De-juicing PED Mythology

Speaking of steroids and baseball, there is a terrific and exhaustive new analytical site, called Steroids and Baseball, that aims to bust a little scientific method onto the widely held myths and claims about "performance enhancing" drugs and their effects on players and even The Children. The author, Eric Walker, is a longtime statistical analyst who used to work for Billy Beane's A's. His conclusion on steroids' actual effects on performance?

There is no evidence that steroid use has altered home-run hitting and those who argue otherwise are profoundly ignorant of the statistics of home runs, the physics of baseball, and of the physiological effects of steroids.

I whizzed too fast through Walker's "spliced power factor graph" to really understand it, but my less-mathematical forays into that study have produced similar conclusions. But what about the medical side effects?

[T]he risks of harm [PED]s present have been--whether from simple ignorance or from what could only be called political agendas--grossly exaggerated. The various potential consequences, individually or ensemble, are not grave, are not universal, and are almost all completely reversible.

Yes, but what about the children?

We first need to note that there is scarcely some runaway epidemic of usage. Current adolescent use rates for steroids are about 1.5% (bet you didn't know that) and dropping (bet you didn't know that, either). And those results are from multiple very large-scale scientific surveys. […]

Not that many kids have an athlete as a "role model", and those that do seem actually cleaner than the others. The small percentage of users are motivated by utterly other considerations: beefing up for the girls, scoring the winning touchdown, achieving a visible masculinity that their own minds will never let them find, or just raising hell all over the place owing to some broad-effect deep-seated maladjustment. […]

The idea that how we deal with professional ballplayers (or other adult athletes) could have anything to do with influencing kids to use or not use PEDs is puerile.

Those are just conclusions; it's the data leading Walker there that really turns your head. The whole site, complete with exhaustive links to various studies, is well worth a read for anyone seriously interested in understanding this amazingly obfuscated subject.

Comments

[David]

The small percentage of users are motivated by utterly other considerations: beefing up for the girls, scoring the winning touchdown, achieving a visible masculinity that their own minds will never let them find, or just raising hell all over the place owing to some broad-effect deep-seated maladjustment.

I'm stunned that kids have more varied reasons for trying steroids than try to emulate ballplayers!

Seriously, there's is no amount of legislation that's going to make teen boys avoid something that they think will make themselves more desirable to girls. Or help someone with a self-esteem problem think they look good enough that they don't need to use steroids.

[Episiarch]

But...but...using steroids isn't fair!!!

[Matt J]

Don't try to confuse the issue with facts.

[P Brooks]

I know next to nothing about baseball, and care less, but... Juicing the ball seems to be a much more effective way to increase home runs than juicing the hitters.

[John]

Of course Oakland, home of moneyball was we now know one of the big epicenters of steroid use in baseball. Further, one of the basic premises of moneyball is that you get get contact and high on base percentage hitters and as they develop the power will take care of itself. Gee how could that be? I think this guy because he was a part of the moneyball mafia out in Oakland has a definite motivation to diminish the effects of steroids.

As far as the statistcal effects of steroids, baseball is a statistically mature sport if there ever was one. Since the live ball erea began in 1920, we have had nearly 90 years of baseball, fifty of which was integrated. If it can be done in baseball, it pretty much has been done. Bonds and Mcquire took a homerun record that had stood around 60 for 60 years and increased it to 70 or 1/6. That is the equivelent of taking the 100 meter record that has hovered around 10.0 seconds for however many years and lowering it to about 8.5 seconds. Bullshit. Further, Bonds and Mcquire did it when they were both well passed the age of 30. That can't happen. Yes, player can do well into their late 30s because they get smarter and more skilled, but not because they get more athletic. It is as if Michael Jorden started jumping higher and running faster at age 35 and broke Wilt Chamberlain's 50 ppg record. It makes no sense in any other athletic context. Bonds was undoubtedly a great hitter before 1998. He was also a workout fanatic. It is not as if he were some fat lazy guy who all of the sudden discovered the gym. He was one of the best conditioned and hardest working players in baseball. Then, he go huge and almost doubled his homerun production. To say that doesn't involve steroids is to defy credulity.

[Brett]

Guess I'll have to dig through this, b/c I am beyond skeptical that PEDs did not have something to do with the jump in HRs of Mcgwire, Bonds, Sosa, even Brady Anderson, etc., especially those that did it at an advanced age as John mentions. But I'll give it a read as a juiced ball would certainly have had an effect.

[NP]

Matt, this is probably way more info than most of us need, but still it's a very thorough analysis. Pity your WaPo article doesn't carry a link to the page.

