What you’re getting yourself into:
5,500 words, 18-36 minute read time
- There is no way to know for sure how much muscle it’s possible to build drug-free. It’s naive to assume that pre-steroid era bodybuilders reached the absolute limits of drug-free muscularity, but now that Pandora’s box is open, we can never be totally sure about modern lifters who claim to be drug-free. We’ll never know for sure where the limit lies.
- Instead, we can estimate the likelihood that someone is drug-free or not based on their degree of muscularity. How? Math, of course. Don’t worry, we made some calculators that do all the heavy lifting for you.
- Over the course of a training career, it seems like steroids allow you to build about twice as much muscle as you’d be able to build drug-free.
This is the third installment in our steroid series.
The first, The Science of Steroids, laid the groundwork by discussing how steroids work, both mechanistically and via the powerful expectancy effects they exert (if you give people a placebo and convince them they’re taking steroids, they still make pretty eye-popping strength gains).
The second, Steroids for Strength Sports: The Disappointing Truth, examined the magnitude of effect steroids have on strength gains – smaller than many would expect, but giving a pretty huge advantage at the elite levels.
In hindsight, I should have written this current article as the second installment, since the strength advantage is primarily attributable to the additional muscle you can gain when you use steroids. Once you get an idea of the muscular advantage steroids provide, it’s pretty straightforward to come up with reasonable estimate for how much of a strength advantage they’ll provide since muscle mass and strength have a very close relationship (assuming you’re comparing two lifters of similar skill levels). We explored that relationship in this series:
This article seeks to shed some light on a common question: It’s obvious that steroids help people gain more muscle than they’d have been able to gain drug free … but how much more?
I want to make it clear from the outset that I’m not going to attempt to answer this question in the common (and quite frankly, lazy) manner people attempt to answer it.
The most common approach is simply to compare bodybuilders before the steroid era to bodybuilders today.
On its face, this seems like a pretty reasonable approach. However, upon closer examination, there are two fatal flaws with this approach:
- In nearly every other sport in existence, results of the very best improve over time as the sport gains more exposure and as incentives increase.
- This approach only tells you about the difference between drug-free muscle growth and muscle growth with every drug under the sun. Most people who use gear are never going to run a drug cycle comparable to today’s top bodybuilders, and this type of comparison doesn’t tell you what would be reasonable to expect from a more restrained approach to PED usage.
The first critique is very straightforward. When the talent pool in a sport grows, freakier freaks come out of the woodwork. The top performers get results several standard deviations better than the average person. The further you move from the mean, the lower your odds of finding someone with that level of talent in a given talent pool. The more athletes you have in a sport, the higher your odds of finding someone a bit further from the mean.
This is easy to see if you compare pro athletes in the 1940s and 1950s to pro athletes today. Here’s the best running back in the NFL in the 1940s vs. the best of this generation – Steve van Buren vs. Adrian Peterson.
Has Peterson had better training than van Buren? Almost certainly. Is there the possibility that Peterson used pharmaceutical enhancements than didn’t exist in van Buren’s day? It’s certainly possible. But I don’t think anybody would seriously contend that van Buren could compete in today’s NFL (even if he had all the advantages afforded to modern players), or that Peterson wouldn’t have made the NFL of the 1940s his stomping ground.
This trend is clear across the board in the NFL. This site has a neat illustration showing the average size of NFL players over time.
This trend is also true in essentially every sport. Very few pro athletes in the ’40s and early ’50s would even beat the top high school athletes today. This is subjective in team sports, but you can compare the high school track records to the world record progressions in the same races; most world records “passed” the current high school records in the mid-’50s to mid-’60s. Heck, Doug Hepburn became the first 400lb bencher in the early ’50s, and there are plenty of high schoolers who can bench press over 400lbs now.
As sports grow in popularity and as incentives increase, more people compete, so the best of the best keep getting better and better.
