Another Look at the Fat-Free Mass Index (FFMI): Part 2

Today I want to continue looking at the Fat Free Mass Index (FFMI) and the idea that exceeding a cutoff point of 25 kg/h^2 is indicative, suggestive or even proves anabolic steroid use.  In Part 1, I defined the concept and looked at the original paper that kicked all of this off.  I also looked at some of the counterarguments against the idea that have been made both online and in an article by Jan Todd in Iron Game History.

Today I want to continue with a re-examination of the topic by first making an addendum to the last part that is critically important.  Then I want to  look at an important physiological distinction and then examining a slightly different issue which is the upper limits of FFM that a human might carry to begin with.  I’ll finish by asking a cliffhanger question to set up the third and final part.

First the addendum to Part 1.

Testosterone was Synthesized in 1937

As was pointed out to me by my friend Broderick Chavez, testosterone was originally synthesized in 1937 and was commercially available by 1940.  According to the Wikipedia (a source that should always be taken with a grain of salt) entry for anabolic steroids:

Clinical trials on humans, involving either oral doses of methyltestosterone or injections of testosterone propionate, began as early as 1937.[201] Testosterone propionate is mentioned in a letter to the editor of Strength and Health magazine in 1938; this is the earliest known reference to an AAS in a U.S. weightlifting or bodybuilding magazine.[201]

Now reference 201 is a 1995 Scientific American article titled “The History of Synthetic Testosterone” (this link will take you to the PDF).  In reading the article, I find no mention of the Strength and Health letter.  Various other sources online make the same statement about the letter existing but I am unable to verify that this is actually true or not.

I would find it odd that someone made up this statement out of the blue but this is the Internet after all where you often find someone saying something and then it being repeated without verification by sources copying the original.  It would certainly not surprise me if the letter was written but, again, I have no way to verify this.  If it does, I would love to see it.

Even if this is not the case, by 1945 a book on testosterone had been written titled “The Male Hormone” which is thought to have made athletes aware of the potential benefits of testosterone.  It is rumored but unproven that steroids were used in World War II (1939-1945)  and the simple fact is that once the drugs were available, they were being used.  And not just clinically.

And this is relevant due to when many of the exceptions that I’ve mentioned came after that time point.  Ok, so the early strongmen were mostly in the late 19th and early 20th century, prior to the synthesis of anabolic steroids.  But I still take issue with any of the claimed measurements and their accuracy.  Even if some or all of those 9 athletes mentioned in Jan Todd’s article do cross the threshold, it’s still only a miniscule number of people.

At most they represent a literal handful of exceptions. At worst, reliance on any self-reported weights or heights is misleading because part of the showmanship of the day was exaggerating EVERYTHING about yourself.  Just as today, people wanted to see larger than life heroes.  And the best way to be larger than life is to lie about how large you are when nobody can verify your claims.

But going back to the Mr USA competitors in the Kouri paper the above dates are interesting.  The list of Mr. USA competitors starts in 1939 with 20 years of data being presented to 1959.  The 1939 competitor does not cross the FFMI threshold and many after that point do not either.  However, according to Jan Todd’s reanalysis, John Grimek surpasses the 25 FFMI cutoff in 1940 at 27.6.  By 1941, a mere one year later, he had hit an FFMI of 31.9 which is well above Arnold Schwarzenegger at his peak (I’ll come back to this).

Grimek was 5’8″ or 1.7m and, if I did the math right, would have given him an FFM of 80 kg (176 lbs) at an FFMI of 27.6 (to derive this divide FFMI by height in meters twice) in 1940.  To get to an FFMI of 32 would require him to have increased his FFM to 92 kg (203 lb).  Todd states that he weighed 221 on stage and this math works out.  221 lb at 8% bodyfat yields exaclty 203 lbs of LBM.

This means that he gained 12 kg/27 lbs of muscle in one year, despite being already highly trained (untrained men don’t win bodybuilding shows).  And this impossible to believe increase just happened to occur a few years after the synthesis of and commercial availability of synthetic testosterone.  With an alleged letter to the editor having been written to the dominant bodybuilding magazine as early as 1938.  Sure it’s all circumstantial that he was using.  But do you really think squats and milk got him those gains?

Other examples in her article such as Bill Pearl (who hit an FFMI of 29.8, nearly equal to Arnold Schwarzenegger at his peak, in 1956) also came after that time with Pearl, in a different history of steroids article, admitting to using in 1958.  Was he using prior to that?  I can’t say.  But the point of this is that steroids were involved in bodybuilding from the middle of the 20th century.

As the Scientific American article states:

According to anecdotal reports, West Coast bodybuilders began experimenting with testosterone preparations in the late 1940’s and early 1950’s.   News of the efficacy of these drugs apparently spread during the early 1960’s to other strength-intensive sports from the throwing events of track and field to footbal.

