And that was a very different era. A comparably great player today like say, Jeff Bagwell, gets 41% on the first ballot.

Carl Yastrzemski ended his career with 3,419 hits, 452 HRs, he won a Triple Crown, he made 18 All-Star games, and he won 7 Gold Gloves. Jeff Bagwell ended his career with 2,314 hits, 449 HRs, 4 All-Star appearances, and 1 Gold Glove. If you think these two resumes are comparable to BBWAA members, then you clearly don't understand how BBWAA members judge baseball players.

Carl Yastrzemski ended his career with 3,419 hits, 452 HRs, he won a Triple Crown, he made 18 All-Star games, and he won 7 Gold Gloves. Jeff Bagwell ended his career with 2,314 hits, 449 HRs, 4 All-Star appearances, and 1 Gold Glove. If you think these two resumes are comparable to BBWAA members, then you clearly don't understand how BBWAA members judge baseball players.

True, but if you think 90's/2000's players are going to be treated fairly and given as much support as they deserve by the BBWAA compared to past generations, then you clearly haven't been paying attention to the voting and written articles over the last few years.

Pedro's WAR in his two best years, 99-00, is 18.5. And WAR relies a lot on being above a low replacement level

If you are using bWar, the replacement level is silly high, not low.

if you think 90's/2000's players are going to be treated fairly and given as much support as they deserve by the BBWAA compared to past generations

They're not going to be "fair" to "known" steroid users - McGwire, Palmeiro. But their treatment of Bagwell is broadly consistent, if not better than, their treatment of, for example, Ron Santo and Johnny Mize, neither of whom the BBWAA even bothered to elect (and they'll be electing Bagwell within the next 2-3 years). I don't see why they won't be as fair to Pedro as they were to guys like Koufax or Gibson from the past. But, as I noted upthread, that doesn't mean they're going to be giving Pedro 95% ballot support: Bob Gibson was under 85% in his first (and only) year on the ballot. The BBWAA has ALWAYS been tepid in their support of guys with peak cases who don't hit major career milestones (and they've always done a fairly poor job of distinguishing well-deserving peak cases from less deserving peak cases - e.g., Dizzy Dean, Catfish Hunter).

To throw another name out there, if the BBWAA doesn't elect Mike Piazza in his first year on the ballot (and I doubt they will), it will be entirely consistent with their treatment of Yogi Berra, Carlton Fisk, and Gary Carter, none of whom were elected in their first year on the ballot, not indicative of the BBWAA being harder on Piazza than they were on catchers of the past.

They're not going to be "fair" to "known" steroid users - McGwire, Palmeiro.

What about suspected but never proven steroid users like Sosa? 609 homers including a 5 year peak of 66, 63, 50, 64, 49. An MVP, plus a 2nd place finish. Normally, 1st ballot all the way (by HOF standards, not HOM). But now he'll be lucky to get the 20% that McGwire is getting. Think that's fair?

What about Biggio? 3000 hits has been pretty much a 1st ballot lock in the past. He could make it next year, but I sure wouldn't consider him a lock.

What about suspected but never proven steroid users like Sosa?

That's why I put the word "known" in quotes. As far as the vast majority of the BBWAA is concerned, they "know" that Sammy Sosa used steroids. They will absolutely be unfair to Sammy Sosa. I agree with that.

What about Biggio?

Let's wait and see how they actually treat him, before we judge it fair or unfair.

Has anybody suggested that Pedro Martinez used steroids, though?

People remember Randy Johnson as clutch in the postseason.

I'm sure some people do. When I think of Randy Johnson in the postseason, my first thought is him getting blasted by the Angels in the rain. And then having another stinker against Detroit the next year.

And then you have post-season stuff (clear advantage Pedro).

Not seeing this at all. Pedro is 6-4, 3.46; Unit is 7-9, 3.50 - but Pedro stayed fairly consistent as the playoffs went along, and Johnson got significantly better in each subsequent (and thus more important) round. If you don't weight the games, they're roughly even, unless you think it's fair to blame Johnson for losing twice while giving up 3 runs in the '98 NLDS. If you do weight them by championship importance (I like to do this, but it's not obvious that you should), Johnson comes out ahead.

Pedro has Game 5, ALDS 1999. That's good enough for me.

(Also, I'm biased.)

Pedro has Game 5, ALDS 1999. That's good enough for me.

And Unit has this series, in which he outduels both Maddux and Glavine.

(Also, I'm biased.)

