THE BOOK--Playing The Percentages In Baseball

There’s evidence, and there’s PERMISSIBLE evidence. 

Evidence is simply anything that changes your estimate. 

Being mentioned in the Mitchell report is evidence, in the sense that, all things being equal, being mentioned increases the probability that player X took steroids.

That evidence may be unfair to use against player X in certain contexts, but it’s still evidence.  Statistical evidence, or circumstantial evidence, or hearsay evidence ... but still evidence.

There are times when it’s unfair or rude to use statistical evidence as if it were specific evidence.  I’ll bet you better than even money that, statistically, African-Americans do eat more collard greens per capita than whites; Jewish-Americans eat more Matzo balls per capita than non-Jews; and and Hispanic Americans eat more tacos per capita than non-Hispanics. 

But if a black person I’ve never met comes over for dinner, and I make collard greens just on the basis of a (probably true) statistical stereotype ... that’s rude.  And there are very obvious reasons it’s considered rude.

On the other hand, if I decide to send more Matzo Meal per capita to supermarkets in New York than supermarkets in New Orleans, that’s probably OK—and good business sense.

So, it’s not Evidence vs. Not Evidence.  It’s Evidence that’s OK To Use In This Particular Context, versus Evidence That’s Not OK To Use In This Particular Context.

I meant using the Mitchell Report as it relates to UNMENTIONED players (Edgar and Jeter), and trying to decide on the population of people Edgar and Jeter belong to, and relating that to the population of players noted in the Mitchell Report.

But, your example of the dinner guest is exactly correct. 

By that standpoint, if Phil sits down at my dinner table, I will say:
“Phil, there’s evidence that you are an a$$hole, that you don’t pay all your taxes to Revenue Canada, that you kicked kids off the playground, and that not only watch Glee, but like to watch Glee.”

Indeed, I can apply that to any Canadian.

How exactly is that “evidence” in any useful sense?  Would you really argue for evidence on that basis in court?  Aren’t you going to lose all 12 jury members with that kind of “evidence”.

The reality is that we need to have REASONABLE definitions, useful definitions, definitions that allows us to communicate effectively.

There’s evidence that a statistician will use and be able to come up with a probability rate of .000001 to .999999 for whatever event we are talking about.

And then there’s the definition of evidence that allows us to talk like regular human beings.

And by that definition of the word evidence, there is NO evidence that differentiates Jeter and Edgar in particular.

I’d say there is.  Jeter has been subject to random steroid testing since 2005.  He has not failed a single one. 

Edgar last played in 2004, the last year before testing.  So was he afraid to face testing?  Or should a 41 year old DH whose OPS+ drops to 92 leave the game anyway?

Just kidding there.  But if we’re looking for all types of evidence, then 7 years of clean tests should put Jeter at a lower probability than Edgar.

What I’m saying is, evidence of a population IS evidence of an individual.  If you find that 40% of Canadians have ripped off Revenue Canada, then, knowing nothing else about me, your prior should be that there is a 40% chance that *I* have ripped off Revenue Canada.

If there were more steroid users during Edgar’s time than Jeters, then, yes, that IS evidence that you should give Edgar a higher prior than Jeter.

But ... that’s just probability.  What we’re really concerned with is *fairness*.  And it’s unfair to accuse Edgar of anything without evidence that’s *specific to him*, rather than just statistical evidence of the groups he belongs to.  (I’m sure there are rules about how that kind of evidence can be used in court.)

Moreover, even if you did use that statistical evidence, it’s *weak* evidence.  On the other hand, if I’m the only Jew in town, and you found traces of gefilte fish breath at the murder scene ... well, that’s *stronger* evidence.

So, I’m saying that Edgar’s size, his era, and his stats ARE evidence.  But, (a) it’s weak evidence, and (b) because it’s weak and statistical, it’s rude for us to evaluate him only by that evidence.  (That’s the definition of stereotyping: concluding based on weak and statistical evidence.)

I think we agree on most of this.  The difference is that I’m saying that “unacceptable” evidence is still evidence, and you’re (I think) saying that it’s so weak it’s not worthy of the term.

P.S.  I’ve never seen Glee, I’ve never kicked kids off the playground, and I don’t rip off Revenue Canada.  I might, however, be an a$$hole.

Phil: I can’t take as evidence your claims of what you’ve never done, since everyone in the population makes the same claims (absent actual evidence against them).

That is, your own testimony of what you have done, which could be presented in court as evidence, has to be rejected in the statistical sense because everyone else would make the same testimony.

So, the ironic thing is that Phil’s evidence is only evidence if you reject Phil’s definition of evidence!

It’s not clear to me what evidence Rob was talking about, unless it’s just each reader’s own imagination of what might incriminate Martinez and exonerate Jeter.  I’d be inclined to take Rob’s argument more seriously if he talked about anything specific.

When Colin addressed a specific piece of evidence, Rob said that wasn’t what he was talking about, but he didn’t say what he WAS talking about.  So, it’s tough to engage Rob on this if he wants to remain that vague.

Tango/5: On ripping off Revenue Canada, I agree with you.  Everyone would make the same claim, and therefore it has no value as evidence.

But some people would admit to kicking other kids off the playground.  And many people would admit watching Glee.  So, on those, you should update your prior based on what I said, because your prior does change.

Also, you could update based on how trustworthy you find me.  If 50% of people watch Glee, but only 10% of those 50% admit it, then the probability a denier watches Glee is 45/95, or 47.3%.  But, if have evidence that suggests I’m more or less likely than most to lie about it, then you should update based on that, too.

As an aside ... is Glee-watching-denial that common?  How come?  Is it a a masculinity thing?

Again, I don’t see that this is “evidence”.  Certainly not evidence you’d present in a court of law.

These are characteristics of bias. 

The evidence shows that a portion of the population of men does watch Glee, and that population doesn’t admit to it.

Because Phil shares the one bias of this population (he’s a man), then the evidence shows that he “may” watch Glee, but not admit to it.

Is this what we mean by “evidence” as it relates to Phil?  Or shouldn’t we just treat this as evidence as it relates to the huge population that Phil just HAPPENS to belong to, through no choice of his own?

Do we really want to use this as evidence against Phil?

***

Now, Rally makes a fine point that Edgar never submitted to PED testing, while Jeter has.  But neither did Andre Dawson nor Willie Mays.

While single out Edgar here?

Another perfect joke, this time from Colin Quinn:

http://www.tumblr.com/tagged/colin-quinn?before=1313265545

Chinese computer hackers have broken into the military’s computer system and shut down a satellite.

Such an act could only be the work of a mathematical genius.

Chinese authorities investigating the case have narrowed it down to 600 million suspects.

I’d guess the evidence Neyer has in mind re: Edgar is his late-career power surge.  It’s not uncommon to start hitting for more power at age 26-27.  But Edgar had a big surge at age 32, which is pretty unusual.

Wait. So jokes about LGBT people are universally “banal, mean-spirited, and dated” but “Asians are good at math” jokes are “perfect”??

Guy/10- Whitaker and Dwight Evans are the other examples I can think of.  Both of whom presumably pre-date the PED era.  It’s unusual, but it does happen.  Not co-incidentally, these 3 would be the 2B, RF, and DH on the all hiatorically under-rated team....

tmp: you missed the entire context of the joke if that’s what you took away from it.

I’m relating the ridiculousness of Quinn’s application of Bayes based on whatever population group he’s creating (similar to the Jews/Matsoballs that Phil noted), to show “evidence” of something.

In any case, joking that Chinese people being GOOD at math is somehow “mean-spirited”?  I already noted that the way Don Rickles does jokes (equal-opportunity offender) is the right way to make jokes of groups.  It’s whether you are mean-spirited or not.

Did your first post have to be an attempt at a “gotcha” post?  Really?  Why are you even here?