These dates are consistent with the 1945 book and, frankly, throws a real wrench in the claims that any bodybuilder after the mid 1940’s (and perhaps earlier) got across the FFMI cutoff clean.  Steroids were completely legal until the early 1980’s and there was no major moral stigma against drug use in sports until about the late 1960’s (and even this was only in the Olympics at that point).  Once available, they would have been put into use.  From at least the mid-1940’s and possibly a bit earlier, any claims that bodybuilders or athletes got past the FFMI cutoff without anabolics are questionable at best and should be dismissed at worst.

In that vein, Todd states that the steroid era of bodybuilding didn’t start until 1958 but I’m not sure this is accurate.  Dianabol, an oral derivative of testosterone, was approved by the FDA in 1958 which is probably where this date comes from. It’s commonly accepted that it was given to American Olympic Lifters by 1959 in an effort to combat the Russians (rumored to have been using since 1954).  But it’s clear that testosterone was both available and, most likely in use in bodybiulding, at least 10-15 years before that.

From the first day of sport, athletes have tried anything they thought would make them better and this tracks back to the first Greek athletes. Before testosterone was synthesized but after someone figured out that something about bull testicles made them into bulls, men were getting orchic injections (basically injections of testicular matter) into their nuts.  It didn’t do anything although the men swore it made them more virile.    There are STILL orchic supplements being sold.

According to yet another history of steroids that I closed the tab on and can’t find again, one late 19th century athlete tried this therapy to try to improve his performance.  Sure it was inert but it makes the point: when something that can improve performance becomes available, athletes will use it as soon as they can get their hands on it.  It is the nature of sport.

Once testosterone was synthesized, available and known about, it would have been used by any athlete or bodybuilder who could get it.  And it was by roughly 1940.  Does this prove that every lifter so far mentioned was on drugs?  No and I’m not going to copout and say it does based on what is admittedly circumstantial evidence.

It’s just damn interesting that a large majority of the exceptions mentioned so far by people arguing against the FFMI cutoff are referring to lifters who were at their peak both after testosterone was synthesized, was commercially available and was known about while also being completely legal to use.  Just things that make you go hmmm….

And with that out of the way, it’s time to delve back into the topic of the Fat Free Mass Index (FFMI), starting with a slightly pedantic but necessary physiological note.

All FFM is Not Muscle

I need to make a quick note that is important and will become importanter below and in Part 3.  And that is that all FFM (or Lean Body Mass, LBM) is not skeletal muscle.  Glycogen, water, bone, organs and minerals all count as FFM/LBM (anything that is not bodyfat is FFM).  Quite in fact, on average, of total FFM, skeletal muscle only makes up about 45-50% of the total with women’s values being a little bit less than this due to their having higher organ mass.

And this is important because when we are talking about natural limits to FFM I doubt many give much of a damn about glycogen, water and bone.  We care about muscle mass.   Even focusing just on glycogen and water, it’s fairly easy to skew the measurements of FFM/LBM with dehydration or carb-loading.  So with 2-3  days of no carbs, an average male might lose 2-3 kg (4-6ish pounds) of water and this will show up as FFM ‘loss’ lowering his FFMI. But is this what we are really talking about or care about?

By the same token if you carb-load someone you might put that same weight on them in a very short period of time.  One very silly book I read years ago actually claimed that it could “….put 5 lbs of muscle on someone in 3 days” and all it was was a carb-load protocol.  Yes it’s FFM but it’s not muscle.  And even if FFMI goes up because of it, is this really what we are talking about in the natty or not debate?  If I sodium and water load someone and make them gain a ton of water, their FFMI goes up.  Does this matter?

When creatine was first popular, you saw claims that it could put 3-5 lbs of “muscle” on someone in as little as a week.  Of course it was all just water.  This didn’t stop a supplement guy I met an an NSCA conferences ages ago from trying to tell me that “No, it’s actual contractile tissue.”  Sure, dude, sure.

Even some of the early blood flow restriction (BFR) work shows this enormous (and frankly impossible) “gain” in FFM.  I’m talking several percentage points in a week or two.   And it’s just pump growth and fluid accumulation (because it all goes away about a week later).

Hell, even the Haun paper I looked at in detail on the Training Volume and Hypertrophy series found that the “FFM” gains were mostly water when volumes got absurdly high.  Does it raise FFMI?  Sure.  Is this what we are discussing when we talk about FFMI and and natural limits?  I think not.

Now, this may seem like a pedantic note but it isn’t.  This is because when we are talking about the issue of the FFMI or natural training versus steroid use, what we are talking about is gaining actual muscle tissue, in the sense of actual contractile tissue.  We are not talking about gaining a bunch of glycogen and water ‘FFM’ because you creatine or carb loaded, got a shitpile of pump volume from BFR or stupid volumes or turned into a big fat boy.

