Fisking Feinstein

John Feinstein

I have no problem with people who believe that performance-enhancing drugs affect baseball. There is no doubt that some players have used drugs such as steroids, HGH, amphetamines in an attempt to improve their performances. If they work, players have every incentive to do so. I think this is wrong, and I wish it didn’t occur. However, I find John Feinstein’s grumpy old man condemnation of baseball’s “steroid era” in today’s Washington Post to be seriously lacking in many areas. Given his condemning tone, I think he should be more careful.

It’s time for a fisking.

Welcome to the latest chapter in Major League Baseball’s ongoing nightmare: steroids, steroids and more steroids.

Welcome to a rant about steroids that takes advantage of the public perception that steroids are a huge problem in baseball and that great performances can only be explained by use of PEDs. No, there’s nothing new or interesting here. Tomorrow: how rap music degrades women and glorifies gang culture.

Chances are good that Kirk Radomski, the former Mets “clubbie,” isn’t the only guy out there who sold drugs to ballplayers who is going to turn evidence to try to cut himself a break with prosecutors. Chances are also pretty good that there are going to be some significant names showing up at some point soon.

Chances are that if a petty pusher is a big fish in a steroids investigation has been operating for 4-5 years, that there isn’t much here. With all the leaks surrounding this issue, the fact that actual big names aren’t out yet leads me to believe there won’t be any. Give me some names and details before I get excited.

Those who played in the big leagues during the 1990s are now starting to talk about circumstantial evidence they saw — but for the most part ignored — during that period. As they talk, it is becoming more and more evident that those who didn’t partake were probably the minority.

OK, now we’re getting to the evidence. Maybe there is something worthwhile in here after all.

“It’s at the point now where you wonder if anyone was innocent,” said Ron Darling, who pitched in the majors from 1983 to 1995. “The only reason pitchers from my era are pretty much in the clear is that no one had figured out back then that steroids could help us too. Clearly that changed somewhere along the way but I was out of the game by then.”

What? The “evidence” is that guys are using today because you know from personal experience that guys in the past didn’t use?

Darling has a clear memory of when he noticed the sport changing….When I got to Oakland in 1991…. As soon as the game was over, most of the guys would come inside, change out of their uniforms and go down the hall to pump iron for an hour. At least. Then they’d get up the next morning, workout again and eat 4,000 calories at lunch. And they never got tired.

“Looking back now, it’s pretty clear what was going on. Back then, I just didn’t give it much thought. I was worried about getting ready for my next start.”

Hmm, so working out more equals steroids. Certainly, working out is needed for steroids to work, but you find very few baseball players who don’t work out these days.

Furthermore, if all of this was going in in 1991, why didn’t the jump in home runs spike just in Oakland in the early 1990s? Speaking of data, let’s take a look at how Oakland’s power number changed over the years. I’ll start in 1980 to give a five-year sample before Jose Canseco joined the team.

Year	HR	HR/G	AVG	SLG	Iso-Power
1980	137	0.85	0.259	0.385	0.126
1981	104	0.95	0.247	0.379	0.131
1982	149	0.92	0.236	0.367	0.131
1983	121	0.75	0.262	0.381	0.119
1984	158	0.98	0.259	0.404	0.145
1985	155	0.96	0.264	0.401	0.137
1986	163	1.01	0.252	0.390	0.138
1987	199	1.23	0.260	0.428	0.168
1988	156	0.96	0.263	0.399	0.136
1989	127	0.78	0.261	0.381	0.120
1990	164	1.01	0.254	0.391	0.137
1991	159	0.98	0.248	0.389	0.141
1992	142	0.88	0.258	0.386	0.129
1993	158	0.98	0.254	0.394	0.140
1994	113	0.99	0.260	0.399	0.140
1995	169	1.17	0.264	0.420	0.157
1996	243	1.50	0.265	0.452	0.187
1997	197	1.22	0.260	0.423	0.163
1998	149	0.92	0.257	0.397	0.140
1999	235	1.45	0.259	0.446	0.187
2000	239	1.48	0.270	0.458	0.188
2001	199	1.23	0.264	0.439	0.175
2002	205	1.27	0.261	0.432	0.171
2003	176	1.09	0.254	0.417	0.162
2004	189	1.17	0.270	0.433	0.163
2005	155	0.96	0.262	0.407	0.145

Average	167.73	1.06	0.259	0.408	0.149

There doesn’t appear to be anything suspicious about 1991 or its surrounding years. Granted, you could parse numbers many different ways, but nothing obvious appears to be going on. I’d say Darling’s conjectures don’t have much support.

