteristics of fat-free tissue. Athletes can violate those assumptions because hydration, bone mineral, glycogen and tissue composition differ from general-population reference values.

This is why a 1970s body-density estimate, a 1990s skinfold estimate, a 2017 DXA estimate and a 2025 four-component estimate should not be treated as four perfectly comparable points on one biological line chart.

So, Have Athlete FFMI Values Increased From 1970 to 2026?

The evidence does not support a precise universal answer. Elite sport has changed: resistance-training knowledge, nutrition, athlete selection, professionalization, specialization and performance-enhancing-drug practices have all evolved. It is plausible that some elite sport populations today carry more fat-free mass than comparable populations decades ago. But the published FFMI literature does not provide a harmonized, representative series that can quantify that change across all athletes.

The defensible trend statements are narrower:

1. FFMI measurement became more common after the mid-1990s.
2. Modern datasets show much larger sport-to-sport variation than a single cutoff implies.
3. Values above 25 occur in some high-level male athletic populations.
4. Women now have better, though still incomplete, sport-specific FFMI reference data.
5. Multicomponent methods are improving confidence in modern body-composition estimates.

How Athletes and Coaches Should Use FFMI Research in 2026

For practical use, compare an athlete against themselves first, then against a relevant sport, sex and position reference. Repeated measurements should ideally use the same device, protocol, time of day, hydration status and pre-test conditions. A change from 21 to 22 on the same validated protocol may be more useful than comparing a single 22 against an internet chart built from unknown methods.

FFMI also works best alongside performance data. A powerlifter may gain useful fat-free mass if strength and recovery improve, while a runner or climber may not benefit from maximizing FFMI. More is not automatically better.

Explore related FFMIPro tools and databases: FFMI Calculator, FFMI Database (10,000+ Athletes), FFMI Distribution Charts, Population Percentiles, Natural Bodybuilder FFMI Database, Age-Adjusted FFMI Norms, and FFMI for Different Sports.

Limitations of This FFMI Trends Review

• There is no annual or decade-by-decade standardized FFMI surveillance sample from 1970 onward.
• Different studies use different body-composition methods and height adjustments.
• Many athlete samples are convenience samples rather than representative national populations.
• Natural/enhanced status is often unavailable, self-reported or outside a study's purpose.
• Sport, sex, position, ethnicity, age, skeletal frame and training history can shift FFMI distributions.
• Contest-condition physique athletes may have unusual hydration and glycogen status.
• FFMI measures fat-free mass relative to height, not skeletal muscle alone.

Key Research Sources

Research review only. FFMI is not a medical test, anti-doping test, or proof of natural/enhanced status. Study estimates should be interpreted in the context of the population and body-composition method used.

Frequently Asked Questions About FFMI Trends