And this goes to the comment brought up above by Khouri about the FFMI equation likely NOT holding for individuals carrying a lot of body fat.  Because when you get fat you hold a lot of non-contractile FFM such as water, glycogen, minerals and connective tissue.  And when the whole issue of natty or not comes up, this is not relevant to the discussion.

We are interested in the genetic limits of gaining MUSCLE MASS and
how steroids might or might not impact that as indicated by the FFMI.

And in that vein, let’s look at a slightly different but related question.

What Are the Upper Limits of FFM?

In keeping with the above, other research has shown some rather clear limits in how much FFM can be carried naturally (kind of) and this would give at least some indication of maximal FFMI values. Much of this work comes from, believe it or not, Sumo wrestlers wh due to their often enormous weights probably represent some of the upper limits possible.  In one study professional Sumo wrestlers were found to have a bodymass between 77.0 and 150 kg (170-330 lbs) with a FFM of  59.1 and 107.6kg (130-236 lbs) of FFM.

In another on 37 professional Sumo wrestlers, 6 were found to have over 100 kg (220 lbs) of FFM.  The average body fat percentage was 26.1% (which is hard to believe when you watch even a single video of Sumo wrestling).  The largest had an insane 121 kg LBM (266 lbs) at a weight of 181 kg (398 lbs).  His height was 186 cm (1.86 m) which gives him a normalized FFMI of 34.6 well well above the 25 limit. But he was also 33% bodyfat (60 kg or 133 lbs of fat) and that matters.  I suspect that BF% estimate is also low.

Because recall Khouri’s original comments: the FFMI may not be accurate at high BF% levels.  I’d go further and say it absolutely isn’t  due to the fact that a lot of the FFM gained when you get fat is not muscle tissue.  Rather it represents what is called ‘inessential FFM’ which means connective tissue, glycogen, water, minerals, etc. that all support the increased weight.  And this can make up to 25% of the total weight gained when people gain fat.  This depends on a host of factors including diet, initial BF%, training, etc.  But the point is that not all of the huge amount of FFM is actual muscle tissue when folks get really fat.

Which means that if you take the Sumo wrestlers and diet them down, they lose a LOT of their ‘extra’ non-muscle mass LBM, possibly up to 25% (this is the average amount of FFM gained when you get fat, whether you train or not and it tends go go away when you lean out).   If you simply subtract 25% from the above monster’s LBM (which is an estimation and nothing more), that’s a 30 kg FFM decrease which brings him to 91 kg of FFM, just over 200 lbs, at the same height).

And that done, his FFMI drops from 34.6 to 26.3, all of 1.3 units over the proposed cutoff.  Fine maybe he’s another 1-2 points above this because this is just a rough estimate, but does it really matter?  The point is that being big and fat skews the values up and, once again, that’s not what I think most of us are interested in here.  And no, I have exactly zero clue of whether or not there is or is not drug use in the sport of Sumo.

Even there, I remember when the idea of a near maximum of 110-120 kg (roughly 220 lbs)  muscle mass was thrown about (based on I believe one of the Sumo studies), a common argument was “No way, there are football players at 350 lbs who clearly have more muscle than that.”   Except that those big boys are invariably carrying a ton of bodyfat, probably 35%+, giving them at least 122 lbs of bodyfat.

Diet them down to very lean levels, some of the extra FFM comes off too and they end up at a lean 230 or whatever at maybe 10-12% bodyfat which gives them about 100kg/220 lbs of LBM or in that range.  I mean, at 300 lbs and 30% fat that’s 90 lbs of fat.  Lose most of the fat and you end up in ~210-230 lb range with ~200-210 lbs of FFM.  Just under 100 kg/220 lbs.  Weird.

Height and FFM

I mentioned above that there is research examining the relationship of FFM and height and this can be used to estimate some maximum FFMI values.  One of relevance is a 1989 paper done in on 4 elite female and 8 elite male bodybuilders.  It found that that elite males had a LBM to body height relationship of 0.45 kg/cm (women had 0.31 kg/cm but FFMI values or cutoffs for women have not been determined to my knowledge so I’ll only look at males).   The average BF% of the males was 8% and I’ll use that value for another quick back of the envelope calculation.

Let’s assume a male bodybuilder who is 5’8″ tall or 172 cm.  This gives him a LBM of 77.4 kg (170 lbs).  At 8% bodyfat his total weight is 84 kg (185 lbs) and his normalized FFMI is 26.6 which is 1.6 points above the cutoff point.  The bodybuilders are listed as having won state-level bodybuilding championships or placing top 5 in national championships (they are also listed as training 4+-1 hours PER DAY 6 days/week).  Were they natural?  I have no idea and any speculation would be only that and probably reflect my bias so I won’t.