Before you cry that all of that progress is due to drugs, let’s step back for a moment. Drug testing in the Olympics didn’t start until 1968, and the World Anti-Doping Agency wasn’t created to combat the ongoing problem of drugs in the Olympics until 1999. Drugs can certainly explain some of the progress since the advent of steroids (first used widely in competitive sports by the Soviets in the 1952 Olympics), but if they were the key factor, you’d expect the records from before 1968 to be untouchable, and you certainly wouldn’t expect people to keep setting records after 1999.
In fact, likely because steroids would be expected to give women a bigger boost, the female shot-put record set in 1987 is almost a full meter farther than any throw since 1999, and 22 of the top 25 throws of all-time occurred before 1999. WADA certainly doesn’t catch everyone, but that’s solid evidence that it is doing its job fairly well.
Sure, athletes still use, but they can’t use as much anymore, and they certainly can’t use as openly. At the very least, they have to cycle off for international competitions, which impacts performance; yet performances keep improving. Training has likely improved, but most of the progress is attributable to larger athlete pools.
In other words, assuming that the pre-steroid era bodybuilders of the ’40s and early ’50s represented the absolute peak of drug-free muscular development is extremely naive. Bodybuilding’s athlete pool was certainly more limited than that of Olympic sports. Calling bodybuilding an underground sport in the ’40s and ’50s would be an understatement. The modern gym with weights didn’t really take off until the ’60s. It’s impossible to get an exact count of aspiring bodybuilders in that era, but the number was certainly tiny – certainly tinier than the athlete pools in the other sports producing (comparatively) amateurish performances during that era.
Modern bodybuilding has way more visibility, larger incentives, and a dramatically larger athlete pool than it had in the ’40s and ’50s. Comparing the top pros of the ’40s and early ’50s to the top pros of today certainly tells you something about the effects of steroids, but the comparison is hopelessly confounded by the effects of a dramatically larger talent pool. If you added steroids to the NFL in 1946, you still wouldn’t get talent comparable to the NFL in 2016 (although obviously the NFL has way more visibility and way larger incentives than modern bodybuilding).
The second problem with comparing bodybuilders of the ’40s and early ’50s to today’s pros is that such a comparison only gives you hints about what is maximally possible with huge amounts of drugs. The average steroid user isn’t on the same drug stack as IFBB pros.
There’s a clear dose-response relationship with steroids. The more you do, the more muscle they help you gain. If you’re not planning on using the same drugs in the same dosages that pro bodybuilders use, seeing how much of a boost they get from their stack doesn’t give you realistic expectations for what to expect from a more moderate approach.
Before we dig into the data, let’s just briefly discuss hypertrophy in a general sense.
When you first start lifting, you gain muscle pretty quickly. Over time, the gains taper off. You may gain a little faster when you’re in a calorie surplus, lose a bit of muscle in an aggressive deficit or if you took a break from the gym, or gain muscle fairly quickly for a period of time after your newbie gains if you found a program that really clicked with you; but on the whole, if you graphed your hypertrophy progress, it would look like a curve that initially rose rapidly, tapering off until it started approaching a horizontal asymptote that corresponded with your genetic potential.
It’ll look something like this:
When you add drugs into the mix, you initially gain muscle very rapidly. It’s sort of like another period of newbie gains, and perhaps even more extreme.
For example, in this study, reasonably well-trained men (90-110kg/200-240lb bench and 100-125kg/220-275lb squat on average) went on 600mg of testosterone per week for 10 weeks. The people who used test while also lifting gained about 6.1kg (about 13.5lbs) of lean mass, on average. That’s a ton of muscle! That’s more than the drug-free high responders (the people above the 85th percentile or above for mass gain) in this study on untrained men.
However, those easy gains obviously don’t carry on forever. It’s very similar to drug-free training – rapid gains initially, tapering off over time.