Every pitcher you talk to from that era insists that the baseball’s changed radically after the strike. …Mike Mussina, now with the New York Yankees, then with the Baltimore Orioles, remembers cutting open two baseball during a rain delay as a time-killing science experiment: “No comparison,” he said. “The core of the old ball just laid down flat on the table. The new ball, the core literally sat up it was so lively.”

Juiced balls, smaller stadiums and juiced players led to the monster home run years enjoyed, not only by Mark McGwire and Sammy Sosa, but non-home run hitters who began reaching out and hitting home runs to the opposite field. Now, belatedly, baseball is paying a price for closing its eyes to what everyone could see just by sitting around a clubhouse and noticing everyone was in the weight room. And never getting tired.

What do balls and stadiums have to do with “steroids, steroids, and more steroids” mentioned in the first paragraph? I guess the lively ball and smaller stadiums kept players from tiring. Maybe they got some greenies from the old-timers who who made the game so great.

[Bud Selig] has to turn his back. And please let us not continue with this foolish notion that Bonds and the others must be convicted beyond a reasonable doubt.

I’ll agree to this if you promise to provide compelling evidence that the rise in offense in the modern era could not have been produced by other factors—including the other factors (smaller ballparks and a lively ball) that you include in this article.

As for Bonds, I think it’s a shame that he has to live with illegally leaked damning evidence that he cannot respond to because of on-going legal process. It doesn’t look good for him, but what does this say about the rest of the league?

If you are paying even a little bit of attention — ala Darling in the Oakland clubhouse in 1991 — it is pretty clear that the number of players using steroids in the last 15 or 20 years has been epidemic.

I might be skeptical and wrong, but it’s not because I’m not paying attention. How about you? What about expansion diluting the talent pool of pitchers? Does Darling’s story check out? You’re not even reading your own article (small ballparks and lively ball)!

Evidence of rampant steroid use does not exist, and there are many other possible explanations for why the game has changed. And given that MLB has been testing for drugs for two seasons and the power numbers didn’t change, I’d say the steroid theory is the weakest one out there right now. And please, don’t even think of bringing up HGH.

[Bonds is] not alone by any means. In fact, every day it is becoming more and more apparent that he was probably a part of the majority. He’s just the symbol of an era that baseball can’t forget soon enough.

The problem is it is going to be a long, long time before anyone is able to forget. Baseball has been stained. Perhaps not forever and probably not irreparably but in ways that won’t be forgotten, and shouldn’t be forgotten, anytime soon.

Where is the damning evidence again? Complaining about steroids in baseball simply in fashion. All the cool kids are doing it, so let’s all jump on. Baseball fans deserve better than this. People sometimes ask me how it is that I can be so skeptical of the impact of steroids in the game. Well, people like John Feinstein make it easy. If I had to choose sides just based on the commentary, it would be an easy choice.

I can’t prove that the game hasn’t been affected by steroids. Indeed, it is possible that it has changed the game. But until I see some better evidence, I remain skeptical.

19 Responses “Fisking Feinstein”

  1. Andrew says:

    I would like to know what Feinstein thinks about usage in other leagues (i.e. NFL) because players test positive much more frequently in the NFL than in MLB.

  2. rob says:

    to put an economics bend to this, isn’t performance enhancing (i.e., maximizing efficiency and output) logical? it’s what business people all over the globe spend their days doing. every other business does it, why would baseball abstain? emergency room doctors use peds to work extended shifts. wall street traders are notorious for taking peds to keep their energy up. business create software unique to their company to provide competitive advantage. we know (meaning there is hard evidence) that athletes in many major sports across the world use peds.

    when looking for evidence that baseball players are using today, based on the above i’m not sure we need conclusive, court of law standard evidence. if we believe economics is how to satisfy unlimited wants with limited resources, that human behavior is to try to maximize ends while using scarce means, isn’t it rational to conclude that baseball players will do this? or at least they would until it became illegal in the game (setting moral objections aside)? the more players get out of their bodies, the better their performance on the field and the more money they make. i agree that we can’t conclusively show how peds have directly affected baseball’s numbers. but we can be pretty close to certain that players do (and always have, in the form of greenies or spitballs, or something else) use performance enhancers today.

  3. Gordon says:

    Good comments on Feinstein. But while I think a juiced ball and smaller parks may well be factors in the 1993-94 offensive surge, expansion was not in fact a significant cause. Overall, HR/game increased by about 45% from 1992 to 1994 (48% in NL, 42% in AL). If expansion was a main cause, the 1992 pitchers should experience a much smaller increase in HR allowed. But that’s not the case.

