Skip to content

Players busted for PEDs don’t show any notable uptick in performance

May 14, 2014, 2:35 PM EDT

Toronto Blue Jays' Cabrera heads to the batters box during the first inning of their MLB baseball game against the San Francisco Giants in San Francisco, California Reuters

A week after ESPN’s Buster Olney decided that Melky Cabrera‘s excellent-so-far 2014 season justifies people assuming he’s back on PEDs, ESPN’s Dan Szymborski dives into the numbers and shows that such a decision is not based on any evidence at all:

Despite the rhetoric surrounding PEDs, players caught for steroid/testosterone use do not show a pattern of overperforming their projections in the years leading up to the drug suspension or a pattern of underperforming their projections in the years after a drug suspension.

All of this is based on Szymborski’s ZiPS projection system, which shows that Cabrera is doing about what you’d expect him to do this year. And that, as a group, guys busted for PEDs don’t really deviate too much from their expected performance one way or another before, during or after they are caught for PEDs.

Obviously there are a lot of caveats in play here. Small sample sizes, imperfect data about when guys start and stop taking PEDs and, of course, the flaws any projection system, even one as generally reliable as ZiPS, brings to the table.

But if you’re making a case for something — and Olney and others who are questioning the legitimacy of Melky’s performance this season are clearly making a case for something — it’s incumbent upon you to present some evidence. Szymborski’s analysis doesn’t necessarily prove anything about the efficacy or lack thereof of PEDs. But it has far more evidence on its side than anything people are hurling at Cabrera lately.

  1. sportsfan18 - May 14, 2014 at 2:39 PM

    Can’t let facts get in the way of what people THINK now…!

    • Bob Loblaw - May 14, 2014 at 2:49 PM

      Wait, “Szymborski’s ZiPS projection system” is fact now? Holy mother of sabermetrics!!!!!

      • Craig Calcaterra - May 14, 2014 at 2:55 PM

        Yeah, go ahead and ignore all the stuff I — and Dan, for that matter — has said about what projections systems are and are not, along with any of the caveats that go along with them.

      • Bob Loblaw - May 14, 2014 at 2:58 PM

        Craig, did you see the item I responded to…the one where the author wrote about “facts getting in the way”? My post wasn’t a response to your article. It was a response to that comment alone.

      • paperlions - May 14, 2014 at 3:32 PM

        Didn’t read the article, did you?

        The point, is that player production (real data, also know as facts) doesn’t deviate from expectation any more for players busted for PEDs than they do for players that are not busted for PEDs. Which, again, is a fact.

        Szymborski isn’t the only person to look for a signal of steroid use and fail to find one. Everyone that has looked has failed to find one, which doesn’t mean they don’t have an effect, but it does mean that if they do…the effect is pretty small. The large effect that people act like PEDs have would be so easy to detect so as to not even require rigorous analysis.

      • term3186 - May 14, 2014 at 3:38 PM

        Paper, why SHOULD the data deviate from expectation? If I’ve been
        taking PEDS my entire career, the expectation is BASED on my PED usage. If I continue taking PEDs after I get suspended, my numbers SHOULD match the previous PED-based expectations!

      • paperlions - May 14, 2014 at 3:54 PM

        Yeah, they would…if you keep taking steroids and manage to not be busted again while being tested far more often.

        Szymborksi also looked for deviations before and after testing began for players in the Mitchell report that we know were using. Just looking at physiques, it is clear that were way fewer muscle bound players after testing began. HR rates didn’t go down when players shrank and steroid testing began, and there was no signal in that data either.

        This isn’t the first time someone has tried to find a signal associated with steroids and could find it. It is just another try that failed.

  2. jbriggs81 - May 14, 2014 at 2:57 PM

    I don’t understand how the Zips projections could be accurate if you don’t know when a player began taking PED’s and stopped taking PED’s. If those projections are based upon PED numbers, how would they be accurate projections?

    • paperlions - May 14, 2014 at 3:37 PM

      That is the point, the projections wouldn’t be as accurate for PED users that are busted as they are for other players. Thereby testing the assumption of a PED effect.