With more drugs, unsurprisingly, you gain more muscle. As previously stated, there’s a very clear dose-response relationship. If people could just use 500mg of testosterone per week and keep gaining muscle indefinitely, there would be no reason for people to take more and more compounds at higher and higher doses.
The effect is basically the same – a quick spurt, followed by leveling off. The second (third, fourth, fifth, etc.) growth spurt generally isn’t as big as the first one, assuming someone is increasing their dosages gradually, but the general effect is the same.
There are two key points from all of this before we move forward:
- There’s a very strong dose-response relationship for both short-term and long-term hypertrophy. This was the second (lesser) problem with comparing old-school bodybuilders to today’s pros. The more you’re on, the bigger the gap will be between drug-free and drug-enhanced results. It’s helpful to get an idea of the effects of a more reasonable level of usage, and not just compare drug-free results to extreme steroid usage.
- We need to separate short-term and long-term results. It would be great if the people who gained 6.1kg of lean mass in their first 10 weeks on steroids could keep up that rate of progress forever, but that’s just not what happens. The effects of a certain level of usage resemble the process of gaining muscle without drugs – fast initial gains that progressively diminish.
Now it’s time to get into the meat of this article and answer the key question: In the long run, how large of a hypertrophy boost do people get from “normal” levels of steroid usage?
I came across three studies that shed some light on this question.
The first two use the Fat Free Mass Index, which is a measure of lean mass relative to body size, calculated by dividing fat free mass by height2. This allows you to compare the muscularity of people who are different heights. Obviously, taller people will generally have more lean mass, and they also generally have more lean mass per unit of height. By squaring the height component, you get a slightly better comparison. There’s a little more to it than that (dividing by height2 still doesn’t give a perfect comparison, so a further correction for height is also added), but that’s not a rabbit hole worth going down in this article – just know that it’s a measure of lean mass relative to height.
The third study, unfortunately, didn’t report the heights of the subjects, so it’s impossible to calculate FFMIs. Lean mass is the only thing we have to go off of. However, it’s also the study that employed the most highly trained lifters.
In it, the researchers compared a group of 74 non-users to 83 steroid users. Even though they administered urine tests to ensure their non-users didn’t have any drugs in their system at the time, there’s obviously no way to know for sure that all of the lifters in the non-users sample were, in fact, drug-free. However, that’s not nearly as problematic in research as it would be in drug-tested sports. Eric Helms sums it up well:
In this study non-users were determined by interview. Those stating they never used steroids took a urine test to confirm they were non-users. One could argue dishonest users currently not “on cycle” may have been included as non-users. While plausible that some may have lied about their status and passed the urine test, it seems unlikely. Research participants do not gain recognition. They are de-identified and represented as part of a group. Also, there were no consequences to reporting steroid use. If the individual reported use, they participated in the study as part of the “user” group. Finally, participants would be told what the study would entail before inclusion, and those uncomfortable with interviews would not have likely volunteered. Nonetheless, perhaps due to the generally negative perception of steroid use, some may have lied and went on to pass the urine test.
Sure, there’s a chance that some users slipped through, but the odds that enough users slipped into the non-user sample to meaningfully skew things is very small.
There are two other things to note about Kouri’s study:
- The participants were all recruited from gyms, so they all presumably had some training experience. However, they weren’t required to be super well-trained. The non-user group included a few successful natural bodybuilders and strength athletes, so at least a few highly trained athletes were in the sample, but it’s probably safe to assume that both group averages were a little lower than they would have been if only highly trained lifters were included. However, most people do train for at least a few years before using steroids, so it’s also reasonably safe to assume the users were more highly trained, on average, than the non-users.
- The users’ FFMIs, in particular, were likely depressed considerably. The users included both people who were currently on steroids, and people who had previously used. A full third of the users in this sample hadn’t used steroids in the prior year.