Pierre:
I don’t think Dewey is comparable to Edgar.  Here is Edgar:
Age / ISO
27 0.131
28 0.145
29 0.201
30 0.141
31 0.197
32 0.272
33 0.268
34 0.224
35 0.243
36 0.217
37 0.255
38 0.237
39 0.208
40 0.195
41 0.122

And here is Dewey:
Age/ ISO
20 0.141
21 0.16
22 0.14
23 0.182
24 0.189
25 0.239
26 0.202
27 0.182
28 0.218
29 0.226
30 0.242
31 0.198
32 0.237
33 0.191
34 0.217
35 0.264
36 0.194
37 0.178
38 0.142
39 0.108

Something really changed with Edgar at age 32. I don’t see a comparable dramatic change in Evans’ data.

I had forgotten how late Edgar bloomed.  First season as a regular at age 27, first HOF-calibre season at 29, and didn’t begin sustained excellence until age 32.  Is there anyone in the HOF with a similar trajectory?  (Excluding those impacted by segregation and/or military service.)

"Now, Rally makes a fine point that Edgar never submitted to PED testing, while Jeter has.  But neither did Andre Dawson nor Willie Mays.

While single out Edgar here?”

Didn’t mean to, just those were the players mentioned in the lead in.  If you accept that steroid use was rare or non-existent earlier and became more common between the Canseco-Caminiti MVP awards, then the chances of being users would have to go:

Edgar>Dawson>Mays

"Is there anyone in the HOF with a similar trajectory?”

Not sure about HOF.  Brian Downing was similar, ISO jumping from .129 to .201 at age 31.  The .129 was a normal figure for him pre-1982, and he maintained his new higher level.

Whitaker had a pretty big jump at age 28, another at age 32, and then set his career high in ISO at age 38.  Then he retired.

The league jumped in offense in 1993-94, which doesn’t explain nearly all of Martinez’s increase, but it needs to be accounted for.  Unless you think the league increase was largely because of a sudden increase in steroid usage right then, but that seems implausible.

If I were trying to defend someone against suggestions they might be a PED user, I think Brian Downing is one of the LAST examples I’d want to cite. (And I’m not saying that was your point, Rally.)

My question about “similar trajectory” was really in terms of being such a late bloomer in general:  becoming a starter at 27, and a consistently great hitter at age 32.  This is very rare, I think.

Downing’s a good one.  Evans’ improvement was all-around, I guess, not just in HRs.  Although in his 30s he topped the highest HR total he had in his 20s 5 times.

Mike/17:  Edgar’s big surge came in 1995. That’s 1 or 2 years too late.  Everybody else saw their power increase in 1993 and 1994—but Edgar’s big move was a year later.  So unless it took an extra year for the juiced balls to reach Seattle, something else happened.

I’m not saying this is “proof” of PED use, or even necessarily strong evidence.  But it is evidence, I think, and probably one of the things Neyer was thinking of.  (Though I agree it’s weird he won’t just tell us what he was thinking about.)

Guy/20, I’m not saying that the league power increase completely explains Martinez’s power increase.  I’m saying that you need to baseline his ISO numbers to the league environment.  I assume you’re not just saying that his power jumped from 1994 to 1995 but also that his power from 1995 onward was much higher than his power from 1994 and before, in which case you really need to incorporate the league increase into that comparison.

Rally/3 -

That Jeter passed PED tests from 2005 on does not give evidence of whether he used PEDs earlier in his career or not.

With most PEDs, stopping usage shortly before the test (the exact time varies per substance, and likely to some extent among individuals varies) is sufficient pass. And there’s significant potential benefit from using in the training phase or offseason, even if you abstain during competition. Thus in sports like track and field they do random, unannounced, year-round testing. Baseball has only done that since (I think) 2005, although the effectiveness of the deterrent depends greatly on the details of implementation (e.g. how likely a player is to be selected for random testing). Testing only at major competitions, or simply any announced dates (at one point, and perhaps still, the dates of NFL drug tests were known well in advance) makes it much easier for PED users to pass their tests.

One other point from track: some athletes have passed years of tests only to test positive fairly late in their careers: Regina Jacobs comes to mind here, but there are other examples.

I cite Jacobs to note that even passing tests for years is not proof that an athlete was not using PEDs: some athletes who were using have been able to successfully avoid detection, and it is naive to believe that wouldn’t be true of at least some baseball players, whether it’s Derek Jeter, Greg Maddux, or Edgar Martinez.

We know Jeter hasn’t had any reported test failures, but that is likely far weaker evidence than many people would like, whether applied to Jeter or anyone else in MLB.

About Martinez’s relative surge in 1995, I’d note that he played just 89 games in 1994, and only 42 in 1993. The strike reduced the Mariners’ total to 112 games in 1994, but it seems likely injuries were a significant factor for Martinez, especially in 1993.

In 1992, he won his first batting title, and led the league in doubles (often, but not always, a harbinger of more HR power later in one’s career).

Oh, and 1995 was Martinez’s first year primarily as a DH, as opposed to trying to play 3B also: he played 64 games at 3b in 1994, but never reached double digits in games in the field again.

One can build a narrative saying that Edgar’s improvement came in large part because he stopped trying to play in the field, and/or becoming a full time DH enabled him to stay healthy enough to hit regularly.

Using performances alone as evidence of potential PED usage is at best highly circumstantial. If that’s all Neyer had in mind, I’d be unconvinced. But I also wouldn’t be surprised to hear that any particular MLB player actually did use PEDs.

"That Jeter passed PED tests from 2005 on does not give evidence of whether he used PEDs earlier in his career or not.”

This is true.  It doesn’t even prove that he was clean after 2005.  But I’m just talking about probabilities.  Knowing a player passed a test has to put him at at least a slightly lower probability of juicing than a player who never had to take a test (but played in a time where steroids were prevalent).

Oh, and while I’ve never seen “Glee”, I did sing in a Glee Club in college. Does that count?

Luis Gonzalez had a huge ISO bump starting at age 30.  Of course Gonzalez was reportedly on the leaked list of positive testers on 2003.

OTOH, Some of the players discussed in this thread (including Edgar) had children during their playing careers.  Which based on the documented link between steroids and temporary sterility, at least offers evidence of discrete points in time when they were probably not using steroids.

All that said, I strongly agree with the opening sentence of Phil/1:

There’s evidence, and there’s PERMISSIBLE evidence.

It’s human nature to want to peek at all evidence whether it’s inappropriate / out of bounds or not.  But journalists should be held to a higher standard than that.  Not necessarily a courtroom-level standard, but certainly much higher than the 2am-at-the-bar standard.

Mike/21:  Sure, for a proper comparison of “early” vs. “late” Edgar one should adjust ISO for league context.  I was interested only in the sharp change at age 32, and since that can’t be explained by league context, I thought the numbers told a valid story. 

And I think it’s clear Edgar became a much better hitter in his early 30s.  Here is his OPS+, so we’ve adjusted for league/park:
27 132
28 138
29 164
30 100
31 121
32 185
33 166
34 165
35 158
36 152
37 157
38 160
39 139
40 141
41 92

Edgar was literally a better hitter at ages 39-40 than he had been at age 27-28.  That’s pretty strange.  How many hitters has that true for?  And his age 27 performance wasn’t a fluke—he was a high-average hitter in the minors, but also averaged just 10 HRs per 600 PA in AAA (in over 2800 PAs).  Even if we grant he had injury problems at 30-31 (which, if true, makes it even more plausible he turned to PEDs at age 32), he still took a big step forward at age 32.  And becoming a DH should reduce, not increase, offensive production.  Edgar’s trajectory may not literally be unprecedented.  But it’s damn unlikely, and guys who did unlikely things in 1995 will tend to come under suspicion.

Well, I think people saying “Jeter has had to pass test, Edgar did not” part of the point .. I suppose, that Neyer was trying to make.

It would be easy for somebody to formulate evidence that lead them to believe Edgar was for more likely to be a PEDs user.

Edgar never had to take a drug test.

Edgar late bloom of his career.