      If a player is using steroids, and steroids affect performance, and after being busted a player stops taking steroids, then that player would be expected to under perform his projections (which would be “steroid enhanced”) after being busted more often than a non-using player during the same time frame. Of course, there are many assumptions, that players top using, that players that are not busted are not using, etc. But if the assumptions hold for the majority of cases, there should be a noticeable difference in the deviations between expectations and performance of the two groups.

      • davidpom50 - May 14, 2014 at 5:06 PM

        But there’s no evidence to suggest that players stop using steroids. There’s no knowledge of when they started, or how much or what substances they used (many positive tests are for masking agents, not PEDs themselves). Szymboski doesn’t have nearly enough meaningful evidence to draw any conclusions, and states as much in his article. His research doesn’t prove that steroids increase statistical performance, but it certainly doesn’t DISPROVE it either. It tells us exactly nothing.

      • paperlions - May 14, 2014 at 5:09 PM

        All true, but then, it is an analysis, analyses don’t prove anything, they simply provide support for an idea or they don’t. These don’t provide any support for the idea that steroids have an effect. Importantly, all players don’t have to quit for an effect to manifest just like all players assumed to be clean don’t have to be clean. Something that is also true, the next piece of evidence that steroids do increase offensive production in baseball will be the first evidence of that effect.

      • davidpom50 - May 14, 2014 at 5:34 PM

        “These don’t provide any support for the idea that steroids have an effect.”

        It’s very, very, VERY important to realize that this statement is not at all the same as saying the study provides support for the idea that steroids have no effect. The study proves nothing, the study disproves nothing. The study had extremely flawed data inputs. This is all we know about the study.

      • paperlions - May 14, 2014 at 5:53 PM

        It is.

        But this is just one more in a long line of studies that have failed to find such a signal….studies of individual players as well as studies of league-wide performance.

        In short, every time someone has gone looking for an effect of steroids on offense in baseball, they couldn’t find it…even though most assumed it would be there when they started.

      • davidpom50 - May 14, 2014 at 5:56 PM

        Again, as Term has been saying over and over again… NONE of those studies had any worthwhile data. Not one. We can’t know exactly who used and exactly when they used, and therefore we can’t know if using had an effect.

      • bh192012 - May 14, 2014 at 6:31 PM

        I just did a study on if scoring more is relative to winning. I did this by looking into my pocket. I didn’t see anything. The evidence just doesn’t support the idea that scoring more helps a team win.

  3. sabathiawouldbegoodattheeighthtoo - May 14, 2014 at 3:03 PM

    PEDs seem to be a conditioning shortcut. They don’t let a guy do anything he could not have done otherwise, they simply shorten the time frame. They let guys lose fat faster; build muscle faster; heal from injury faster; recover from fatigue faster; etc etc.

    Perhaps some of these guys went into “hard work” mode, or perhaps they are all n drugs again, or maybe the drugs didn’t really do any good in the first place. Who knows (except Olney, of course).

    BTW, have any Boras Corp clients made their way into Olney’s cross-hairs? I can’t imagine he would be eager to bite the hand that feeds him, but I am not sure about who is represented by whom.

    • schmedley69 - May 14, 2014 at 5:05 PM

      Not to mention die faster.

  4. term3186 - May 14, 2014 at 3:14 PM

    Yeah…. about those ‘caveats’. I think a more accurate term here would be “fatal methodological flaws”. Without knowing exact PED types and dates taken, how could any of this possibly be worth anything? How can you project “Player X without PEDs” when you don’t (and can’t) know what his performance would be like with or without PEDs, since you can’t know when Player X was taking (or not taking) PEDs? There are some fatal assumptions: Maybe players aren’t outperforming projections because those projections were based on players taking PEDs at the time! Maybe they don’t underperform because they are still taking PEDs of some sort,
    hoping to show people that they still have “it”.

    • dluxxx - May 14, 2014 at 3:19 PM

      That’s a lot of caveats….er… maybes there.