I also don’t know offhand what types of bodyweights pro bodybuilders were competing at at that point which might give some indication if the sample above was natural or not.  But there is no doubt that they, albeit slightly, exceed the cutoff point.  They are also elite bodybuilders, by definition the top 1% of the 1% of competitors in the sport.  And they still only made it 1.6 points over the claimed cutoff.  Natural or not it kind of makes the point that the majority of humanity won’t even get close to the cutoff point much less above it naturally.  Many won’t even get there with drugs (a finding in the original FFMI paper that is often missed).

Hell, consider Arnold Schwarzenegger a bodybuilder who has admitted to steroid use.  His competition height and weight are listed as 1.88m and 107kg. Assuming 6% bodyfat, that gives him an LBM of 100kg (220 lbs BTW) and a normalized FFMi of 27.9 (at 4% this goes to 28.5).  This is 3 points higher than the cutoff yet somehow lower than Grimek in 1941.  Now we know that Arnold used anabolics which raises even more questions about this.

Was Grimek just a genetic god?  Was he using?  Did training and nutrition somehow devolve in the intervening 30 years?  Yeah, something is weird here and it’s probably several things including claimed bodyweights, the problem with visual estimates of BF% (what both Kouri et. al. and Jan and Terry Todd used), etc.  But if you want to argue that Grimek somehow surpassed one of the greatest bodybuilders in history clean (while doing it a year after testosterone was avaialble), well…I think you know what I think of that argument.

Regardless, do you really think that the grand majority of naturals are even getting close much less past the cutoff?  Me neither.

Other Exceptions to FFM Limits

Ok, what about other exceptions?  One that comes to mind is a recent case study on the world’s strongest RAW powerlifter.    Competing in the superheavyweight class he holds holds world records in the squat (477.5 kg/1050 lbs), deadlift (392.5 kg 863 lbs), and total (1105 kg/2431 lbs).  Standing 1.84 m (6 feet) tall at a bodyweight of 183.1 kg (402 lbs) his BF% was estimated by Ultrasound (rough at best) at 24.3% .

This gives him a calculated FFM of 138.6 kg (304 lbs), 17 kg higher than the sumo wrestler described above and 38 kg higher than Arnold.  It also gives him an FFMI of an astounding 41.  Even if we assumed that the Ultrasound was screwed and his BF% was as high as 30% (giving him an FFM of 128kg), his FFMI would still be 37.6.  But once again this is at a high BF% level which throws everything off.

Because, just like the Sumo wrestler, dieted down this would drop. If, just as I did with the Sumo we simply take 25% off his calculated 138.6 kg FFM (again, this is nothing more than a ROUGH estimate), his FFM drops to 104 kg and his FFMI drops to 30.7.  If he started with only 128 kg of FFM due to being fatter than the Ultrasound showed  and take 25% off of that to account for non-muscle FFM.

Now his FFM drops to 96 kg which gives him an FFMI of a much more realistic 28.3 (about 3 points above the cutoff, similar to Arnold at his peak). I’m not saying these calculations are completely accurate and they probably over-estimate the true drop in FFM, I am simply making the point that a high BF% makes the FFMI cutoff concept meaningless in this case.

Basically, FFMI gets skewed way higher when you carry a shitpile of body fat and the claim of it representing natural vs. steroid use was never about that situation.  And with some *very* rough assumptions being made, even in the biggest of the big, FFMI comes down to much more normal levels when you take all the fat off and lose the 25% of non-contractile FFM that tends to be gained.  It may still be above the cutoff point but it’s a much more reasonable amount over.  And these still represent elite genetic exceptions, the top 1% of the top 1%.

Now, I want to make three points before moving on.

  1. This is clearly Ray Williams, an absolute beast of a man and freak of a lifter (and I’ve heard a super nice guy).
  2. He clearly represents the absolutely elitest of the elite.  By definition, the strongest RAW powerlifter on the planet has to be an exception to the rest of humanity.  Few have squatted 1000 to begin with and even fewer raw.  He’s in a club with a handful of other people.  The majority of humanity is not him.
  3. He’s black.

Before everyone loses their shit, #3 is not meant as anything but a statement of fact.  And I think it matters since it’s possible/likely that baseline FFMI (or possibly the upper limits of FFMI) varies between ethnic groups.  I’m only aware of one study in this regard and it wasn’t on athletes, but it found that ethnically, FFMI was highest in African American adults and lowest in Asians with Caucasians and Hispanics roughly equal to one another and in the middle.  The difference wasn’t huge about 1 point between the African American and Caucasian/Hispanic individuals but it was systematically higher.