In other words, the data from Kouri’s study is certainly more useful than merely looking at the short-term effects of steroids, but it certainly has its flaws; the participants certainly weren’t untrained, but we’re not entirely sure just how well-trained the average participant was, and it’s impossible to know how much muscle the people lost who had previously used, but weren’t using at the time of the study. In all likelihood, the means of both groups were a bit lower than would be expected comparing people approaching their muscular potential, both with and without drugs.
With that in mind, the average non-user in Kouri’s study had a Fat Free Mass Index of 21.8±1.8, and the average user had an FFMI of 24.8±2.2.
Here’s how those two curves look:
As you can see, the steroid users were considerably larger on the whole, but there’s a fair amount of overlap, especially in the 21-24FFMI range.
Our second study is by Brennan. It was very similar to Kouri’s study, insofar as it compared the FFMIs of users and non-users. It likely did a better job of getting relatively well-trained athletes, since participants were required to bench at least 275 to participate – not a huge feat, but certainly one that requires a fair amount of training for most folks. Additionally, training ages were reported. Most participants had at least 6-7 years of experience in the gym.
One drawback to Brennan’s study is that, unlike Kouri’s, there was evidence of attempted deception. Several people had to be excluded from the non-users’ group because they were either definitely on steroids (they failed the urine test), or probably on steroids (in the researchers’ words, they had “implausibly high muscularity and low body fat despite denial of AAS use”). However, even if a couple of users slipped in, they probably didn’t skew the data too much. The sample size for nonusers was big enough (131 people) that a few bad apples wouldn’t spoil the whole bunch, and the researchers were excluding suspiciously jacked people anyways, so if a few users did slip in, they apparently weren’t swole enough to raise any red flags.
Brennan’s study compared three groups of lifters: non-users, people who had used steroids but not human growth hormone or IGF-1, and people who had used steroids along with hGH and/or IGF-1.
The average non-user in this study had a Fat Free Mass Index of 22.8±1.9, the steroid-only user had an FFMI of 23.3±2.3, and the average person who’d used hGH and/or IGF-1 along with steroids had an FFMI of 26.2±2.8.
Here’s how those three curves look:
Surprisingly, the people who had exclusively used steroids weren’t very much bigger than the non-users in this study. The average FFMIs are only separated by 0.5 points. However, this may be due to the fact that most of the steroid-only users hadn’t been using for very long, with a median of about 6 months. The folks who added hGH and/or IGF-1 into the mix, though, had been using for much longer, ranging from 2 to 7.5 years, with a median of close to 3.5 years.
The last study by Yu does not report FFMI. It only reports lean mass. It’s unfortunate that we don’t have FFMI data for this study, because it was on the most highly trained athletes out of the three. All the drug-free participants were national-level powerlifters, and all of the users were strongman and bodybuilding competitors. Most of the users were on a pretty fair amount of anabolics.
What we do know is that the drug-free athletes had 74.6±6.8kg of lean mass, while the users had 89.8±8.2kg of lean mass – a 15.2kg (~35lbs) difference. If we assume they were the same height, that would mean a difference of about 4.5-4.8 FFMI points. My hunch is that the FFMI gap would be slightly smaller since bodybuilders and powerlifters tend to be similar heights, while strongmen (who were only in the users’ group) tend to be quite a bit taller, but we can’t say for sure.
So, taking these three studies in totality:
- We’re looking at a total of 4 comparisons: between the users and non-users in Kouri’s study, between the non-users and steroid users in Brennan’s study, between the non-users and the people who used steroids along with hGH and/or IGF-1 in Brennan’s study, and the comparison between users and non-users in Yu’s study.
- Of the four comparisons, Yu’s would probably tell us the most about long-term potential with and without steroids since it used the most highly trained lifters, except for one pesky detail: heights weren’t reported. It’s impossible to tell how much of the difference in muscle mass is attributable to steroids, and how much may have resulted from the people in one group just being taller.
- The comparison between non-users and people who only used steroids in Brennan’s study isn’t much use to us here. We want to get an idea about the long-term impact steroids have on muscle growth – how much do they raise the ceiling? The steroid-only group in that study hadn’t been on gear long enough to help us answer that question.