      • term3186 - May 14, 2014 at 3:28 PM

        No kidding. This report doesn’t seem worth the paper it’s printed on. It’s certainly not “compelling”. It doesn’t even get to the level of “mildly interesting” if looked at with a critical eye. Though Craig does love the “PEDs don’t help performance” narrative, so I suppose I’m not terribly surprised that the problems are pretty much glossed over.

      • dluxxx - May 14, 2014 at 3:38 PM

        I was pointing out all of your maybes as being a lot. I can look at the report, and see that a players numbers (something concrete and measurable) pan out pretty much exactly where they would be expected to. I don’t have to do any mental backflips to convince myself that someone might still be cheating because they want to prove it.

        In fact, I would guess that a player who has been suspended would become less likely to cheat, especially with the increased scrutiny. Common sense says that would happen, and Occam’s Razor would support that. Essentially, in the essence of uncertainty, go with the fewer assumptions made, the better.

    • paperlions - May 14, 2014 at 3:49 PM

      There was no attempt to project players without PEDs.

      All you do is project player performance. If the steroids had an effect, and guys stop taking them after being busted, then their actual performances should fall below the projections more often than guys that were not busted for steroids (which assumes that those guys were not using). All it is, is a comparison of expectations to performance. If steroid use affected performance, then projections would be artificially inflated, and the same player clean should under perform (on average) more often than players that were clean the entire time.

      • term3186 - May 14, 2014 at 4:02 PM

        But that assumes they stop. Which is not the small assumption that it’s being made out to be. If we’ve learned anything from cycling and Bonds, Clemens and the WADA, it’s that risk of being caught is relatively low. Especially since in the arms race between dopers and catchers, the dopers are almost always one step ahead.

      • paperlions - May 14, 2014 at 4:06 PM

        I agree.

        But, using those examples, we should just assume that everyone is using because it is hard to get caught and the other players are just better at cheating than the guys that got busted.

      • paperlions - May 14, 2014 at 4:10 PM

        To be clear, the assumption that players keep using after being busted is no more valid that the assumption that they stop using.

        Again, in the study, he also used earlier data associated with Mitchell report players. We know that steroid use is less common now as players show less physical evidence of using (i.e. players are much smaller now than they were pre-testing). There wasn’t any signal associated with that data either.

      • dluxxx - May 14, 2014 at 4:18 PM

        It’s obvious paper, EVERY baseball player is using steroids. Only some of them get caught. That’s why the numbers don’t lie!

      • paperlions - May 14, 2014 at 4:20 PM

        Indeed.

        I really feel for the children.

      • dluxxx - May 14, 2014 at 4:29 PM

        Awww, you can’t pull the “Think of the Children!” card already. I was going to, but figured I’d hold back for a little longer…

      • dluxxx - May 14, 2014 at 4:29 PM

        And honestly, I was going to use it on the “Let’s just make EVERYTHING legal” comment…

      • term3186 - May 14, 2014 at 4:26 PM

        I wouldn’t go so far as to say that the assumption they keep using is “no more valid”. We KNOW of a number of players who have continued to use after being caught (and we know because they were caught again). But my larger point is this: We don’t know they stopped. We can’t know they stopped. Therefore the study is fatally flawed, and can’t be compelling/relied upon/used to make a point.

      • paperlions - May 14, 2014 at 4:34 PM

        What study would convince you of anything?

        Even league-wide analyses show no association between steroid use and offensive production. HR rate increased almost over night from 1 per 45 PA to 1 per 30 PA in 1993. HR rates remained at that level for 15 years, the last 5 of which were after PED testing started. Did everyone start using at the same time? Did the proportion of players using stat constant from 1993 through 2008? Did players suddenly stop using in 2009 when HR rates started to go back down? Essentially, none of the data are consistent with any kind of expectation if steroids have a measurable affect on player performance.

      • dluxxx - May 14, 2014 at 4:35 PM

        Well, one can reasonably assume they stopped unless they get caught again. I mean, to assume that all players who have used have continued to use, and no player that hasn’t been caught has never used is pretty asinine. We know that’s not true because new people get caught all the time. If we can’t tell if someone is cheating by the current testing that they’re using, then why even test? Why bother at all?

        You may as well believe that every player is using and only some get caught. There’s no more proof of that then saying that players that have been caught have continued to use.