It should be fairly clear, simply by observing the real world that blacks (of West African descent) are often more muscular and leaner than whites (although they are as likely to be more obese and these factors are likely related).  Those athlete’s  invariably dominate sports with an explosive component (for various reasons) and I don’t think the issue can be ignored in this discussion.  Unfortunately, the ethnicity of the subjects in Khouri’s paper was not indicated but I am going to *assume* that they were all or mostly white.  There is almost assuredly an ethnic/genetic effect here.

Let’s Count it Up

Ignoring that for now, with the above add a handful more big time exceptions to the FFMI cutoff. A couple of Sumo wrestlers, 8 elite male bodybuilders and the strongest RAW powerlifter in the world.  Certainly there are more but this brings us to a total of about 36 exceptions so far.  This assumes that we can take at face value the strongmen of the late 19th and early 20th century and if you still believe that even a majority of the post-testosterone availability Mr. USA competitors were clean.

If we take out the Mr. USA competitors who are likely to have been using by at least the early 40’s, we’re left with 9 early 20th century strongmen (still questionable to me), a handful of Sumo wrestlers, potentially some elite bodybuilders and Ray Williams.  So about 15 identifiable (yes, there were assuredly more) people over a 100+ year span.   This makes less than a compelling argument to me.  No the cutoff isn’t absolute in a literal sense.  But are we really giving 15 or so total individuals that much weight in this argument?

An Analysis of Top Natural Bodybuilders

I’m going to start with an assumption that I know not everybody agrees with which is this: bodybuilders are likely to have the highest levels of muscle mass of any athlete as they train specifically towards that goal (for other athletes, gaining muscle is a means to an end, not an end unto itself).

The counterargument to this is that the top athletes are more likely to go into money sports like football or basketball and are more likely to have the highest level of muscle mass.  Certainly the data that exists (including older papers on football and basketball players I couldn’t get the full texts of) support this with often high levels of FFM being seen.

But bodyfat levels are often much higher which skews this as I keep repeating. Yes, Shaq is a monster.  He also carries a lot of bodyfat.  Since I can’t get the raw data, I can’t analyze it or present it.  You can agree with my assumption or not and that’s fine.  I am as likely to be right or wrong so far as I am concerned.  IF you have those papers send ’em to me and I’ll make an addendum.

But based on that assumption, I think it’s worthwhile to look at the FFMI in top natural bodybuilders as being at least representative of maximum or near maximum levels.   And I’m going do to this by looking at an analysis of top natural bodybuilders.

Now, I forget exactly when I did this analysis or why I have this list of competitors in the first place (I like to think I got it from some major natural competition but I honestly can’t remember).  I know it was a few years ago and may be slightly out of date and I don’t even know how I obtained the height and weight to begin with (either I obsessively combed the Internet for it or, more likely, found a list and just did some extra math).

I am not presenting this as proof, it is meant to be illustrative only.  What it is is a list of the top natural bodybuilders (well bodybuilders competing in a natural federation) in different weight classes.  And I thought it would be interesting to calculate their FFMI since this is likely to be illustrative of near maximum values in natural bodybuilders who I think are likely to represent the upper limits of FFM/FFMI to begin with.  Since most of the people who care about the FFMI limit are seeking maximum muscle mass, this should at least be indicative.

What I did was take their listed competition weight and height and calculate out their FFM based on an assumed body fat percentage of 4% so that I could calculate both the raw and normalized FFMI values.   The list goes from the lightest to heaviest classes and I’ve marked all the lifters who surpass the 25 cutoff point in red.

Note immediately that the majority of those lifters in the heaviest classes with only two being in the lighter classes.  This is important as there is a height issue at work here with heavier competitors typically being taller.  Even that skews FFMI up because overall bigger people have more organ mass, bone mass, and other non-muscle mass simply as a function of being bigger to being with.

Bodybuilder FFMI ChartLooking at the non-normalized FFMI value, a mere 5 of the 14 cross the 25 FFMI value. Using the normalized FFMI, 6 of the 14 cross 25 as Jeff Nippard goes from 24.4 to 25.4.  Looking at the normalized values, the lowest value is 22.1 while the others are in either the 23 or 24 range.  They get close but don’t get across it and these are elite male bodybuilders who won their class.  And less than 50% cross the cutoff.

Presumably the other competitors in their class weren’t at a higher FFMI (yes, I KNOW that bodybuilding is not judged only on mass and someone could be bigger but less conditioned or symmetrical or whatever so spare me that strawman).  This is the best of the best and the majority still don’t make it across the cutoff point.