    I looked at all pitchers who threw 200+ innings in 1992, and also pitched at least 100 IP in 1994, creating a sample of 40 pitchers. This group of pitchers allowed .64 HR/9 in 1992, but .98 in 1994, a 53% increase. And the increase was true at all talent levels. Here are the 1992/1994 HR rates for likely HOFers:
    Randy J .56/.73
    Glavine .24/.54
    Smoltz .62/1.0
    Clemens .40/.79
    Mussina .60/.97
    Maddux .24/.18 (!)
    Only Maddux resisted the trend (only 6 of the 40 saw their rate decline — 3 of those by a trivial amount — while 34 increased).

    It’s clear that it wasn’t just new pitchers giving up the extra HRs — the pre-expansion pitchers saw their rates rise at a comparable level. Expansion was, at most, a tiny factor.

  4. JC says:


    The data you provide is interesting, and this is an excellent comment. I hadn’t thought about looking at the question in this way. Just playing around with the data and tweaking who qualifies, I get the numbers to jump around quite a bit. The numbers seem to be sensitive to cutoffs and sample. Depending on the sample and cutoffs I get home runs rates averaging about 8-30% more after expansion than before it, a bit smaller than you suggest. And this is without taking into account aging, weather, park changes, etc.

    One problem with this is that we can’t correct for is that bad pitching spills over. It taxes good pitchers who must pick up the slack. But, there is no such thing as a perfect study.

    I have a good bit on how expansion affected several aspects of the game in my book. The dilution of pitching talent is pretty remarkable. It might not be the entire answer, but I think expansion’s contribution is real.

  5. Someone should send a copy of The Black Swan to Feinstein and all the other writers who are convinced that the game has been ruined by steroids. It seems as though PED use fits their story ex post, and thus they have decided to run with it. This hypothesis (which writers insist is true) is merely an example of confirmation bias.

  6. Frank says:

    A little grumpy today JC? Little one keep you up late last night? πŸ™‚

    Here’s a cure–take a look at the Braves batting stats and ignore Feinstein. Kelly Johnson has a 1.05 OPS; Thorman and Francoeur are both above .900. Lots to feel good about there. With a new leftfielder and an upgrade to the back of the rotation, this could be a pretty fun year in ATL. Of course, it’d be much easier to fill one of the holes if JS hadn’t wasted $2.5 mill on Sturtze, Woodward, and Redman.

  7. JC says:

    Steroids could be a black swan, but then so could any of the other explanations. Thanks for mentioning the book, Josh. I’ve been meaning to pick up a copy.

    The funny thing is, when I started looking into the distribution of talent across leagues, it didn’t initially occur to me that this might fit with home runs and hit batters. I was just trying to compare eras for the fun of it. I remember putting two and two together and going “Holy crap! Why didn’t I think of this earlier?” If I had only been doing this 15 years ago, I could have predicted it. πŸ˜‰

    I think Gould’s theory holds a lot of promise. It’s tough to know exactly what is going on, but the rush to tout steroids as the only possible explanation makes me very grumpy. Sorry Frank, not even a good Braves team can cheer me up about this. πŸ™‚

  8. Gordon says:

    JC: I suppose that starting with only the 200+ IP pitchers in 1992 could produce a small bias, so I expanded my sample to all starters who pitched at least 100 IP in both 1992 and 1994. (1993, as you know, was a transition year: more HRs than 1992, but not as many as in 1994 and thereafter.) This adds another 22 pitchers to the sample, and the pattern is basically the same: .73 HR/9 in 1992, 1.0 HR/9 in 1994, an increase of 40%. Combined results for all 62 pitchers: .67 in 1992, .99 in 1994, +48%, almost exactly the same as the actual 45% MLB-wide increase (and a +.31 delta, vs. MLB-wide +.32).

    I hope you will share your results. But it’s hard to see how you would get anything significantly different, as these pitchers account for about half of all starters’ IP in 1992. Age and park might have small impact, but likely to be a wash.