      • term3186 - May 14, 2014 at 4:48 PM

        Finally! Something resembling agreement with both of you! As to HR production and whatnot paper, I can’t comment intelligently, I don’t know the numbers and don’t really have the time right now to research them (aside to say maybe PEDs helped pitchers too?). Do PEDs actually help performance? I have no idea. The only thing I’m saying is that THIS STUDY doesn’t tell us.

        dluxx, we’re saying similar things, but in opposite ways. I’m not “assuming players kept using” rather, I’m NOT assuming that they stopped. Which is what the study assumes. That’s all I’m saying. Don’t make the assumption they stopped (especially when the opposite has proven true in a number of cases).

        As for the ‘why test’ question… a full articulated response is more than I’d really care to write, and is an entirely separate discussion, so the only thing i have to say to that is Nirvana Fallacy.

      • bh192012 - May 14, 2014 at 6:41 PM

        The beauty of it is, depending on how and why they took PEDS (to keep from slipping?) they might actually CAUSE ZIPS to be MORE accurate.

  5. stlouis1baseball - May 14, 2014 at 3:38 PM

    Well this settles it. I am now convinced. Legalize all PED’s. Hell…legalize everything.
    I want pitchers dosing acid (ala Dock Ellis) and I want those not dosing lsd to start doing lines (ala Doc Gooden, Daryl Strawberry and Steve Howe). Down with natural ability! Up with performanced “enhanced” substance abuse.

    Signed,
    The Ostrich

  6. paint771 - May 14, 2014 at 3:44 PM

    For all the people sniping Szymborski’s back-of-the-envelope take on it, I’m curious: what’s your method for determining that PEDs do in fact have a statistically significant impact on performance? Now that we’re all sudden sticklers for methodology, I’m kind of interested in hearing it.

    This is a bit of shadow boxing, in that those that hold that PEDs may not have a particularly significant impact on performance are suddenly held to standards of exacting scientific rigor, while the “boo PEDs” folks seem to be able to rely on “you know – big necks!”

    • term3186 - May 14, 2014 at 4:11 PM

      It’s impossible. To know the exact impact of PEDs on performance you would know to know, at the very least, which players, what PEDs, and what dates they were taken. For hundreds, if not thousands of players. This is the only way to play the science/numbers game. Absent this kind of information (which I feel comfortable saying that no one will get) there is no way to scientifically measure the impact of PEDs. None. Which is why I find Craig’s appeal to “science” and “the numbers” so irritating.

      • [citation needed] fka COPO - May 14, 2014 at 4:42 PM

        To know the exact impact of PEDs on performance you would know to know, at the very least, which players, what PEDs, and what dates they were taken.

        No you don’t. You didn’t read the article did you, because Dan’s methodology is explained precisely on how he came about his numbers. Also you don’t need to know what the player took, because it’s irrelevant. Unless you think some PEDs are better than others…

      • term3186 - May 14, 2014 at 5:01 PM

        I think it’s very possible that stimulants have different effects than steroids. I’m not a doctor or a biochemist so I won’t pretend to know. And my entire point is that the methodology is flawed. Because you don’t know the things I mentioned. So you have to make assumptions. And facts based on assumptions (especially assumptions that have been shown to be incorrect) aren’t facts at all.

      • [citation needed] fka COPO - May 14, 2014 at 5:06 PM

        I think it’s very possible that stimulants have different effects than steroids.

        You should have just admitted you didn’t read the article:

        To test this, I took the 27 major league hitters who were suspended for non-amphetamine PED use since 2004 and compared their ZiPS projected OPS+ to their actual OPS+ (including only stints of at least 100 plate appearances in the majors).

        Because you don’t know the things I mentioned. So you have to make assumptions.

        Please go read the article, read his methodology and then come back.

      • term3186 - May 14, 2014 at 5:13 PM

        Facepalm. The comment I was replying to was about “method for determining that PEDs do in fact have a statistically significant impact on performance”. Not this study in particular. Generally speaking, when attempting to measure the impact of PEDs on performance, they should be separated by class. (I mean, ideally they’d be separated by individual drug, but that is probably a bridge further than is actually necessary, class ought to sufffice).