Edgar’s injury history and that some admitted PEDsters have indicated they used because of injuries.

Plenty of evidence to show Edgar was a a more likely PEDster than Jeter.

Note, when I use evidence here it should be quoted. The “evidence” is somebody’s idea of evidence. Very much like the convoluted logic that some HOF voters use. It isn’t valid yet they use it to invalidate somebody.

Now if somebody said something like, “I never seen Martinez cheat. I have seen Jeter act like he was injured in an attempt to deceive an umpire about getting hit by a pitch. If Jeter is willing to lie about something that would so obviously be shown to be false, it leads me to believe he would not have a problem lying about PED usage and going to steps to hide it.” that might carry a bit more weight. Yet, it still isn’t evidence.

Edgar was better in the 2nd half of his career, which is rare.

Edgar performed better as a DH than he did as a position player, which is rare.

Edgar could just be a unique case.he could have used PED’s.

We’ve seen that being a nice guy and like able doesn’t mean they haven’t used PEDs.

Edgar followed two partial seasons with a string of highly productive seasons at an advanced age during the steroid era. Not unlike McGwire.

There’s enough there to have suspicion IMO, but maybe not enough for accusation.

Often being very healthy following injuries has been an indicator of PED use. In Edgar’s case it could also have been due to not playing the field.

My heart wants to think there’s no chance, my brain says there’s some chance.

In regards to the HoF Edgar is sorta unique because he’s one of the first, if not the 1st player to be primarily a DH. David Orriz will be the next case. I don’t think the voters know what they want to do with these guys. Edgar was a league average fielder, so he didn’t DH because he couldn’t field. Whether DHing prolonged his career significantly is another story. But the league created the position, so for him to be penalized for playing that position doesn’t seem right.

Some deserving players increased their vote, but is this just due to the thin ballot? When mega stars enter the ballot in upcoming years will the increase cease or reverse?

I freely admit that I love Edgar Martinez. Before Albert Pujols, there was Edgar Martinez as the guy who had a flatter swing path than most power hitters, enabling the barrel of the bat to stay in the zone longer resulting in power to all fields, particularly oppo.

Unfortunately for Edgar (and his fans) there are some reasons (evidence?) to have suspicion.

1. His rare career path; becoming a better player in the second half of his career.
2. The increase in health (could also be due to DH’ing).
3. His body change. He was very slight in his prime age season, very stocky in his later years. Some of this is natural I suppose.
4. He played in the steroid era which featured some awkward career paths ... guys getting better and more healthy in their aging years.

Edgar martinez in 1990.
http://shop.sportsworldcards.com/ekmps/shops/sportsworld/images/seattle-mariners-edgar-martinez-645-donruss-1989-mlb-baseball-trading-card-38648-p.jpg

The Edgar we remember:
http://i.usatoday.net/communitymanager/_photos/daily-pitch/2012/01/05/martinezx-large.jpg

To be fair, Jim Thome went through a similar body change, as did Jeff Bagwell. By itself it is not a major piece of evidence as it didn’t seem to occur in one off-season or so. Mark McGwire underwent perhaps a more serious change, and we are under the assumption that he wasn’t really using steroids in 1988-1990 (like canseco said) when he he turned into a physical monster (and even moreso in the mid 90s).

I think having a teammate widely presumed to use steroids (Brett Boone) factors in.

It is possible that Edgar is just a unique situation in that he got better during his past-prime seasons and he got healthier via moving to DH, and he simply filled out well as he aged (like many of us do).

The problem for Edgar is that the guys that did something similar to him, experiencing a string of good health following injuries and partial seasons, adding a fair amount of weight, and having better production during aging seasons than they did in their prime were often linked to steroids (McGwire, Bonds, Caminiti, Gary Matthews Jr to a lesser extent, etc).

Again, it’s possible that Edgar is unique, and I’d love to believe that whole-heartedly, but when other have used PEDs to attain the same unique path/performance, it makes it difficult to assume that he more likely did not use PEDs.

The career path is the big piece of “evidence”. It’s just so abnormal. Fluke seasons are one thing, fluke career paths are harder to understand.

I don;t really see Lou Whitaker as being in an all that similar situation. The thing that seems weird about Whitaker’s career is that 1987 was a down year for him, when the rest of the league seemed to find new offense, presumably because of the baseball.

I have a lot of probably random and unconnected thoughts on this subject.

First, I was disappointed in Rob Neyer’s comments on this subject because he is someone I greatly respect even though there are several things (including Edgar’s HoF case, which I believe to be just short of slam-dunk quality and Rob has gone back and forth on and thinks is very very borderline).  The use of the word “evidence” was disappointing, not because of the use of THAT word itself but because of a lack of appreciation for what KINDS of evidence we normally allow to be considered, for reasons of fairness.  Many others here have made that very point.  I realize this is not a legal proceeding, but the analogy to the weighing of “probative value” of evidence to its “prejudicial effect” in determining the admissibility of that evidence in a FAIR trial is a very good one.  There is simply NO WAY you can bring Edgar’s name into this discussion in any kind of fair way, given the lack of weight of the evidence against him and the highly predjudicial effect as opposed to the very, very slight probative value it may hold.

I believe the “evidence” against Jeff Bagwell is similarly weak and outweighed by its highly prejudicial nature, but there is certainly more evidence against him than Edgar.  Google Bagwell and Kelly Blair, and there are certainly some whispers, and investigators (mostly surrounding the Clemens investigation) snooping around.  There is nothing comparable with Edgar, except the Shane Monahan story, which did not implicate Edgar (or anybody) directly and only painted a broad brush picture of greenie use (not steroids), and was sidely disputed by players who were part of the Mariners of that era for many more games that Shane Monahan.  Others have pointed out that Edgar’s considerable “gap power” is often a precursor of home run power that will develop later, and that was evident even in MiLB.  Edgar had 1159 PA in AAA, hitting .344/.450/.495, with 95 XBH (68 2B, 6 3B, 21 HR).  He had good power - .495 SLG compares pretty favorably to Bagwell’s .446 SLG in 597 PA in AA (his highest MiLB level).  I don’t think there is enough on EITHER guy to feel comfortable going on anything but their (excellent) numbers, but there is a lot more smoke surrounding Bagwell than Edgar and I think it is fundamentally unfair to drag him into the morass surrounding Bagwell’s candidacy.  He’s got enough problems just overcoming the DH bias.

The other thing you see people do - Guy is doing it here, and Jeff Fletcher did it in response to Dave Cameron’s Fangraphs piece on Edgar - is point to Edgar’s 1995 jump and strongly insinuate that, since this was the 1990s, well.....  Edgar’s career arc is not that unusual except that the front half (pre-age 27) isn’t there, for reasons that are not really his fault given the level of performance he established in AAA (playing 3B).  Plenty of great players had sustained high levels of performance until age 40, just as Edgar did.  The difference is he really didn’t begin in earnest until age 27.  His lower levels of performance in 1990 and 1991 can be explained by adjustment to being a MLB regular. 

Before 1990, he had less than 300 PA in MLB. In 1990 and 1991, he hit a combined .305/.401/.443 (ISO .138 compared to a league ISO of .132) before his breakout season in 1992, when he won the batting title and increased his ISO to .201 (compared to the league ISO of .126 that year).

Then he got hurt.  1993 was a series of injuries and struggles, and 1994 began the same way, getting hit on the wrist by a pitch in the last exhibition game of that Spring.  The regression in his ISO in 1993 and 1994 (1993 ISO of .141 vs. a league ISO of .141, and 1994 ISO of .197 vs. a league ISO of .163) was ENTIRELY injury-related.
In 1995, he picked back off where he had really left off in 1992, winning another batting title, and seeing his ISO jump to .272 (vs. a league ISO of .158), and things pretty well leveled off at that point at a level pretty well established in what were essentially his 3rd (1992) and 4th (1995) full, healthy MLB seasons.