[John]

For the entire history of professional sports in this country, athletes peaked when they were in their late 20s or early 30s. That is when they have the experience of five or more years of top level competetition but still have their peak physical powers. Ruth had his monster statistical years in the early 20s when he was in his late 20s and was I believe 31 or so when he hit 60 homeruns and he never approached 60 again. Mickey Mantle was 25 or so when he won the tripple crown in 1956, one of the best offensive years in history, and 30 when he had his homerun chase with Marris in 61. His numbers went down after that. The list goes on and on until you get to the late 90s and Bonds and company who had their best years and added 10 or more pounds of lean muscle mass every year all in their late 30s. That is steroids. There is no other explanation.

One other thing, how does this guy explain Jason Giambi? He was a product of the Oakland moneyball system. He was when he first came up a skinny first baseman who had a great eye, walked a lot and a was a very good on base percentage guy. Then he goes to Balko gets on steroids and turns into a 45+ homerun guy and wins and MVP. After he got sick and Balko got busted he had to stop using roids, got skinny again and in the last couple of years with the Yankees is exactly what he was when he started, a light hitting first baseman who gets on base a lot. If steroids don't help wiht power, how do you explain Jason Giambi?

[Pain]

Actually when it comes to strength, athletes do not usually peak until their 30's. Just look at powerlifters. Most of the top ones are in their 30's. Not only does it take time to put on muscle, but it requires a solid understanding of one's own body to put it on right. Not something 25 year olds have as much knowledge in.

This is not to say that some of these guys didn't take steroids. But I don't believe it makes as much difference as people seem to attribute to them. I imagine better diets and strength training regimens are the main reason for the increases.

[bartman]

Steroids aren't harmful? Let's ask those East German shot-putters from the 70s. What, they all died in their 40s? What coincidence!!!

[BlueBook]

Don't know about baseball, but have you seen the training regimens for Olympic athletes these days? The uniforms are ergonomically designed and made from the latest composites, the diet is strictly controlled for nutritional content, exercise is precisely tailored to maximize performance, and every movement is analyzed by high-speed cameras and computers to find and eliminate mistakes. All of which is considered perfectly legitimate and sporting; but god forbid that an athlete take a synthesized version of a chemical that the human body produces naturally, and which has been used for decades.

[Virgil]

The 'roids and bball website uses circular reasoning to hammer the statistics into the shape it wants. First it postulates the notion that the ball is 'juiced' in certain years then it divides out the statistics from the 'juiced' years. Voila!, the upward trend is neatly eliminated.
Piffle.
Anyone looking at the performance of certain players before, during and after the years of using 'roids can see the difference.
Mark Twain's comment on 'Lies, damned lies, and statistics' is as true today as it was 150 years ago.

[R C Dean]

Let's ask those East German shot-putters from the 70s. What, they all died in their 40s?

Traditional web discourse calls for a link to support a factual assertion such as this.

[R C Dean]

Then he goes to Balko gets on steroids and turns into a 45+ homerun guy and wins and MVP.

Radley, it sounds like you've got some 'plaining to to do.

[stuartl]

Whatever your views on the legality of taking steroids, the argument that taking them does not increase physical performance is confusing. If they do not do any good, why are world class athletes taking them? Why risk losing your medals, or being banned from your sport?

I'm working so I also only have time to skim the articles, but it seems that the reasons given for taking steroids is to bulk up. Since hitting home runs requires strength (it is the big guys that are home run hitters) why are baseball players taking steroids if not to hit more home runs (or run faster, or throw harder)?

[Bob Smith]

If they do not do any good, why are world class athletes taking them?

Even if steroids didn't improve performance world class athletes would still take them. Why? Because anabolic steroids promote healing and recovery from intense workouts.

[R C Dean]

Because anabolic steroids promote healing and recovery from intense workouts.

Isn't this just the (main) reason why steroids increase performance?

Put another way, how can you heal and recover better from intense workouts without increasing your performance?

[stuartl]

Because anabolic steroids promote healing and recovery from intense workouts.

To further what R C Dean said, healing and recovery from an intense weight lifting workout builds muscle mass. AKA, strength. Useful for hitting things harder, etc.

FWIW, the "anabolic" in anabolic steroids refers to building up of tissues, in particular bone or muscle mass.

[stuartl]

Out of curiosity, I took a look at Eric Walker's steroids and baseball page. The first two claims he makes are weak at best, intentional lies at worst. Baseball statisticians love to argue this stuff, I just was curious about the weird claim that steroids don't help hitting home runs.

Walker immediately quits the home run statistic, he switches to a power factor (total bases per hit). While this could be fine, the fact that he doesn't correlate his numbers back to home runs is somewhat suspicious.

More importantly, he looks at a chart of year to year power factors, does eyeball (only!) statistics, finds 3 splices (which he attributes to juicing the ball), and does a set of adjustments based only on the splices. He ignores other equally obvious spikes/data anomalies: the spike up in 76 is real, but the one down in 82 is not? He ignores a spike up in 83 and back down in 88. He ignores a negative plateau from 88-92. I'm not a professional statistician, but this kind of eyeball analysis of noisy data is crap. He uses his 3 spikes (up only, none down) to show a small negative trend in power. Frankly, the downward trend is so small that given the way he manipulates the data, adjusting only for spikes up, suggests to me that power is going significantly up, not down.