      • [citation needed] fka COPO - May 14, 2014 at 5:22 PM

        So you’re trying to refute a study that hasn’t happened?

      • term3186 - May 14, 2014 at 5:43 PM

        No no no. paint771 asked for a scientifically sound way of determining PED impact on performance. To do that, you would look at the numbers after learning: What players took PEDs. When they took them. What types of PEDs they took. (this has the added bonus of comparing the benefits – if any – of different types of PEDS!). There might be several other questions that need answered, my reply was just off the top of my head. But for any rigorous analysis you certainly NEED the first two, and in large enough numbers to overcome sample size issues.

      • [citation needed] fka COPO - May 14, 2014 at 5:48 PM

        Ok, now I understand and that makes more sense.

  7. paperlions - May 14, 2014 at 3:56 PM

    I see that most of the people here must represent the portion of the population that still believes the sun rotates around the earth (shockingly, still 25% of the US population).

    • unclemosesgreen - May 14, 2014 at 4:50 PM

      You expected more from the deadliest parasite ever known to the planet?

      • paperlions - May 14, 2014 at 5:02 PM

        Nah, just noting to correspondence between expectation and performance.

  8. grumpyoleman - May 14, 2014 at 4:17 PM

    I project you could give steroids to many of the people on here and they still couldn’t hit a ball off of a tee.

  9. Detroit Michael - May 14, 2014 at 5:07 PM

    Regarding Melky Cabrera specifically, he should perform better than his ZiPS projection assuming that the projection system just saw his 2013 performance and doesn’t know that he was playing with a major tumor.

  10. davidpom50 - May 14, 2014 at 5:21 PM

    I think the point Term is trying to make (and term, please correct me if I’m wrong) is that without a rigorous double-blind study of PED use on MLB caliber baseball players, we’re guessing. Szymborski is guessing in this study. There is no science here, and to call it such is silly.

    Obviously we cannot measure exactly the results of PED use. We can know the following, though:
    1. PEDs increase strength in human beings.
    2. There is a strength component to both hitting and throwing a baseball.

    These are facts. You cannot deny these facts.

    From that, we can deduce that PED use, in the absence of all other variables, will give any baseball player an advantage in hitting and throwing over his baseline talent. As Paper is so fond of pointing out, there is not an absence of other variables. Many of these variables could, in theory, make a player WORSE (for example, a player who changes his swing to attempt take greater advantage of his added strength, only his new swing is much worse) The bottom line is that it is ridiculous to claim PEDs cannot and do not give any player an advantage, and it is equally ridiculous to claim that PEDs will turn every player into Barry Bonds or Roger Clemens, or even that PEDs will definitely make any player better.

    • [citation needed] fka COPO - May 14, 2014 at 5:25 PM

      Obviously we cannot measure exactly the results of PED use. We can know the following, though:
      1. PEDs increase strength in human beings.
      2. There is a strength component to both hitting and throwing a baseball.

      These are facts. You cannot deny these facts…..Many of these variables could, in theory, make a player WORSE

      The variable doesn’t make the player worse, the player changing something would be worse. And channeling Dan here, wouldn’t you assume that if Player + PED = X performance, than Player – PED = X-n performance?

      If you assume the player would perform worse without them, and the player doesn’t, maybe the player didn’t get a benefit from the PED(s) to begin with?

      • davidpom50 - May 14, 2014 at 5:30 PM

        No, I wouldn’t assume that at all. There are far too many variables in play, and far too much faulty data in play for Szymborski’s methodology to come close to a worthwhile result. Maybe it’s YOU that needs to re-read his article… as he pointed out to me on Twitter yesterday, he’s very clear that his study doesn’t prove or disprove anything (which is why I question why he even bothered to write it up)

      • [citation needed] fka COPO - May 14, 2014 at 5:44 PM

        No, I wouldn’t assume that at all. There are far too many variables in play, and far too much faulty data in play for Szymborski’s methodology to come close to a worthwhile result. Maybe it’s YOU that needs to re-read his article

        So what would you assume, that we just don’t know?