Guy suggests that “if” it is true that injuries derailed his 1993 and 1994 seasons (which is an easily confirmed fact), then that just makes it more “plausible” that he used PEDs in 1995.  First, that ignores his 1992 season, and secondly, 1995 is just as easily explained by more and better training (widely reported at the time), better health, and an ability to focus solely on hitting.  Injuries and poor performance followed by health and better performance is NOT so much “evidence” of PED use as it is, at best, a possible explanation of a potential MOTIVE to use PEDs.

If you forget about the fact that Edgar performed for four years at a very high level in AAA at a time when other players of his abilities were establishing themselves in MLB, then you get this picture of Edgar as an unprecedented outlier who peaked extremely late and sustained it to a seeminly unnatural age.  But his career wouldn’t look a whole lot different than other comparable hitters who played well into their late 30s and early 40s if their careers had not been allowed to begin until 27.  We can’t know for sure that he would have suceeded in MLB earlier if given a realistic chance, but his AAA numbers (and subsequent MLB performance once given a proper chance) suggest he would have.  His career arc suddenly doesn’t look so strange if you ignore his age and just look at 1st through 15th full years, 1990-2004. 

I also take issue with the description of Edgar as this huge guy.  He’s probably 6’ tall, maybe a tad more, and always had a fairly thick build that aged just as I would expect that build to age.  He never looked bulked up, to me (and I saw a LOT of Mariners games in his career).  YMMV.

By the way, her are Edgar’s PA, ISO, League ISO, and ISO Difference year-by year:

87-90: 280 PA, .098 ISO, .133 league ISO, -.035
1990: 570 PA, .131 ISO, .129 league ISO, +.002
1991 642 PA, .145 ISO, .135 league ISO, +.010
1992: 592 PA, .201 ISO, .126 league ISO, +.075
1993: 165 PA, .141 ISO, .141 league ISO, +.000
1994: 387 PA, .197 ISO, .163 league ISO, +.034
1995: 639 PA, .272 ISO, .158 league ISO, +.114
1996: 634 PA, .268 ISO, .168 league ISO, +.100
1997: 678 PA, .224 ISO, .156 league ISO, +.068
1998: 672 PA, .243 ISO, .160 league ISO, +.083
1999: 608 PA, .217 ISO, .167 league ISO, +.050
2000: 665 PA, .255 ISO, .164 league ISO, +.091
2001: 581 PA, .237 ISO, .158 league ISO, +.079
2002: 407 PA, .208 ISO, .159 league ISO, +.049
2003: 549 PA, .195 ISO, .157 league ISO, +.038
2004: 549 PA, .122 ISO, .161 league ISO, -.039

I fail to see how this career arc is really “strange,” once 1993 and 1994 are explained.  1995 and 1996 are the peak of a great player’s career.  It is a jump of approximately the same distance from league average versus the previously established healthy level (1992) as was the regression to his post-peak 1997-2001, before his decline began.  Yes, he was old when he hit all of these points - as were many other great players. 

I fail to see any kind of convincing “evidence” of PED use.

Interesting semantical argument. And that is all it is semantical.

There are two arguments:

One, is the argument over whether a statistical inference, like, “Blacks as a whole are more athletic than Jews, therefore so-and-so (a black person) is more likely to be an athlete than so-and-so (a Jewish person),” evidence that the black person is an athlete and the Jewish person is not?

Semantical argument.

Two, is whether it is true that black people, as a group, are more athletic than Jewish people. I think it is true, but I could be wrong.

Colin is arguing about #2.

Most of the people in this thread are arguing # 1.

I agree with some of the posters above that the issue is whether it is “fair” (rude, etc.) to use statistical inference to “accuse” someone of something.

Arguments like “Blacks commit more crimes than whites, therefore this black person is more likely to be guilty than this white person,” are not allowed in court not because they are not “evidence.” It is because the inflammatory/prejudicial nature of the argument outweighs the evidentiary value.

I agree with you regarding “convincing” evidence if convincing means that one changes their opinion to believe that he did use PEDs.

All of my comments are in response to the idea that there’s absolutely no reason to have any suspicion that Edgar used steroids.

I also take issue with the description of Edgar as this huge guy.  He’s probably 6’ tall, maybe a tad more, and always had a fairly thick build that aged just as I would expect that build to age

Edgar was definitely not “thick” in his early MLB years, but the opposite.

Edgar is listed at 6’0 175 in his earlier MLB years.
http://shop.sportsworldcards.com/ekmps/shops/sportsworld/images/seattle-mariners-edgar-martinez-645-donruss-1989-mlb-baseball-trading-card-38648-p.jpg

Chase Utley is 6’1 200. Chase Utley has to have the smallest waist for a power hitter since Eric Davis. i.e., he’s a slightly built guy.

Now, I do not know what Edgar was listed at during his aging years, but he added significant bulk to his legs and hips (which generate power). I’ve seen Edgar listed at 210, but IMO he looks more like 220-225. He’s solid, particularly in the lower body and neck.

MGL: I would argue the courts don’t allow it because the merit of the argument is not specific to the individual.

That simply applying Bayes doesn’t amount to evidence against an individual.

A blindfolded jury member, after hearing all the arguments, is 95% sure that the accused is guilty.  This jury member has a 97% threshold.  He then takes off the blindfold, and sees a black person.

Bayes would then suggest that this person is 98% guilty, if the jury member thought he was 95% guilty, race-neutral.

I don’t see how this is “evidence”.

Pete I don’t know that I’d necessarily view Edgar’s performance in 95 and 96 as simply a “player’s peak”, especially considering those years are when the steroid era essentially gets it’s start, building up to the explosion around 98-00.

Players in his division were using steroids in the late 80s, and if the reputation is earned, Texas Rangers in the 93-94 years. The transformation Juan Gonzalez underwent leading to an incredible offensive peak don’t really work in Edgar’s favor.

The AL West seems to be the “birthplace” of baseball players using steroids.

Probably not by coincide Jose Canseco was with Oakland in the mid/late 80s and even less surprisingly TEX in the early 90s.

There’s no way other players/opponents wouldn’t notice.

Edgar’s career arc is not that unusual except that the front half (pre-age 27) isn’t there

This is like saying “Randy Johnson’s pitching ability was not that unusual, except for his ability to strike hitters out.” True, but it kind of misses the point.  If Edgar starts his career 6 years later than usual, and yet still has a roughly typical aging curve, that would hardly be evidence in his favor.  Hitters don’t decline after age 27 because they’ve been in the majors for 6-7 years, but because of their AGE.  A late entry to MLB can’t explain why Edgar improved rather than declined in his early-mid 30s. 

Suppose that Edgar hadn’t been called up by the Mariners until he was 32, and posted the same stats but delayed by five years—such that he peaked at 37 and was still posting an OPS+ of 141 when he was 45.  Would you still say there was nothing unusual about his career, because the curve looked “normal?” Of course not.....

Funny you should bring up RJ.  He had a pretty late peak as well, no?  He was finally able to control the strike zone at age 29 onwards.

In what possible way does Jeter’s failure to fail a drug test after 2005 indicate use or non-use before 2004?  The tests don’t measure things in the past.  That argument has exactly zero weight.  A user who stopped would have the same results as a non-user.

"MGL: I would argue the courts don’t allow it because the merit of the argument is not specific to the individual.”

That is just the general rule of evidence. In order for “evidence” to be admissible, the evidentiary value must exceed the prejudicial value. As far as the definition of evidence in a court, it is any information or physical entity that impacts the likelihood of a defendant being guilty (or liable in a civil case).

I don’t think the color of someone’s skin would be inadmissible because it is not evidence.  Only that the evidentiary value is small. Statistical inferences that impact the chances of someone’s guilt or innocence are presented in court all the time (not enough actually, because many lawyers and judges don’t understand them).