Because I am nothing if not a petty prick I’ll mention that one of the most vocal critics of the FFMI cutoff, who I shall not name, doesn’t get to much less past the 25 cutoff point himself at least as of the date of this analysis which was several years ago.  I’m told he’s bigger now but I doubt it’s by much since he’s clearly near the upper limit of his genetics.  I could remath him with a few more pounds of LBM but it won’t change much.  I suspect it still doesn’t get him across the 25 cutoff, just a little closer to it.  I am not saying that dismisses his opinion out of hand, I leave those kinds of ad hominems to the other gurus in the industry.  I am just genuinely being a petty prick about the topic.

Irrespective of this, the fact is that the majority of the list don’t cross it and many fall several points short.  And these are presumably the best natural bodybuilders that exist.  The average lifter doesn’t have a chance anymore than they have a chance of squatting 1050 like Ray Williams.

And even with those 5-6 exceptions, all I think the above proves or demonstrates is that the 25 FFMI cutoff for natural status is not absolute.  I won’t disagree with that and I’m not sure I ever disagreed with that (I certainly hope I didn’t).  Fine, we could handwave that even some of the above guys are using (it’s not as if drug users or former drugs users don’t enter natural contests) but I think that’s a copout on every level.

It’s simply too pat to define any exception as being on drugs since it means that you have defined it in such a way as for there to be no exceptions.  Easy way to be right but intellectually dishonest.  As well, I know three of the athletes on the list “personally”  (inasmuch as you can know anybody online personally) who cross the cutoff and I have no reason to doubt their natural status.  So, I will take at face value that all bodybuilders on this list are natural.

I would note that these are guys in a fully lean and depleted state.  Their FFM due to water and glycogen would be higher in the off-season and this would raise their FFMI.   If we normalized them to the 12% average in the original Khouri paper, their FFMI would be higher and a few more would probably make it across the threshold.  But I’m tired of doing math so I’m only mentioning it for completeness.

At best I imagine many would make it to 25, one or two more might get across it and the top guys would be hit even more insane numbers.  I don’t think it changes the point I’m trying to make since it’s still only like 6-8 total people, the best of the best natural bodybuilders, which is still inconsequential in the big picture.

I also think there is a mitigating factor that explains 4 of the lifters in the list who surpass the cutoff, interestingly the 4 highest values.  Now in the original Excel file, the lifters were all linked to online pictures (probably Instagram) but those didn’t transfer with my screen shot (and I am far far too lazy to put the above in a WordPress chart since the interface sucks).  So here are links to the pages of each of the lifters who surpassed the 25 FFMI cutoff along with their FFMI values.  I put them in order from lowest to highest.

Evan Goodbee (25.3)
Jeff Nippard (25.4)
Marques Morgan (26.2)
Vernon Williams (26.5)
Josh Gilliam (29.7)
Nsima Inyang (29.7)

A first thing to notice is that four lowest values of 25.3-26.2 are consistent with the estimated FFMI from the elite bodybuilder paper above so that seems to pass the reality check (also perhaps suggesting that those elite bodybuilders were in fact natural).   As you can see, the two highest numbers are significantly higher (but still below the claimed FFMI for Grimek which throws even more doubt on his natural status)

Now go click the links in order from top to bottom and perhaps you’ll notice something.

The first two bodybuilders who cross the cutoff (just barely), Evan and Jeff, are both white while the 4 highest values are all black.  The first two black bodybuilders, Marques and Vernon clock in exactly one point higher than Even and Jeff respectively, the same one point difference seen in the paper on ethnic differences I talked about above.  I would suggest that they started 1 point higher when they began training and, thus, ended up the same one point higher when the dust settled.  That is, if Evan and Jeff started at 21 and Marques and Vernon started at 22 and both gained the same 4 FFMI points over their training career, they end up at 25 and 26 respectively.

Then there are the final two bodybiulders, Josh and Nsima who are in another universe, almost clearing 30 which is approaching the values of the dieted down Sumo wrestler.   Nobody can touch Ray Williams at his current bodyweight but, based on VERY ROUGH math that I’m not showing, at contest lean he comes in only 3 points higher than Nsima (~33) with about 7 kg/15 lbs of extra LBM.  Mind you, dieted down that far, he might be higher or he might be lower.  But diet him down and the differences start to evaporate.

Let me note that Nsima’s FFMI actually EXCEEDS Arnold’s at his best by about 2 points which is fascinating in its own right.   This is also a man who clearly has genetics that put him as one of the elite the elite, the bodybuilding equivalent of Ray Williams who just so far outstrips anyone it’s kind of disheartening to look at.

Which again only proves that the occasional genetic freak can exceed the cutoff point, seemingly by a large value.  And most of the bodybuilders on this list still don’t get across it.  Which means that the average lifter has about zero chance of getting near it much less above it naturally.

Summarizing This FFMI Nonsense

So what have we learned so far?  An early study suggested/argued that anybody who exceeded an FFMI of 25 kg/m^2 was on steroids (or rather that the cutoff could be a screening tool) and this is provably/obviously false.  Because in this series alone I’ve identified maybe 30 athletes in the history of the weight game (and again there are assuredly more) that have crossed it although some are debatable at best.