  9. Gordon says:

    FYI, here are my results (hope formatting isn’t too ugly). Part one:

    NAME, 1992 HR/9, 1994 HR/9, DELTA
    Sid Fernandez, 0.50, 2.11, 1.60
    Terry Mulholland, 0.55, 1.79, 1.24
    Juan Guzman, 0.30, 1.22, 0.92
    Mark Langston, 0.55, 1.43, 0.88
    Jim Abbott, 0.51, 1.35, 0.84
    Cal Eldred, 0.36, 1.16, 0.80
    Dave Fleming, 0.51, 1.31, 0.80
    Mike Moore, 0.81, 1.58, 0.77
    Bobby Witt, 0.78, 1.46, 0.68
    Dave Stewart, 1.13, 1.76, 0.63
    Greg Swindell, 0.59, 1.21, 0.62
    Kevin Brown, 0.37, 0.95, 0.58
    Zane Smith, 0.51, 1.03, 0.52
    Bob Tewksbury, 0.58, 1.10, 0.52
    Andy Benes, 0.54, 1.05, 0.50
    Bill Swift, 0.33, 0.82, 0.50
    Tim Belcher, 0.67, 1.17, 0.50
    Mark Portugal, 0.62, 1.11, 0.49
    Bill Gullickson, 1.42, 1.87, 0.45
    Charles Nagy, 0.39, 0.80, 0.40
    Roger Clemens, 0.40, 0.79, 0.39
    John Smiley, 0.63, 1.02, 0.39
    John Smoltz, 0.62, 1.00, 0.38
    Charlie Hough, 0.97, 1.35, 0.38
    Scott Kamieniecki, 0.62, 1.00, 0.37
    Mike Mussina, 0.60, 0.97, 0.37
    Melido Perez, 0.58, 0.95, 0.37
    Ron Darling, 0.65, 1.01, 0.36
    Orel Hershiser, 0.64, 1.00, 0.36
    Steve Avery, 0.54, 0.89, 0.35
    Alex Fernandez, 1.01, 1.32, 0.31

  10. Gordon says:

    Tom Glavine, 0.24, 0.54, 0.30
    Ramon Martinez, 0.66, 0.95, 0.30
    Chris Bosio, 0.82, 1.08, 0.26
    Danny Jackson, 0.40, 0.65, 0.25
    David Cone, 0.55, 0.79, 0.24
    Dennis Martinez, 0.48, 0.71, 0.24
    Jack Morris, 0.67, 0.89, 0.22
    Darryl Kile, 0.57, 0.79, 0.22
    Kevin Appier, 0.43, 0.64, 0.21
    Jose Rijo, 0.64, 0.84, 0.20
    Todd Stottlemyre, 1.03, 1.22, 0.18
    Randy Johnson, 0.56, 0.73, 0.18
    John Burkett, 0.62, 0.79, 0.17
    Scott Erickson, 0.76, 0.94, 0.17
    Doug Drabek, 0.60, 0.77, 0.17
    Ken Hill, 0.54, 0.70, 0.16
    Kevin Gross, 0.48, 0.63, 0.15
    Bill Wegman, 0.96, 1.09, 0.13
    Mark Gubicza, 0.65, 0.76, 0.12
    Erik Hanson, 0.67, 0.73, 0.06
    Kevin Tapani, 0.70, 0.75, 0.05
    Mark Clark, 0.95, 0.99, 0.04
    Chuck Finley, 1.06, 1.03, -0.03
    Butch Henry, 0.87, 0.84, -0.03
    Tom Candiotti, 0.57, 0.53, -0.05
    Greg Maddux, 0.24, 0.18, -0.06
    Jack Mcdowell, 0.72, 0.60, -0.13
    Joe Hesketh, 0.91, 0.71, -0.20
    Jimmy Key, 1.00, 0.54, -0.46
    Ben Mcdonald, 1.27, 0.80, -0.47
    Ricky Bones, 1.49, 0.90, -0.59
    AVERAGE, 0.67, 0.99, +0.32

  11. JC says:

    Well, the main difference is that I’m not just picking 1992 and 1994. Here is are two samples. The first includes pitchers who threw more than 200 IP in 1992 and 1993 (it’s an easy way to catch pre and post expantion pitchers). The second includes lowers the cutoff to 100 IP. 94 is excluded due to the small sample.

    200 IP Sample
    Year	N	HR9
    1986	8	0.85
    1987	7	1.00
    1988	9	0.71
    1989	9	0.63
    1990	13	0.71
    1991	22	0.65
    1992	29	0.61
    1993	29	0.75
    1995	7	0.74
    1996	13	0.81
    1997	10	0.75
    1998	11	0.84
    1999	6	0.84
    2000	5	0.84
    86-92		0.74
    93-00		0.80
    %Diff		0.08
    100 IP Sample
    Year	N	HR9
    1986	35	0.88
    1987	34	0.99
    1988	42	0.69
    1989	54	0.70
    1990	63	0.75
    1991	72	0.74
    1992	95	0.69
    1993	95	0.87
    1995	59	0.92
    1996	53	1.01
    1997	42	0.94
    1998	39	0.98
    1999	33	1.12
    2000	23	1.13
    86-92		0.78
    93-00		1.00
    %Diff		0.28
  12. Gordon says:

    Great, our results are totally consistent. Your data show a 1992-1993 increase of 23% (200 IP) to 26% (100 IP) among incumbent pitchers. The actual 1992-93 MLB increase overall was 24% — .723 in 1992, .896 in 1993. So again we see that the returning pitchers experienced the same increase in HR rates that we see overall. The influx of weaker pitchers can’t explain the increased HR rates for Glavine, Smoltz, et. al.