"Facts” are not evidence in and of themselves. They become evidence only when embedded in an argument, that is, as part of the construction of a model, including assumptions, which purport to explain what we have observed. What was disappointing about Neyer’s blog posting is that he fails to make any clear argument. People here are trying to fill in the blanks for him.

I myself am not impressed by the career trajectory arguments, because we are still far from understanding the interplay of many variables involved. I won’t be persuaded by comparisons which are weighted with the careers of players from decades ago, because too many things have changed which impact recent players’ ability and opportunity to perform besides access to steroids.

Didn’t expect to see Regina Jacobs get mentioned here ... If I remember correctly like most athletes who test positive, she claimed she was innocent. But Jacobs is a nice illustration of how evidence and counter-evidence are embedded in competing models. What does passing or failing a drug test actually reveal about whether an athlete’s performance has been improved by past use of drugs, when the efficacy of countermeasures to evade detection are not known, and the likelihood of errors in testing are not known ...?

"Chase Utley is 6’1 200. Chase Utley has to have the smallest waist for a power hitter since Eric Davis. i.e., he’s a slightly built guy.”

B.J. Upton, or Soriano.

“In what possible way does Jeter’s failure to fail a drug test after 2005 indicate use or non-use before 2004?  The tests don’t measure things in the past.  That argument has exactly zero weight.  A user who stopped would have the same results as a non-user.”

Why is this so hard to understand?  I’m not proclaiming Jeter as innocent or accusing Edgar of anything.  Just making a simple statement of probability.  A player who has taken and passed a steroid test is less likely to have used than one who never took such a test.  Maybe only half a percent, maybe more, but it’s there.

If I call the period from 1985 (Canseco) to 2004 (last year before testing) the steroids era, then Edgar played more years in the steroids era than Jeter did.

Without knowing anything else about the player, a a baseline for whether a player did or didn’t would depend on what % of players are estimated to have used in this period, and how long the player played in that time.

This isn’t necessary for the argument above, but the fact that Jeter’s performance didn’t drop after 2004 indicates that:

A) He didn’t use steroids before testing either
B) He’s very good at concealing his steroid use from the testing process
or
C) Steroids don’t help you anyway

D) Steroids help other guys but not him.
E) Steroids did in facr help him, it’s just not blatantly obvious from the stat lines.

Judge: Mr. Hutz, w’ve been in here for four hours. Do you have any evidence at all?

Lionel Hutz: Well, Your Honor. We’ve plenty of hearsay and conjecture. Those are kinds of evidence.

I wonder if Edgar has the all-time largest proportion of his career value as an older player?  Seems possible.  I looked at WAR after age 32 (min 3000 PA), and here are the top 10:

Barry Bonds 77.7
Babe Ruth 62.4
Willie Mays 57.6
Hank Aaron 44.3
Tris Speaker 41.2
Ty Cobb 40.7
Ted Williams 39.1
Edgar Martinez 38.8
Stan Musial 38.2
Eddie Collins 37.6

Clearly, Edgar’s career total is the lowest on this list.  He produced 58% of his WAR starting at age 33.  The closest I can find is Downing, at 54%.  But I haven’t done a systematic look.

Hiz/44: darn it, I should have opened with that.

Guy, your research in your post in #45 provides me with the substance for the response I wanted to make to your post at #36.  The reason that Edgar’s total WAR is the lowest of that group isn’t that he had this shitty first half of his career, it’s that he had that career in AAA while similarly talented players were establishing themselves in MLB.

My point was, among great players (like those you list in #45), it isn’t that unusual to perform at a high level through the late 30s, even early 40s.  His performance at those ages is not unprecedented, at all.  What is unusual is for a player like that to not play regularly in MLB until 27.  If Edgar had had a crap MiLB career, I would completely agree with your point.  But he didn’t - he had an OUTSTANDING MiLB career, and together with his performance after he became a MLB regular, is a pretty good indication that his career arc is within what we’ve seen before from great players.

@CircleChange11, #35:  1995 and 1996 were Edgar’s peak years, whether you measure that by OPS+, ISO vs. league ISO, or wRC+ (actually 1992 is a tick higher in RC+ that 1996, but...).  OF COURSE that coincided with the advent of steroid use in baseball.  His peak could have occurred at any time in the 1990s and there would be some kind of argument to be made that it coincided with the advent of steroid use.  But IMO there has to be something MORE than that to rise above the level of coincidence.

And, as far as body type and physical changes, sorry, I just don’t see it.  Edgar was never a guy anybody would have mistaken for Eric Davis’ body type (or even a young Barry Bonds or Mark McGwire).  Yes, he was thinner, but (at least to me) he always had that build that looked like he was going to add bulk as he aged - as he did, but never at any kind of alarming rate of change (like Bonds, or McGwire, or even Bret Boone, about whom that is just about the only evidence outside of a mention in Canseco’s book).

My basic retort to those who say “Edgar had a huge second half to his career and that’s suspicious” is, would you please show me where that first half is?  Because it is mostly in AAA, to the tune of .344/.450/.495 over 1159 PAs.  Once he began playing regularly, he played well but at a lower level than he would in his prime for two seasons, broke out in his third, and except for two seasons plagued by injury (and strike-shortened) built upon and then maintained the level established in that break out season for another 7-9 seasons.  That is not an unusual career arc, except for the age at which it occurred, but even that (sustained high level performance through the 30s) is not unusual among great players. 

While it may “evidence” in the sense that it describes facts which might support a theory, it has so little evidentiary weight or probative value as to be nearly useless and is completely unfair to Edgar - more so than even the shit being thrown around about Bagwell.

Edgar was a late bloomer.  He should have been given more of a shot a bit earlier, such as 1988-89.  He was a better option than playing Jim Presley.  But I’m not convinced he missed out on great seasons there.  In 1989, he had more playing time in the majors than he did in the minors, and didn’t hit much in the majors.

While his batting average was .344 in AAA, that was in the PCL, and he didn’t hit for any power.  My guess is if you bring him up midway through 1987 he gets the adjustment struggles out of his way, then gives you an extra 2 years at the front of his career hitting .280-.300, getting on base a lot, 5-10 homers, while playing first base.  He’d still have the great bulk of his value at the back end of his career.

I don’t think this has any bearing on whether he was a steroid user or not.  It’s not exactly uncommon among high-schoolers, so to single out guys who played better later in their careers doesn’t make a lot of sense.  He was just a late bloomer.

My basic retort to those who say “Edgar had a huge second half to his career and that’s suspicious” is, would you please show me where that first half is?  Because it is mostly in AAA, to the tune of .344/.450/.495 over 1159 PAs.

That’s a fine hitting line.  But he was in AAA, he was age 24-26, and he didn’t hit for power (11 HR per 150 G).  This is not a “usual” performance for someone who, from age 32 on, was one of the 7 or 8 greatest hitters in baseball history.  It is much, much, much worse.  Look at the guys most like him after age 32:  Williams slugged .588 in AA, from ages 17-19 (!).  Musial slugged .626 in AA (age 20). THAT is what great hitters typically do as young men.

I don’t know if Edgar used PEDs.  But I do know his trajectory was highly unusual. 

I don’t think this has any bearing on whether he was a steroid user or not.  It’s not exactly uncommon among high-schoolers, so to single out guys who played better later in their careers doesn’t make a lot of sense.  He was just a late bloomer.

Rally, what do you mean by “among high-schoolers?” Edgar bloomed at age 32.  In what sense is that not uncommon?

I agree with nearly everything you said there, Rally.  Edgar didn’t really take full advantage of the early opportunities he DID get at MLB, but even then he was outperforming Jim Presley and his performance at AAA showed he really had nothing more to gain/learn by going back there as much as he did.  I think if he had been given more regular time, he would have performed better, and had a couple of seasons worth of additional MLB PAd at essentially the level you describe.  That, plus another 300 or so games at 3B would make a big difference in the quality of his HoF case for the DH biased and those who highly value traditional counting numbers for Hall consideration.