I still think that the strongmen are questionable based on the (non) records of the day, by the 1940’s anabolics were in use bringing most of the Mr. USA winners into question (Grimek was somehow bigger than Arnold on top of gaining 27 lbs of muscle in a year).  Then we have 8 elite bodybuilders, some Sumo wrestlers, the best RAW powerlifter on the planet and 6 current natural bodybuilders.

Depending on how you add it up and how generous you are, it’s maybe 35 people tops.  Yes, there are probably more than this out in the world who aren’t represented here.  We all know about that dude who benches 600 lbs for reps in his cousin’s basement who won’t go to a meet because he trains for himself, right?  Or the friend of your best friend’s girlfriends first cousin who is 300 lbs and ripped with 26″ arms but doesn’t have a camera on his phone to prove it.

Yes, I’m being obnoxious. There are assuredly more folks out there who naturally and provably exceed the 25 FFMI cutoff without carrying a ton of BF% or artificially skewing the value in some other fashion.   But let’s do a reality check on this.  Those are 35 people, and hell I’ll be generous, let’s say we’re missing another 65 folks who weren’t measured and call it 100 exceptions to the 25 cutoff.  Hell, let’s call it 200.  The elitest of the elite in bodybuilding or the strength sports or maybe some team or power sports.  The top 1% of the top 1%.

How many provably natural athletes over the history of the 20th and early 21st century have lifted weights or strove (strived, striven? what’s the past tense of strive?) for maximum muscle mass.   I’d bet it’s in the tens of thousands or possibly more.  Maybe even hundreds of thousands, I mean how many men are lifting right now for maximum muscle mass?  I can’t begin to fathom the numbers.

Men have always been more or less interested in weight training and getting big, Strength and Health magazine dates back to 1932 and there was an absolute boom in the 70’s when Arnold made it fashionable to bodybuild.  Add to that the absolute EXPLOSION of interest in physique sports in the last decade (not all being bodybuilders seeking maximum size mind you).  We are not talking about a small population now.  I don’t know how big it is but it is absolutely enormous.

And the best anybody can do is throw a few dozen, or 100/200 if I’m being nice, exceptions to the rule?  C’mon.  This doesn’t disprove the 25 FFMI cutoff at all.  If anything it makes the point.  More specifically, it makes the following point:

Even if a 25 FFMI cutoff is NOT a literal absolute and can NOT indicate if someone is using steroid without exception, it might as well be.  I mean, let’s say there have been as few as 100,000 males seeking muscle mass over the past 100+ years.  And you can come up with at best 100 who got across the threshold?  That’s 0.1% of the total lifting population ever.

ZERO POINT ONE PERCENT.

That’s assuming 100,000 total folks over the decades is a good number.  If it’s lower, the percentage goes up a bit.  If it’s higher, the number drops even lower.    Hell, let’s say my estimations are low, which they may be and probably are, and all of 1% of the total lifting population naturally gets across the 25 FFMI cutoff without being fat as hell or manipulating FFM with creatine or carb loading or whatever.   Hell, let’s be super generous and say it’s a whopping 2% who get there.

For all practical purposes, it might as well be zero.  Yes, there are handfuls of exceptions, I can’t say how many.  But they are ultimately completely and utterly insignificant in the big picture.  Because while I might not be able to say with 100% certainty that anybody over the 25 FFMI cutoff is using anabolics, I can say it with roughly 98-99.9% certainty.

Now, if I told you a rule held in 98%-99.9% of cases and that the exceptions were too few to worry about, would you spend much mental energy on the exceptions or make a big stink about those exceptions?   If I told you that you had a 1% chance of surviving jumping out of a place with a parachute (people have done it) would you put much stock on you’re being one of them?

Even if my numbers are wrong and 2% get across the FFMI cutoff…would you jump out of the plane with a 2% chance of survival?  No, no you wouldn’t.  You’d be happy to consider the rule absolute because it might as well be.  Sure, I can’t predict for any given individual if it holds but statistically I’m more likely to be right than wrong in assuming you’ll die.

What I am saying is this:

  1. Yes, exceptions exist to the FFMI cutoff of 25 in terms of they’re being natural or not. Of that there is no doubt.  But all those exceptions are are exceptions and, in a real world sense, the number is likely to be so small it might as well be zero.
  2. The 25 FFMI cutoff is not absolute in a literal sense.
  3. But it might as well be in a practical sense.

Because I can say without hesitation that if I meet someone relatively lean who is above the 25 FFMI cutoff, there is a  ~98-99.9% likelihood is that they are using steroids. I might be wrong.   But statistically speaking I won’t be anymore than expecting someone to die if they jump out of a plane without a parachute.    If you asked me to bet on it, that you’d die jumping out of a plane or that someone got past the 25 kg/h^2 is using steroids, I’d certainly be likely to put money down that you would/they were.