      • davidpom50 - May 14, 2014 at 5:48 PM

        I guess my only assumption is that there is a strength component to successfully hitting or pitching a baseball. I KNOW steroids have a large, measurable impact on human strength. I KNOW that without a rigorous study, we can’t pretend to have any idea of the overall effects of PEDs on a large sample of baseball players.

    • term3186 - May 14, 2014 at 5:45 PM

      I agree with your first paragraph. As for PED use in general, I haven’t done a boatload of research and don’t know and don’t claim to know. All I’m saying is that this study is so methodologically flawed as to not tell us anything.

      • davidpom50 - May 14, 2014 at 5:51 PM

        I’m sorry, I meant to make clear that from the second paragraph on was my own thoughts. Didn’t mean to put words in your mouth!

  11. chip56 - May 14, 2014 at 5:24 PM

    I may not be a statistician but I can read.

    Melky Cabrera as a Yankee playing in hitter friendly Yankee Stadium hit around .270 with a slugging percentage in the .390 range. With Atlanta he hit .255 with a .354 slugging.

    After being non-tendered and signing with Kansas City his batting average jumped 50 points and his slugging percentage jumped 120 points. The following year his BA jumped another 40 points and his slugging percentage did as well.

    So I’m not sure exactly what this guy is looking at, going from fringe major leaguer (which is what he was in Atlanta) to All Star the following year to one of the top outfielders in the game the year after certainly seems like a performance spike.

    • [citation needed] fka COPO - May 14, 2014 at 5:33 PM

      So I’m not sure exactly what this guy is looking at, going from fringe major leaguer (which is what he was in Atlanta) to All Star the following year to one of the top outfielders in the game the year after certainly seems like a performance spike.

      Cabrera
      2011 (dirty?): .305/.339/.470, .164 ISO, .350 wOBA, 118 wRC+
      2012 (busted): .346/.390/.516, .170 ISO, .387 wOBA, 150 wRC+
      2014 (clean): .329/.365/.515, .186 ISO, .384 wOBA, 142 wRC+

      Granted there’s a GIANT SSS issue with his ’14 season, and I really wish Dan didn’t use Melky, but his ’12 and ’14 numbers seem awfully close.

      • term3186 - May 14, 2014 at 5:37 PM

        This is the problem:

        2014 (clean): .329/.365/.515, .186 ISO, .384 wOBA, 142 wRC+

        What in the world gives validity to your/Dan’s assumption that Melky is currently clean?

      • [citation needed] fka COPO - May 14, 2014 at 5:46 PM

        Apologies, meant to put an asterisk and note “presumably”. And call me naive, but I just don’t think as many people are using as some think. Do you think all 27 people that Dan studied are still using or were?

      • term3186 - May 14, 2014 at 5:52 PM

        No! People seem to want to say that I’m assuming people are using. This is false. However I am NOT assuming they’ve STOPPED using (THIS IS NOT THE SAME THING!!!!!!). Dan has assumed they have. Why he makes that assumption, I’m not sure, since a number of players have tested positive twice. This critical assumption (especially combined with the number of other problems with this study – sample size, projection formula, etc.) makes this study effectively useless, and not worth the paper it’s printed on. It doesn’t tell us anything useful.

      • chip56 - May 14, 2014 at 11:45 PM

        And Buster’s point on Melky was – are we sure he’s clean? Given that the only other time he’s put up these numbers was when he was dirty/busted versus where he was the rest of his career – that’s the benefit of the doubt that Olney (and some others) are not willing to give Melky. Their assumption is that he’s just found a better way of cheating.

      • chip56 - May 14, 2014 at 11:56 PM

        This is the point that Buster was trying to make a few weeks ago (and Craig took him to task for it). If you look at a player like Melky who has had such a drastic career arc that was tainted with a steroid bust then comes back and puts up the same numbers he was putting up when he was busted vs. the numbers he put up throughout the early part of his career then it’s completely fair for people to question whether he’s dirty or not.