The only minor quibble I have is with your statement that he didn’t show power in AAA.  He didn’t show the kind of power he would have later (which is quite common), but he showed SOME power.  At 25-26, he was bouncing back and forth between AAA and MLB, but he had 11 HR in 444 ABs over those two seasons, and 34 other XBH on top of that. 

If he’d gotten the earlier chance, I’d have pegged him as a guy more likely to post 10-15 HR than 5, and to hit 20-30 doubles.  Not terribly unlike what he did in 1990-91, maybe a tick lower across the board.

What I agree with you MOST about is that there is nothing about his career arc that has any bearing on the question of whether he may have used PEDs.

Guy, Edgar didn’t bloom at 32.  He bloomed at 29 (1992: .343/.404/.544), after approximately 1300-1400 MLB PAs, and he had two very solid (if not quite Hall-worthy) seasons before that (1990: .302/.397/.433, and 1991: .307/.405/.452) at 27-28.

You keep ignoring that earlier performance and saying he came out of nowhere in 1995.  The only reason it even APPEARS that way is because of his 1993 and 1994 years, which are easily and entirely explained by injury.  1995 doesn’t look out of place as a follow-up to 1992, it only looks out of place because of the injury-plagued 1993 and 1994 seasons....

"Rally, what do you mean by “among high-schoolers?” Edgar bloomed at age 32.  In what sense is that not uncommon?”

It is uncommon.  But I don’t think by itself, it’s a red flag for steroid use.  Pete Incaviglia was a dominant high school and college hitter, had a few decent years in the majors, and faded quickly.  I don’t think he’s any less likely a steroid candidate than a late bloomer like Edgar.

How come we’re not talking about Carlton Fisk yet?

"At 25-26, he was bouncing back and forth between AAA and MLB, but he had 11 HR in 444 ABs over those two seasons, and 34 other XBH on top of that.”

In the Pacific Coast League, that’s not power.

Go look at what immortals like Brandon Wood and Sean Rodriguez hit in that league.  Or any number of players.

I don’t know what Calgary’s park factor was, but the team led their league in HRs both of those seasons.  And Edgar was not among the team leaders. 

Rally, I agree Edgar’s weird trajectory is not a red flag “by itself.” But it’s not by itself.  We now know that several other comtemporary players who also improved their power production relatively late in their careers—Bonds, McGwire, Palmeiro (I would add Sosa)—used PEDs.  I think it’s reasonable to suspect use by players in this time period whose late career performance is far better than expected (e.g. Luis Gonzalez)—although that wouldn’t keep me from supporting anyone for the HOF. 

I don’t get your Incaviglia comparison.

Tango, what’s your point on Fisk?  His strong age 40-42 performance?

But Guy, how do you establish any kind of MLB baseline for Edgar from which you can reasonably say that his late career performance was “better than expected”?  I think looking at the totality of his career, he established a good level of performance in his first 1000 or so PAs as a MLB regular, then took off, and thereafter maintained a solid and very consistent level of performance from his take-off point (1992) with the exception of two injury-plagued years, until he fell off a cliff (2004) and quit.

I don’t see how it was “better than expected” unless you somehow see 1993 and 1994 or pre-1990 as his true talent level - which I’m sure you aren’t claiming.  1992 doesn’t look out of the ordinary following his 1990 and 1991 seasons, 1995 seems like a natural “peak"/progression from the level established in 1992, and everything after than is about as consistent as anything you’ll ever find.

And I should have added, with the possible exception of Palmeiro, all of the players you mention who had these late-career improvements (and later were exposed as PED users) all had a previously well-established level of performance (and in some cases, a decline from that level not explainable by injury, or at least not entirely) before hitting a new, higher level after the time of their PED use. Look at the Cumulative WAR by Age graph here:

http://www.fangraphs.com/graphsw.aspx?players=1109,302,1008559,1086

Pete:
Let’s try it this way.  We’ll ignore his age 30-31 seasons, per your concern about his injuries.  Try to find me either:

1) players who were roughly as good as Edgar from age 32 on, but who were not much, much better than Edgar in their 20s.

OR

2) players who were roughly as good as Edgar through age 28 (adjusting his minor lg stats accordingly), and who became anywhere near as good as Edgar from age 32 onward.

My guess is there are very, very few players—if any—who meet either set of criteria.

If you adjust that to 29 (the year he really blossomed, not his age 32 year), and if I can find the time, I will take you up on it.

I don’t doubt that you are right, though.  Edgar’s skillset is a fairly unique one.  His excellence is found more in the breadth of things he did well, but not exceptionally well (except for OBP, perhaps).  And I’d expect just about any hitter who demonstrated the ability he showed at AAA would have been called up and become a MLB regular far sooner than 1100 AAA PAs - which isn’t Edgar’s fault, really.

Off the top of my head, Lance Berkman got a fairly late start, IIRC, and had a damn good year last year at what, 35?  He’s got a ways to go to match Edgar, but might.

Here’s one that fits:

Dolph Camilli came up at 26, became a regular at 27, and posted OPS+ seasons right around average until age 29, at which point his seasonal OPS+ went 161, 169, 137, 144, 145, 163, 143, 132....  It’s kind of hard to tell the quality of his minor league performance, because there is data missing, but he appears to have had a good, but not great minor league career.

So the syllogism seems to be:

A) Unprecedented growth curve
B) Played in steroid era
C) Therefore:  steroids!

I don’t buy it.  For one thing, at the very left end of the bell curve of talent (I’m at the right end), it’s absolutely uncertain that every spot has been taken.  Absent the DH rule, steroids or no steroids, Edgar is an oft-injured pinch hitter/3B/1B, Lenny Harris without the defensive flexibility, after 1992.  With it, he leads the league in games played in 1995 and is reasonably durable through 2001.  Add in two expansions, changing ballparks, divisional realignment and changes in schedule, and it’s not surprising you can’t find reasonable comps for a player of his magnitude.

Derek Jeter has no reasonable comps either.

Breadbaker:  No one suggested “C: steroids!”, only that A + B = some reasonable suspicion of PED use.  I don’t personally care whether Edgar used PEDs (other than that I’m curious to know who was helped by them and by how much).  But I don’t see any way you can put the probability much south of, say, 20%, and probably higher than that.

What does his playing DH have to with anything?  And in what way does Jeter have no comp?

Pete:  Camilli seems like a reasonably good comp. Pretty dramatic change in approach: at age 28 he had 65 BB and 113 Ks, then at age 29 he improved to 116 BBs and 84 Ks.  No change in power, but big BA jump.  Maybe he’s the first PED user in baseball!  Seriously, seems like a guy who really figured something out. 

I would just say 2 things: 1) with the coaching today’s players receive starting at a young age, it’s even less likely (but possible) that a player would suddenly “figure it out”; 2) it’s likely harder to “figure out” how to hit HRs than how to distinguish balls and strikes.

Guy, or “no one” as you refer to yourself, you put a number on the probability of something about which you have no idea at all.  What steroid?  When?  For how long?  Was it during the time he was fathering three children? 

If you don’t get the DH argument you don’t seem to understand baseball history.  Edgar got injured in spring training 1993 and every time he tried to get back in the lineup for the next two years he got injured again.  So he became a full time DH, an option unavailable to his potential comps before 1973, who either became pinch hitters or insurance salesmen. 

Per B-R, Jeter’s closest comp number for his career is Roberto Alomar at 847, and the closest shortstop is Barry Larkin at 778, neither of which is a close comp.  From his age 26 season, he has no comp closer in a season over 901.  Last year, it was 786.  Through his age last year of 37, there is no career comp that was a shortstop his entire career.