    So we’re back to the ball (likely the biggest factor, given the sudden nature of the change), bats, weight training, parks, etc. as causal candidates. I think we can cross expansion off the list.

  13. Norman Shatkin says:

    An additional point: while steroids may well convey an advantage to hitters, the most important aspect of their skill is not strength, but eye-hand coordination. His strength does not help Bonds square up on a 90+ fast ball, without which he doesn’t get the extra feet of flight that steroids may provide him.

    But pitchers CAN gain a significant advantage based upon simple strength — the ability to throw the ball faster. While pitching also has a high level of skill, it also relies to a much greater degree upon velocity. Jamie Moyer is an exception but to almost everyone else, a live fast ball is pretty much essential.

    Therefore, if steroid usage is as rampant as it’s supposed to be, we should be seeing hitting DECREASE, since it seems to me it would be more adversely affected if everyone playing the game gained x% more strength than would pitching.

    On the other hand, juicing up the ball and bringing in the fences WOULD provide exactly what we’re seeing — more home runs without a major increase in batting averages.

  14. JC says:

    Why should we focus on one single year? The analysis is a bit more complicated than you are making it out to be. Even what I have done (and I disagree with the way you are interpreting what I’ve shown, because the average change in HR rates is much more smaller than you suggested) is far too modest to reveal exactly what is going on so that we can exclude expansion.

    Also, just because a player stays around or sees his totals go up doesn’t mean expansion is not part of the explanation. Some of these pitchers may include those who should have thrown fewer innings or retired.

    What is interesting is that I think you’ve pointed out that maybe something else is going on, which I don’t disagree with.

  15. Gordon says:

    Ah, now I see your multi-year average calculations. I was just looking at 1992 vs. 1993.

    The 200 IP sample is of course extremely small. And more importantly, you have a survivor bias problem. In 1992 and 1993, this sample gave up HR at about 85% of the league rate. But in 1995 and after, the remaining pitchers are MUCH better than that, suggesting the ones still in your shrinking sample are the best pitchers (as we’d expect).

    In any case, if we look at your larger sample, or just a 1992 v. 1993 or 1992 v. 1994 comparison, they all tell the same story: Incumbent pitchers gave up a lot more HRs than they had previously, and new pitchers played little or no role.

  16. Gordon says:

    “Some of these pitchers may include those who should have thrown fewer innings or retired.”

    Doubtless true in a few cases. But look at the pitchers in posts 9 and 10. Just 5 saw a decline of .1 or more, while 50 saw an increase of .1 or greater. Many of them were in their prime or young enough to still be improving, yet still saw a massive increase in HRs allowed.

  17. Lou says:

    I really can’t stand Feinstein and thank you for breaking down his article in such a fashion that questions his motives.

    I do think however, that if at least one shocking name is leaked or made public that the steriod craze will pick up again. What happens, just throwing a name out there, if Andy Pettitte is found to be purchasing steriods. The craze will be like it never has been before.

  18. Gavin says:

    Reading the comments, it is clear several people share my concern that Feinstein is singling out hitters as the main culprits of his PED’s era.

    At the same time, he is “interviewing” an old friend, Ron Darling, and giving the ex-pitcher a forum to defend all pitchers against these monsterous hitters who were ruining the game.

    The real danger of this column is that it perpetuates the myth that only hitters have taken advantage of PEDs.

    It isn’t the strength issue for pitchers. The major benefit of PEDs is reduced healing time. That is the biggest benefit to a pitcher, especially a relief pitcher.

  19. Nolan says:

    1) Any writer who quotes a player’s analysis of the way the ball has changed is just irresponsible. In my mind, the hysteria about the ball is worse than anything that’s been said about steroids – it’s basically moon-landing conspiracy theory stuff and its silly.

    2) JC wrote: “As for Bonds, I think it’s a shame that he has to live with illegally leaked damning evidence that he cannot respond to because of on-going legal process.” I’m going to have to disagree here. There is literally nothing legal preventing Bonds from saying: “I didn’t do anything.” He hasn’t done that and he can’t do that (honestly) because he clearly has done something.