    • paperlions - May 14, 2014 at 5:57 PM

      If you look at Melky’s career. Much of the variation in is BABIP (which affects BA, OBP, SLG…everything). His K rate is the same, his BB rate is the same, his power mostly changes with his home park. Some years he has just hit it where they ain’t a LOT more than other years. Steroids may do a lot of things for people, but they don’t help to hit the baseball away from fielders…..do they?

      • davidpom50 - May 14, 2014 at 6:00 PM

        In theory… yes, they do. Balls with a higher exit velocity off the bat have a higher BABIP. Swinging harder will produce a higher exit velocity. http://www.baseballprospectus.com/article.php?articleid=15562

      • paperlions - May 14, 2014 at 6:03 PM

        You want to look up Melky’s batted ball velocity for me?

        In theory, if this was true, then the guys that hit for the most power would have the highest BABIP as well, wouldn’t they? Because they are always hitting the ball very hard…..how’s that check out?

      • paperlions - May 14, 2014 at 6:03 PM

        Importantly, steroids primarily affect upper body strength. Hitting a baseball with authority is primarily based on lower body strength, hip rotation, and balance.

      • davidpom50 - May 14, 2014 at 6:02 PM

        Whoops, wrong link. Here’s the right one: http://www.baseballprospectus.com/article.php?articleid=15532

      • davidpom50 - May 14, 2014 at 6:09 PM

        “Importantly, steroids primarily affect upper body strength.”

        True, but the effect of PEDS on lower-body strength is still significant and measurable, and upper body strength still plays a role in hitting and throwing.

      • paperlions - May 14, 2014 at 6:12 PM

        From all of this, I think what you are saying is that you hope the Cardinals play well and Wacha throws a gem tonight versus the Cubs.
        :-)

        Time to go do some yard work.

      • davidpom50 - May 14, 2014 at 6:12 PM

        “You want to look up Melky’s batted ball velocity for me?”

        I wish I could. I don’t know where Mike Fast got his data.

        In theory, if this was true, then the guys that hit for the most power would have the highest BABIP as well, wouldn’t they? Because they are always hitting the ball very hard…..how’s that check out?”

        It… checks out very well. Read the study I linked the second time :)

      • davidpom50 - May 14, 2014 at 6:19 PM

        Good God no. I never want the Cardinals to do well, and I have completely irrational hatred for Wacha because some idiot once tried to tell me he’s better than Clayton Kershaw based on one playoff game.

      • paperlions - May 14, 2014 at 7:43 PM

        No no….all the data suggests that you secretly love the Cardinals and have a huge man crush on Wacha…despite the fact that he obviously is no where nearly as good as Kershaw….don’t deny it.

      • chip56 - May 14, 2014 at 11:51 PM

        I’m not sure that you can make that assumption with his power numbers if for no other reason than that for the first part of his career he was playing in a hitters’ haven in NYC.

        As for BABIP they kinda could help if you buy into the theory that you’re stronger and faster it means that lazy fly balls can now bang off the wall or that if you’re quicker with the bat you’re going to square up a ball more than if you’re behind it.

        Melky’s talented, clearly – and he’s at an age where it’s entirely possible that he’s just grown as a player through maturity. That said, I can’t fault someone who looks at Melky’s career arc and says “yeah, guy did roids and they saved his career.”

      • paperlions - May 15, 2014 at 7:10 AM

        The thing is…there are a LOT of player’s whose BABIP numbers vary like Melky’s have varied. Austin Jackson, for example….has a similar up and down pattern. BABIP for hitters can vary a lot between seasons, and it does so for a lot of hitters. Michael Young was a BABIP dependent hitter, when it was high, he was good, when it was low, he was not…and it bounced up and down throughout his career.

        You can’t assume that a pattern is steroid related when it happens with a lot of players for which there is no evidence of steroid use and when it is known that there is a lot of variation due to luck. It would be one thing if there was any evidence that steroids make guys better hitters, but, again, every time someone has used the data to try to find out if that is true or not, they don’t find any evidence to support it.