- Of the two remaining comparisons (Kouri’s and Brennan’s comparing non-users to people on steroids plus other anabolics), Brennan’s subjects were more well-trained. However, there are two major flaws: 1.) The group on hGH/IGF-1 was the smallest of the bunch: 27 candidates vs. 70-80 in both of Kouri’s groups, and 100+ in Brennan’s other two groups. 2.) The group on hGH/IGF-1 had an advantage over the drug-free group beyond steroid usage: They were about 5 years older (32.5 years old vs. 27.8), and they’d been training for about 3 more years (11.8 years vs. 8.6). On the other hand, it’s hard to know which group was at an advantage in Kouri’s study. On one hand, a third of the users had been off steroids for a year or more, which would obviously handicap them. On the other hand, since there wasn’t a minimum amount of training experience required, odds are pretty good that the steroid group contained more experienced lifters; since most people don’t hop on gear as soon as they start training, the odds of having quite a few new-ish lifters in the steroid group is lower.
In these final analyses, I’ve decided to pool the data from Kouri and from Brennan’s comparison of drug-free lifters vs. people on hGH/IGF-1. I personally think that Brennan’s data is better for elucidating long-term hypertrophy differences, but I ultimately feel slightly better analyzing data from larger samples, and the gap between the users and non-users was pretty similar in both comparisons anyways (3.0 FFMI points for Kouri, and 3.4 for Brennan).
Using these average numbers, the non-users had an FFMI of 22.3±1.9, and the users had an FFMI of 25.5±2.6.
Now we can answer a few burning questions:
1) How jacked can someone possibly get drug-free?
The most epistemically honest answer is that I don’t know, no one knows, and there’s really not a good way TO know.
We have a pretty good idea of how jacked a few people got before steroids burst onto the scene, but like I said in the intro, odds are horrendously low that any of the members of that tiny subculture attained the absolute pinnacle of drug-free muscular development. If people have continued to improve in every other physical domain to a degree that can’t be solely attributed to performance enhancing drugs, I can’t see a good reason why bodybuilding would be an exception.
Unfortunately, it’s also impossible to establish a maximum degree of drug-free muscularity with modern data. Quite simply, drug-testing doesn’t catch everyone (and without out-of-competition testing, it only catches people who are really dumb), and it’s easy for someone to claim they’re drug-free without ever having been drug-tested. You can never be 100% confident that an individual – much less everyone in a group – is, in fact, drug free.
In other words, it’s impossible to know for sure.There is no way to know for sure how much muscle it's possible to build drug-free. Click To Tweet
Some people like to claim that an FFMI of 25 is a hard cutoff, based on Kouri’s study, because the researchers didn’t find any drug-free people in their sample with an FFMI over 25. However, this is a very naive way to approach the question, because:
- Their drug-free sample was only 74 people. When you’re asking how big someone can get drug-free, you’re inherently asking about the outliers. In a sample of 74 people, your odds of finding someone at the limits of any trait or ability are horrendously low; you’re probably not going to find someone who runs a sub-10 second 100m dash, or who’s 7 feet tall, or who has Einstein’s IQ. Even in a 74-person sample of exceptionally fast people, exceptionally tall people, or exceptionally smart people, you’re probably not going to find the next Usain Bolt (100m dash of 9.59 seconds), Robert Wadlow (8’11” tall), or Terence Tao (IQ over 200). The researchers DO note that a handful of their subjects were successful natural bodybuilders or strength athletes, but since they didn’t have an experience requirement, it’s highly unlikely that all (or even most) of the subjects were at or near their drug-free muscular potential. Even if they were, it’s still very unlikely you’d find someone at the limits of what could be attained drug-free.