Beating the Dead FFMI Horse

Because, beating that dead horse, if only 1% of the top 1% of all people who have touched a weight make it across the threshold, you probably aren’t one of them.  I mean, let’s look at this from a different perspective that might not carry the same emotional weight for people (who can’t seem to handle the idea of a limit on the ability to build muscle mass).

In the history of the 100m sprint, a sport contested since the early 20th century, a total of 136 sprinters have run faster than 10 seconds by electronic timing.  136 against how many thousands and thousands of sprinters that have ever tried.   The first wasn’t until 1968 and realistically most of those guys were on steroids which were already well in use by that point in sports (remember that they were legal until the early 80’s and we know they are still being used now).

Many of the more recent ones are as due to training as changes in the track and they are still using drugs (see also: the Balco scandal).  I mean, just watch the Olympics any summer.  The best of the best athletes in the world go to the games, maybe the top 80 sprinters in the world at that time, and only a handful have or even have a chance of breaking 10 seconds.  It’s just a miniscule percentage of the total number of sprinters who have tried.

So if I claimed that anybody running faster than 10 seconds in the 100m was on drugs, would you argue?  If I told you that 10 seconds in the 100m (and likely slower) was a cutoff that nobody can get past without drugs, would you argue?  Because even if 1 or 2 on that list was natural (which is a stretch to begin with), that means that 1.4% tops did it clean.  So with 98.6% certainty, I can say that if you did it, you weren’t clean.  And similarly I can say that no amount of work ethic is getting you past 10 seconds without drugs (and realistically with them since it requires top genetics too).

Same thing here.  If only 0.1% or even 1% or 2% of folks can clear a 25 FFMI cutoff without using steroids, it might as well be an absolute rule for all practical purposes.  And the pushback against it (or the idea that there are any genetic limits) is baffling.  Throwing 6 provably natural bodybuilders, or even double that number (or digging deep for late 19th century strongmen), who made it across doesn’t disprove the 25 cutoff at all.

Rather, it makes the point of how strong it is.  Because those 6 exceptions don’t matter in the big picture.  Throwing 30 exceptions in the history of the game, or 100 or 200 doesn’t matter in the big picture because it represents an insignificant number of people compared to the total number who have ever trained consistently, intensely and relentlessly and never gotten even close much less there or past it.

And the only reason I can fathom that nobody wants to accept that 25 is a damn hard cutoff for the grandest majority of lifters in all but the tiniest number of exceptions is because bodybuilding, perhaps moreso than any other sport, is an activity marked by sheer delusion (and an industry that feeds it) about how big you can get.

When the top pros are 290 ripped, a guy at 180 feels small even if he’s huge compared to most of humanity.  And a supplement industry finds it in their best interest to sell false dreams to hopefuls that they can keep growing for ever.

The 25 FFMI cutoff point (or the suggestion that 100 kg is about the upper limit for FFM without drugs) suggests a genetic limit that can’t be surpassed without steroids and bodybuilders and other athletes don’t want to believe such a limit exists.  So you get idiotic arguments trotting out a few dozen exceptions as if that means that the other 99% of lifters or athletes can even get close.  Or that the rule won’t hold in such a vast majority of cases that it might as well be absolute.

A final point before moving on: I am prepared to speculate/argue/accept that African American athletes will have a higher cutoff point. I might handwave based on the one study that it’s 26 compared to 25 (and possibly lower for Asian athletes) since it was 1 point higher to begin with.  In the analysis of bodybuilders I did above, two of the black athletes had an FFMI one point higher than the biggest white bodybuilders and the other two were much higher (with Ray Williams still holding the record so far as I can tell).

This is MY SPECULATION and I’d love to see comparative research. But it
wouldn’t surprise me if this turns out to be the case.

I am also prepared to believe that there may be, err, regional differences I guess.  I mean this in the sense that it’s fairly clear that folks in different parts of the world are often just big, mean and strong genetically.  All of those ver Magnussons from the Norwegian countries along with Island Somoa and Tonga come to mind and it wouldn’t surprise me if they are more likely to get across a cutoff naturally than your basic pasty white guy.

I’d love to see this studied systematically as well.  Just do some comparative work and see what the baseline FFMI of different groups based on region and ethnicity come in at.  Because this is likely to help you find out where any upper limit of FFMI is that can be achieved with training is likely to fall and whether it too varies in this fashion.

So yeah, that was a long way to go for a rather basic conclusion. But I’m not done.  In part 3, I want to address a different issue entirely and ask the following related question:

Is FFMI even what we should be talking about?

Read Part 3