      • chip56 - May 15, 2014 at 9:21 AM

        Paperlions,

        The problem with your comps of Melky to Austin Jackson or Michael Young is consistency. Michael Young consistently hit between .275 and .330 throughout his career. On the other hand Melky as a Yankee and Brave was around .250 – .270 and then jumped to all of a sudden becoming a .300 hitter and then a .350 hitter.

        I think saying he’s just hitting with better luck now than he did for the first half of his career is as silly an argument as those who believe that steroids help you hit a curveball.

        For his sake I do hope he’s clean and I hope he goes on to play like this for the rest of his career – however I can’t find fault with those who question whether or not he’s cheating because of how drastic the last 4 years of his career have been from the first 6 coupled with the fact that the last time he put up these numbers he was using.

      • paperlions - May 15, 2014 at 7:54 AM

        A couple of other things to consider. Old Yankee stadium was not nearly the HR park the new one is. Melky only played for the Yankees 1 year in that park and he set a career high in HRs that year.

        I couldn’t find batted ball velocity, but I did find batted ball distances. Cabrera’s average batted ball distance starting in 2007:271, 275, 283 (new YS), 266, 281, 269, 270, 265.

        So, in his best years, Cabrera’s batted ball distance include his 2nd highest average distance and his career low in average distance (this year). He’s didn’t hit the ball farther in years he used steroids or in years he suspected of using steroids than he did in years when no one suspected him because he was bad at baseball. He’s been consistently hitting more line drives and fewer FBs and IFBs, the last few years, which probably help with BABIP…but there’s no evidence that he’s hitting the ball farther.

      • chip56 - May 15, 2014 at 9:29 AM

        Old Yankee Stadium was still a HR haven for left handed hitters and, as a switch hitter Melky did get that benefit.

        As to line drive rates, again, if you’re using something that can help you get stronger and faster then you’re going to be quicker through the zone and what were lazy fly balls or grounders that you just missed will turn into line drives.

        Back to the original point of the post though – you can’t look at Melky’s career arc as evidence to suggest that steroid use doesn’t improve performance.

      • chip56 - May 15, 2014 at 9:31 AM

        He would have been better off using the career stats of players who stink who were caught using. For example Jesus Montero or Fernando Martinez clearly saw no improvement in their games through PED usage.

      • paperlions - May 15, 2014 at 9:49 AM

        He, meaning Symborski? He used EVERY player that has been caught cheating, not just Melky. He also used EVERY player named in the Mitchell Report.

        It isn’t like these findings are controversial or surprising because, again, every attempt so far has failed to find a signal associated with steroid use…even when potential bias existed because people assumed that steroids had a big effect.

      • chip56 - May 15, 2014 at 10:01 AM

        Sorry for the confusion, I meant Craig when I said “he.”

        As I said though, suggesting that Melky is just hitting into better luck now than he did with the Yankees or Atlanta seems like a fairly big reach to explain the explosion of his performance.

      • chip56 - May 15, 2014 at 10:06 AM

        The problem with Symborski’s article is that he’s assuming that the only times the players he lists cheated were the years they got caught and that goes to the argument Buster made about how do we know that the only tainted numbers were from the year they got caught? How do we know Melky or Grandal or Cruz aren’t cheating now with a different substance that just hasn’t been detected?

        The point is that it’s a fair argument. We all know that for every method of detection, whether it’s drug testing, antivirus software or fraud detection software there are people out there creating ways to get around those set ups. Therefor it’s not impossible to think that there are PEDs and masking agents that are available to players that MLB doesn’t have a detection method for yet.

  12. anxovies - May 14, 2014 at 5:41 PM

    So, why are we banning them?

    • davidpom50 - May 14, 2014 at 5:50 PM

      Because they’re illegal in the United States without a prescription, because there are serious side effects, and because this study doesn’t come close to proving that they don’t have an effect.

  13. tbutler704 - May 14, 2014 at 5:57 PM

    I want some steroids.

Leave Comment

You must be logged in to leave a comment. Not a member? Register now!

Featured video

Final take away from World Series
Top 10 MLB Player Searches
  1. H. Pence (2120)
  2. Y. Cespedes (2074)
  3. J. Panik (1967)
  4. M. Morse (1955)
  5. M. Moustakas (1916)