- Quite a few of the bodybuilders in the pre-steroid era did exceed an FFMI of 25. Some people like to write that off and say that they’re errors due to the researchers needing to estimate body fat percentages visually. However, odds are pretty low that the estimates across the board could have been that bad. Even if the estimate was off by 5%, that would be about 10lbs for someone who’s 200lbs (4.5kg for someone who’s 90kg). Using a “standard” height of 5’11” (180cm), that would mean an error of about 1.4 FFMI points. If you knocked 1.4 FFMI points off of all the pre-steroid era Mr. America winners to assume large systematic errors in measuring body fat percentage, you’d still have 2 people with FFMIs above 25 during the era where there’s almost no chance of them acquiring steroids, and 3 more if you included people during the era where odds would be very low they’d be using steroids.
So if the limit’s not 25, what is it?
I feel very confident that it’s above 28, since that was accomplished by George Eiferman during a time when steroids existed, but weren’t widely used or readily available. During the period before steroids were available at all, the biggest Mr. America winner had an FFMI of 27.3 if you’d like to use that as a “conservative limit” instead.
How much higher than 28 would be possible? It’s hard to say.
Averaging the data from Brennan and Kouri (22.3±1.9), an FFMI just above 26 is two standard deviations from the mean, meaning it’s very unlikely for a particular individual to achieve it, but about 2-3 people out of 100 could do it. 28 is three standard deviations from the mean, meaning it would be attainable for about 1-2 people per 1,000. Four standard deviations has you at 29.9, knocking on the door of an FFMI of 30, with the odds for a single individual down to around 1 in 100,000.
Obviously, you could keep going further and further from the mean, with the odds getting smaller and smaller, until you wound up with infinitesimally small odds that a drug-free person could become as massive as a black hole. However, around an FFMI of 30 – 4 standard deviations from the mean – doesn’t seem to be too much of a stretch. I would not be willing to say it’s either possible or impossible, but it certainly seems somewhat reasonable. (Of course, a human myostatin mutant, lacking any working copies for the allele that makes the hormone myostatin which limits human growth, could perhaps even blow that number out of the water).
Of course, all of this assumes that Kouri and Brennan’s data are normally distributed. Brennan didn’t report the FFMIs of each individual, and Kouri’s 74 individual data points aren’t really sufficient to be overly confident about the kurtosis and skewness of the sample. Unless you have a big enough sample to establish kurtosis and skewness, talking about things happening 3+ standard deviations from the mean should be treated as speculative at best.
It is true that most human traits and capabilities are quite normally distributed, and it’s standard procedure in research to make sure your data are normal enough (making sure they don’t differ too much from a normal distribution) before treating them normally, but the data very well may not be perfectly Gaussian. However, the data would need to be super skewed left (implying some sort of immutable physiological mechanism that halted muscle growth past a certain, well-defined point) for the proposed “limit” of 25 to hold any water. Without a much larger data set to establish kurtosis and skewness, I think the best course of action is just to assume that FFMIs are normally distributed.
Ultimately, I think the entire notion of proposing a “limit” is wrongheaded. Unless there is some physical law that imposes an unavoidable constraint to further growth (I’m not aware of one), the Law of Truly Large Numbers kicks in when you’re trying to see what’s possible for the best of the best in a massive population; the incredibly-unlikely-but-not-impossible becomes a near certitude until you reach something truly impossible. In this case, we simply don’t know where “impossible” starts.
Instead of looking for a limit, we should be more concerned with likelihood. This leads us to the next question:
2) If there’s not a hard limit that tells you someone is definitely on drugs, is there a way to assign a likelihood to someone being drug-free?
Why, I’m glad you asked!
Let’s bring back this chart:
These are the odds that someone will fall within a given FFMI range with or without drugs. The higher the blue line is above the red line, the more likely someone with that FFMI is drug-free, and the higher the red line is above the blue line, the more likely it is that someone with that FFMI is on drugs. So, where the red line and the blue line intersect around 24, that means you have a roughly 50/50 chance that someone with that FFMI is drug-free.