This reminds me of an exchange in “A Civil Action,” a non-fiction book (and John Travolta movie) about a cancer cluster in Woburn, Massachusetts.  The plaintiffs were accusing a large industrial company of polluting the groundwater, leading to the cluster, but the scientific link between that type of pollution and type of cancer was not clear; the plaintiffs had to prove it.  At a pre-trial hearing, the judge tells the plaintiff’s lawyer that he hopes he isn’t going to try to get away with statistical inference in his courtroom; he had to present actual evidence.  In other words, it wasn’t enough to show that there was a high probability of the pollution causing the cluster “just on the basis of some expert’s speculation.” And the lawyer replies, “Absolutely not, we’re going to prove our case on epidemiological grounds.”

There is a difference between “evidence” and “reason to believe.” A fairly narrow range of things counts as evidence, but we have all kinds of reasons to believe all kinds of things.  I believe that my mom is the best mom in the whole world.  I haven’t done any studies or read any research or talked to any experts, but I believe it as strongly as I believe anything.  My experience counts as a reason to believe the proposition, even if it isn’t really evidence.

For me, this distinction is important in daily life.  It means that I don’t have to subject all my beliefs to rigorous analysis and testing.  I’ve tried to do that and found it onerous to the point of debilitation.  Other people’s psychologies are (it seems to me) constructed differently and do not find that effort so overwhelming, but I think they are the minority.  I also don’t believe it’s normative, i.e., that we *should* all strive to reach the point where we can subject all our beliefs--about who serves the best breakfast taco, whether Oswald acted alone, or the revolutionary nature of the Higgs particle--to that kind of pressure because I don’t think it’s possible and the effort will frequently damage.  Rather, there is a middle range of appropriate perspectives; it would be maybe even more bad to subject *none* of your beliefs to the pressure of analysis.  For me, there is a very narrow range of things I care and know about, and my beliefs on those topics are highly organized, consistent, and tested to the best of my ability, but for everything else it’s catch as catch can.

All of this is prelude, though, to the fact that it is ludicrous to believe that you can find even good reasons to believe, let alone evidence, that Edgar Martinez did or did not use steroids.  Even if you find one side of the aging curve or body type argument more persuasive than the other your confidence in any belief that develops from that has to be extremely low in order to remain rational.  You often--usually, even--just don’t know, even in fields in which you are knowledgeable and devoted. (In this context I have no idea what Neyer is saying; our gut reactions are evidence for own prefab beliefs? Tautological. Other people’s beliefs are? A poor practice, and very un-Neyer-ish.) We aren’t on juries and we aren’t required to have opinions about everything.  In the context of the Hall of Fame, an argument how known steroid users (Bonds, Clemens, etc.) should be treated and how everyone else from that era should be treated is appropriate.  An argument about whether someone like Martinez or Bagwell used, though, while perhaps interesting, has no relevance to a discussion of the Hall because there is no possibility of reaching a rational conclusion.

Although I enjoy them all, this is one of my personal favorites:
http://joeposnanski.si.com/2010/01/19/iron-fisk/

"Per B-R, Jeter’s ....  Through his age last year of 37, there is no career comp that was a shortstop his entire career.”

Just because you stand between the 3B and 2B does mean you are a shortstop.

Rally, I agree Edgar’s weird trajectory is not a red flag “by itself.” But it’s not by itself.

That’s the issue. Each aspect on its own is not all that compelling. Is any single piece of evidence alone ever all that compelling? You can take essentially any topic/item and divide it into small aspects, and each individual aspect is not compelling evidence. It’s only the collection of the aspects that is compelling.

Even in a murder trial, one single bit of evidence isn’t compelling? The alleged murderer knew the victim? Well, lots of people did. The alleged murderer owned a gun? Well, lots of people do. The gun was the same caliber that was used in the crime? Well, lots of people have that same gun. The alleged murderer was seen near the crime scene earlier in the day? So, lots of people were seen in that area. The alleged murderer was angry at the victim? Well so were 2 other people. And on and on.

- Late career; unusual path.
- Physique change
- Steroid Era
- Late developing power

Now, we can go by these one by one and probably find a comparable person in history that did ONE of these. But, that’s not the point. But, that so few have done what he did is what makes it “unusual”.

Unusual does not mean guilty, and I want to make that certain. Unfortunately for Edgar, unusual career path combined with physical change combined with steroid era leads to likely suspicion.

My comments have all revolved around the idea that there’s “enough smoke” to wonder if there is a fire.

Edgar had an unusual career path during the steroid era, along with the physique change. It would only be strange, IMO, to not have suspicion given what we know about others during that era that had unusual performances.

To me, the issue with Edgar is more “DH” than “PED”.

BDF/65: I enjoyed that post, and want to make clear that I am NOT suggesting there is good reason to believe Edgar Martinez DID NOT use PEDs so much as I am trying to refute the idea that there is good reason to believe that he did. 

My own view on PED use in the steroid era and how it should effect HoF voting is evolving.  I used to think (as I think Rob Neyer still does) that it is unreasonable to ask voters to ignore what they know or even what they THINK they know, but that there are gradations of “discounts” (if you will) that should be attached to a candidate based on whether we KNOW they used (admitted or caught), strongly suspect with good evidence to back that up (Mitchell Report), suspect based on believable innuendo and hearsay (some would say Bagwell or Piazza fit this), and then those who the suspicion is based on little more than the coincidence that their career coincided with the steroid era and they hit home runs and had a muscular physique.  The higher up that ladder you go, the bigger the discount.  And then, in the case of those you believe were clean, give them a little “boost” if and when a tie-breaker is needed or a comparison with someone you suspect is close.  I think a fair number of voters still vote this way.

I’ve come around to the point of view that, outside of those who’ve been caught or admitted PED use, it is simply impossible to know with any reasonable degree of certainty who used and who didn’t, and what effect PEDs did or did not have on performance.  My system of “discounts” and “boosts” is therefore not much better than purely arbitrary.  The better practice, I now think, is to ignore the question of PED use, measure each player against his peers within the era, and elect those who performed the best. [The possible exception might be known/admitted/caught users, at least if a voter strongly suspects that player could not have achieved a HoF level career without PED use - which leaves room to elect guys like Bonds and Clemens while potentially barring the door for a guy like McGwire.]

But guys like Edgar - or even Bagwell, about whom I think the “evidence” is much stronger - the whole question has or should have little to no bearing on their HoF candidacy and the whole debate is unfair to them based on the kind of “evidence” people throw around.  Both of these guys have good HoF credentials, in my view - Bagwell’s clearly better, and Edgar closer to the borderline for legitimate questions about career/peak length, counting stats, and the DH.  I think there are good answers/rebuttals to those concerns about Edgar, and he is comfortably over the bar, but reasonable minds may differ on that.  I am OK with that; I am not OK with BS accusations of PED use based on the flimsiest and most speculative bases derailing his candidacy.  And, more than anything, what upset me about Rob’s piece was that he is a respected saberist with a national following who is helping to open the door to that.

This has been a fun discussion, and I know that those I’ve argued opposite of are not screamin “Edgar’s a USER! He should be banned from the HoF!” from the rooftops, and probably would even agree with at least some or even most of what I wrote above.  We can agree to disagree about whether PED suspicion surrounding Edgar is reasonable or not, but I hope we can agree that even bringing the subject up in the debate over his candidacy (or other, similarly situated candidates) is damaging in a grossly unfair way.

And largely pointless.  The era is what the era is, much as the 1920s and 1930s are what they were because of rule changes, the “lively ball,” and lack of integration.  Elect the best of the era, and be done with it.

Pete L./69: Thanks.  My view on all this is evolving, too, owing to the passage of time and also to Tango.  I understand the logic of both your and the Rob Neyer position, but I don’t think it will satisfy the deeply outraged. (Is Neyer deeply outraged?  I didn’t think he was, but I don’t know anything.) The reason the deeply outraged feel deeply outraged isn’t that, “Well, Manny Ramirez, is 69 WAR player, but he’s probably only about 48 without the ‘roids, so he’s out.” The “crime,” from the perspective of the deeply outraged, is worth way more than the 21 WAR difference, if that makes sense.  It’s like a guy who breaks into your house and steals your wife’s great-grandmother’s pewter wedding ring and fake family pearls: You want something worse to happen to him than if he just took your TV *even though* the TV is worth more.