His second argument is that the medical effects of steroids would not increase power. Here he is clearly either lying or only reading what he wants to read. He references several articles saying that the effect of steroids is more on upper body strength than lower body. He then correlates this with a baseball physicist, Adair, who says:

The considerable energy . . . transferred to the bat . . . is generated largely by the large muscles of the thighs and torso. The arms and hands serve mainly to transfer the energy of the body's rotational and transverse motions to the bat and add little extra energy to the bat. [Footnoted with:] In particular, the contribution of the hands and wrists to the energy of the bat is almost negligible.

But a careful reading of the actual articles and the actual quotes he uses leads to the exact opposite conclusion. He points to the following statement from Cynthia Kuhn of Duke:

Testosterone increases upper body mass differentially, so performance in tasks like weightlifting should improve more than lower-body tasks or tasks in which aerobic capacity rather than strength are assessed. As expected, the task in which increases have been reported most reliably are in the bench press (Friedl, 2000). Finally, the degree of improvement expected in such studies is generally small. Changes in performance of 1-5% are rarely statistically or clinically significant but they represent the margin of victory for elite athletes.

Several things about this quote, hitting a baseball is all anaerobic, whether you are talking legs, torso, or arms. Second she is not saying that that the lower body is not helped, just that it is not helped as much. Most importantly Walker has done some slight of hand here, Adair's quote says power comes from the torso and legs, not so much from the arms. The torso includes much of the upper body, chest, abs, back, .... In fact the one muscle Kuhn points out as gaining mass is the trapezius, part of the upper back. Kuhn's quote ends by saying saying that even though the effect is small, a small improvement does matter for world class athletes.

There may be more to Walker's claim, but after two fatally flawed arguments, I'm not going any further.

[daksya]

I'm just surprised that H&R approvingly links to someone who treats DEA & NIDA as credible.

[Eric Walker]

Let me address briefly a few things said here.

"I think this guy because he was a part of the moneyball mafia out in Oakland has a definite motivation to diminish the effects of steroids."

Aside from the fact that I left my consultancy over 10 years ago when I retired to rural Washington State, I find the remark one that could only be made on the internet, rather than face-to-face. That is as polite as I know how to say it.

"Further, Bonds and Mcquire[sic] did it when they were both well passed [sic] the age of 30. That can't happen."

Peak career home-run seasons: Hank Sauer, age 37; Hank Aaron, age 37; Rico Carty, age 38; Carlton Fisk, age 37; Gary Gaetti, age 37; Chili Davis, age 37; Edgar Martinez, age 37; and of course, Rafael Palmeiro and Barry Bonds, age 37. And that does not address men with high numbers not quite their career best at late ages. "Can't happen"?

" . . . added 10 or more pounds of lean muscle mass every year"--making them, what, 330 pounds at age 40?

"First it postulates the notion that the ball is 'juiced' in certain years then it divides out the statistics from the 'juiced' years. Voila!, the upward trend is neatly eliminated."

Nothing is "postulated". The evidence is right out in the open. Do we argue that the 1977 Rawlings ball was identical to the 1976 Spalding ball? Piffle. And the 1993 juicing is documented by not one but two independent laboratory examinations of the balls themselves, as is thoroughly documented on the site.

"If they do not do any good, why are world class athletes taking them?"

They do not do any material good to *baseball players*. As the site carefully notes, in other activities, where a hundredth of a second over seveal miles can make the difference, they probably do help--and especially activities where upper-body strength is critical. But there is also the reality that athletes often do not really know even what they're putting in their bodies, much less what exactly it might do for them.

". . . anabolic steroids promote healing . . ."

I guess "everybody knows so"--everybody except those in the medical profession. There is an entire page of the site on purported "healing effects": you might want to examine it.

"Walker immediately quits the home run statistic, he switches to a power factor (total bases per hit). While this could be fine, the fact that he doesn't correlate his numbers back to home runs is somewhat suspicious."

Not if you know baseball. Since apparently a lot of people willing to talk about it don't, I will later tonight be putting up a graph of PF versus simple HR/H, to show that they are as close to identical as doesn't matter--which we expect, as HRs dominate TBs.

". . . does eyeball (only!) statistics . . ."

Wherever did you get that idea? While the disjuncts are immediately obvious to the eye--which is why I used graphs to demonstrate the principles--the graph is not the *source* of the analysis; our knowledge of baseball history is.

Finally, the quotation from Ms. Kuhn seems simple, but I guess it's not, so let me emphasize the key word:

"Testosterone increases upper body mass differentially, so performance in tasks like weightlifting should improve more than lower-body tasks OR tasks in which aerobic capacity rather than strength are assessed." That is, upper-body tasks are improved more than:
a. lower-body tasks OR
b. aerobic tasks.

OK?