However, we can’t leave it there. If we did, we’d be guilty of base rate neglect. The likelihood that someone is on steroids in an entire population will also influence the odds of someone with a given FFMI being on drugs or not.
Here’s a simple way to illustrate this: If 20% of people can reach an FFMI of 25 drug-free, and 60% of people can do it on drugs, then you’d assume that there was a 75% chance that someone with an FFMI of 25 was on drugs. However, if you’re dealing with a population of 4,000 lifters, 3,000 of whom are drug-free and 1,000 of whom are on steroids, then you’d expect 600 drug-free people to have an FFMI of 25 (3000*0.20), and 600 steroid users to have an FFMI of 25 (1000*0.6). In other words, a random person with an FFMI of 25 in that population would have a 50% chance of being drug-free, even though the odds are dramatically higher for any single steroid-user to reach an FFMI of 25. If you reverse the scenario – 3,000 users and 1,000 non-users – then you’d be looking at 1,800 users with an FFMI of 25 and only 200 non-users with an FFMI of 25, making it an 89% that someone with an FFMI of 25 was on drugs.
With that in mind, I made a handy dandy calculator incorporating the FFMI distributions of users and non-users and the percentage of a given population you think is truly drug-free. So, if you are curious about the status of a given bro in the gym, and you think 80% of gym-goers who actually populate the free-weight section are drug-free, you’d enter 0.8 in the first box. If you’re curious about the status of a certain natural bodybuilder who competes in drug-tested shows, and you think only 15% of bodybuilders at those shows are drug-free, you’d enter 0.15 in the first box (we don’t know those likelihoods, so just play around with that number to see how the predictions change). Then the form will do the rest of the calculations for you to tell you the likelihood that someone is drug-free based on their FFMI. If you think in percentages instead of decimal points, just shift the decimal point over 2 places (so 0.053 is the same as 5.3%).
3) How jacked can I personally get drug-free?
That’s a damn good question, and one that deserves its own article about the spread in genetic potential (which is coming eventually). However, it seems like about 2/3 of men wind up with an FFMI between 20.4 and 24.2. Use the calculator below to see what that range means for you, based on your height.
Keep in mind that there’s roughly a 2/3 chance you’ll wind up somewhere between the upper and lower range numbers. That still gives you a pretty decent shot of winding up bigger or (unfortunately) smaller.
4) How much of an advantage do drugs provide for hypertrophy?
The advantage is pretty massive.
The average untrained male has an FFMI of about 18.9.
Without drugs, the typical trained male winds up with an FFMI around 22.3, for a gain of 3.4 FFMI points. That’s about 9-13kg (~20-30lbs) of muscle, depending on height, gained over a training career.
With a reasonable degree of drug usage, the typical trained person winds up with an FFMI around 25.5, for a gain of 6.6 FFMI points. That’s about 20-24kg (~45-55lbs) of muscle, depending on height, gained over a training career.
In other words, there’re a roughly two-fold difference. That doesn’t mean that a user winds up with twice as much muscle; it means that users will typically wind up around twice as far from their starting point than nonusers.
With extreme usage, the gap gets dramatically larger. The top IFBB pros have FFMIs around 40, which is 21.1 points better than the average person. Of course, not everyone who does that amount of drugs will compete in the Mr. Olympia, but the top IFBB pros are about 6x further from the average person than the typical drug-free lifter is.
Naturally, the magnitude of that advantage that drugs provide will be larger or smaller based on the amount of drugs someone takes, their genetics, and how well they respond to drugs.
Wrapping it up
- Unfortunately, it’s impossible to know what the true drug-free limit is. In fact, trying to establish a limit isn’t the best course of action. Rather, these are questions that should be approached probabilistically.
- Over the course of a training career, it seems like people can build about twice as much muscle with steroids, relative to their starting point. The gap is even larger with very high doses.
- Dogs without myostatin are super jacked.