There’s also the fear, even among the non-deeply outraged, that they will elect someone whom they suspected (like Bagwell) only to have their suspicions confirmed years down the line, after his induction.  That’s holding back a lot of trigger fingers, which I think a lot of people think is irrational, although I’m not sure why given the underlying aversion of these people to steroid users being in the Hall.

I think it makes more sense either to ignore steroids entirely and vote in Bonds, Clemens, McGwire, Sosa, Thome, Bagwell, Thomas, everybody, based entirely on their on-field accomplishments and whatever other criteria you’ve always used.  It also makes sense to keep (almost) everyone out NOT because they might have used but in some vague, amorphous way participated in and contributed to a culture you feel was destructive to baseball in some abstract, historical, ill-defined way.  But I can’t get any of the middle roads make sense yet.

BDF/70: I am sure you are right about the depth of emotion on this subject - especially about the guys who were caught (like Manny, who would be a no-brainer but now I doubt ever gets in as a guy caught not once but twice in the era where there WERE rules in place, and reportedly also a third time as one of the 104).  But it doesn’t make sense to extend the moral outrage much further than that, because it can’t really be based on anything anybody can no with any reasonable degree of certainty (and “certainty” is not really the right word; maybe even something closer to “suspicion").

Logically, I think the two possible paths a voter might take that you describe in your last paragraph makes sense.  Practically, the second path (keep everyone out because they either used or enabled the users) can’t happen.  The Hall can’t let it happen, for economic reasons, and shouldn’t let it happen because it is first and foremost a historical museum, and pretending that this era didn’t happen is not good for the sport or the museum.  That’s why I’ve concluded that the only remaining reasonable - if still uncomfortable - solution is to ignore PEDs and vote in the best of the best.

"It’s like a guy who breaks into your house and steals your wife’s great-grandmother’s pewter wedding ring and fake family pearls: You want something worse to happen to him than if he just took your TV *even though* the TV is worth more.”

If that’s what writers think about steroid use, then they need some psychological help.  Seriously.

That’s why they are The Holy Writers.  They are infallible according to scripture.

Baseball’s punishment for steroid use is 50/100/etc game suspensions. No part of the suspension is prohibition from the Hall of Fame.

IMO, the baseball writers are taking it upon themselves to implement a punishment that does not exist, or did not exist.

My opinion has evolved to the point that even those that admitted use or tested positive should still be eligible for the Hall of Fame, both formally and informally (and I don;t rally “like it” that I feel that way because generally I am a “1-strike and your out” type of person when it comes to cheating that requires this much conscious decision and planning).

So, by extension those that are rumored or suspected of steroid use are eligible ... and when I say eligible I mean “without deduction”.

My preference is that grown men recognize that what they are about to do is both illegal and against the rules of their organization and don;t engage in the behavior regardless of whether there’s testing or not. The reality, as a former athlete, is that many athletes are going to exploit any advantage they can to be successful ... either as a team or an individual. Once you’re convinced your opponents are doing it, you can convince yourself that you don’t even have a choice (and there’s some truth to that), if you want to compete.

I think we’re going to have to come to grips with the idea that PED use was possibly more prevalent than just the guys that got caught or have admitted and that it was rather commonplace.

Right, I’ve argued that in the past.  If MLB decides to give an 8-game suspension for using an emery board, then that’s how bad they think it is.

Writers, if they really feel like they do, should turn their backs on Ryan Braun, and never write another thing about him, where they are paid to write about him.

The reality is that The Holy Writers are huge hyprocrites.  They are not in a position to judge anyone.

Pete L./71: We may not know who used what and when and how much it helped, but I’m pretty confident in saying that many otherwise HOF-worthy players used, that the vast majority of baseball people knew what was going on in a general sense, and that they said nothing.  This would be the basis of the exclusionary condemnation.

Rally/72: I see disagreeing with the claim, but I don’t see relegating its believers to the mental illness bin.  Baseball is deeply important to these folks and they feel that steroid use damaged it irreparably and so they’re viciously angry at steroid users and want to keep them out of the Hall as a means of both protecting the game and the Hall as well as punshing.  That may not be the best, most productive way to feel/think, but I don’t think it’s close to mental illness, even accounting for any hyperbole on your end (although you did say “Seriously").

CircleChange11/74 and Tango/75: I may be coming around to this position, too.  I was thinking deeply last night, fairly loaded, about Barry Bonds (whom I loathe and resent as deeply as it is possible to loathe and resent someone you’ve never met and never will) and how on a very real level it is absolutely ludicrous to have a Hall of Fame that he is not in.  I mean, the guy is the best baseball player ever.  He’s not Pete Rose, he’s not S.J. Jackson--he’s the best there ever was.

But, Tango, your anger at the blowhardy hypocrisy and clumsiness of the HWs bleeds into your ability to assess their arguments/feelings.  They are, for the most part, not setting the world on fire with either their prose stylings or synaptic sophistication, but that doesn’t mean there is not merit to their position.  They may not deserve to be the ones filling the role of the decision-maker, but, well, there you are.  Maybe the way it will all shake out is OK.  Maybe the writers will never elect Bonds but the VC will and that delay is enough both to punish Bonds and to “protect” the game.  Maybe the simple outrage of Bonds going in on anything other than the first ballot is enough.  Even fifteen years into this situation it all still feels fairly fluid to me.

"That may not be the best, most productive way to feel/think, but I don’t think it’s close to mental illness, even accounting for any hyperbole on your end (although you did say “Seriously")."

No hyperbole on my part.  I don’t know if the writers take it THAT seriously.  Just denying a “yes” vote on the HOF doesn’t prove anything.

But IF they feel personally victimized (comparable to a robber breaking into their home and taking something of personal value) by some ballplayer’s choice of what to put in his body, then yes, I call that mental illness.

Rally/77: Many of the writers do take it that seriously I’m pretty sure.  That’s why Tango keeps making the religious comparison.  Hell, *I* take it that seriously and it’s not even my career.  Baseball in the abstract, Platonic sense is just incredibly important to a lot of people.  It just is.  Of course this is goofy, but comparably goofy to all the other goofy things people choose to care about immensely.

That having been said, yes, it is an imperfect analogy, as all analogies are (by definition).  A sense of personal victimization would bleed into mental illness, I agree.  But I meant to emphasize just the asymmetric weights of the situation, not the personal aspects.

The idea that this group of players is not worthy of an honor, that’s opinion.  One I don’t agree with, but whatever.

Feeling a need to punish these players - that I think is unhealthy.

There was a huge football player (LT? Reggie White?)… someone big name like that.  When he was up for the Hall of Fame, one of the writers said he’s going to judge him based on what he says about drugs, if he’d apologize, etc.

Anyway, this football player did not apologize, and though he should have been unanimous, he got 31 of 34 votes (or something like that).

Football writers are at least realistic.  They know how the world works.

Baseball writers?  Ugh.

There have been many HOF ‘roids discussions on this site and this, IMHO, has been the best, particularly thanks to the contributions of BDF and Pete L. It would be great if some of the members of the BWAA could thoughtfully partake.

I was in the “big discount” camp prior to reading these posts and thanks to #69, “Elect the best of the era, and be done with it,” and the arguments presented backing that up, I’ve changed my mind. (Does that actually happen on the internet?)

I also think it would be great if the HOF/BWAA could be honest about the era. I would hope they could say that Bonds and Clemens were among the best hitters and pitchers in the *history* of baseball, but they played in a transitional era were things were pretty complicated. I think that that would be most fair to the fans and to the game and to the players from other eras.

So, elect the best of the era, but make sure it’s clear what this era was really about. Then, let the perspective granted by time assess steroids in the larger context of greenies, deadball, integration and whatever next inevitably comes down the turnpike.
To claim some ability to have that perspective now - as many in the BWAA are doing - smacks of egotism.