Fair enough.

I don't think either Johnson or Pedro has a predominantly positive impression in the minds of the writer's re: the postseason. Johnson was considered a postseason choker until the 2001 World Series (he entered the 2001 playoffs 2-6 in the postseason, with both those wins coming way back in the 1995 ALDS - since then he was 0-6), which despite his picking up three wins was and was considered/widely reported as the Schilling Show (Schilling won Game 1, then pitched very well twice on short rest, which became The Topic of the World Series). Pedro pitched pretty well but his team didn't win, and when they finally did in 2004 it was again the Schilling Show.

When it all evens out I don't think either guy really has huge a postseason reputation in one direction or the other at all. When both are thought of they're thought of for their regular season achievements, which is what they'll be elected on.

Has anybody suggested that Pedro Martinez used steroids, though?


Pedro Martinez used steroids. There, it's been suggested. That's about as much evidence as there is against Sosa, Bagwell, or Piazza.

But snark aside, you're right that we need to wait and see what happens starting next year. I'm just not holding my breath for any kind of consistency. I actually gained some respect for the BBWAA over the last decade when players like Winfield, Puckett (not a great choice by SABR standards, but considered worthy by most casual fans), Ozzie, Molitor, and Eck were all elected on the first ballot. All of them had valid HOF cases (again, by traditional standards), but none of them were the kind of player that would be considered "first ballot" by the old time purists. I was thinking maybe the writers were finally abandoning that first ballot nonsense and realizing that a HOFer is a HOFer. But then Sandberg, Alomar, Larkin, Bagwell, etc, were all forced to wait and I'm back to thinking the voters are a bunch of dullards. And possibly only getting worse, as we may or may not find out in the next few years.

#103 Unless I'm very much mistaken, replacement level is set at a level that would produce a 52 win team. At any rate, that's where Bill James set it when he came up with the concept.

Keith Woolner wrote a nice article about where replacement level "should" be, given a normal talent distribution and I'd assume the significant Seans are familiar with the article.

Another very practical way to look at it. Doesn't Andrew Miller's performance last year look like what you'd expect of a replacement level pitcher? Certainly does to me, and he's within spitting distance of replacement level.

Now there is something to be said for using what I've taken to calling VOTW (value over train wreck). These are the worst performances. Conceptually replacement players having a bad year. In practice every year the worst positions are more like 30 runs below average rather than the 20 or so that replacement level is set at. Completely predictable in that normal performance variation among regular players is in something close to 14 runs per 650 PAs. (I haven't checked what it is for pitchers. But they are in fact more volatile than position players)

What this boils down to is that players true talent level is an ERA+ of 75 (well Sean's working with earned runs. Still, a quick check of a few starters with a WAR of 0 gives an ERA+ in the range of 75. Good enough for the purposes of this discussion) will produce results all over the map.

As I said, not crazy to view value in terms of keeping the failures off the field. The problem in practice is that you end up asserting that a replacement level team (by the VOTW definition) is roughly a 37 win team and that's just too low given what we know about teams assembled from freely available talent.

attaching the blame for one of the greatest managerial blunders in baseball history to the pitcher is like blaming Dave Henderson for Bill Buckner blowing a groundball.

The reason that it was a blunder is that Pedro was unable to close out that game. That inability is a strike against Pedro, particularly given he was pitching on 5 days rest. By comparison, Randy Johnson was able to close out game 7 after pitching 7 innings in Game 6. If we're talking about postseason memories, Johnson has Pedro beat.

#16 Look, I think Smoltz is a clear HoF'er but lets use good arguments for this and not the above.

As one who is keen to learn, I'd be happy to see a better reasoned argument than what I presented in #9 above. I sometimes like shooting fish in a barrel, and have been known to go to sites like Bleacher Report to "set the record straight." And sometimes it works -- I was able to convince a writer over there that Tim Raines belongs in the HoF not long ago by presenting a good argument. So improving the argument for future use can only help.

Much obliged.

Should also note that I was able to get someone over at Bleacher Report to at least think much harder about their anti-Ron-Santo-for-the-HoF position.

Sorry to post yet again, but had another question to ask. Is there a repository (perhaps somewhere on this website -- I wasn't able to find it) of succinct and respectfully worded pro-HoF arguments for saber-worthy candidates? Something that can be cited and brought forward in what might be seen as less-educated forums than this one? The best way I can see to sway minds is to do so one at a time (more likely affecting an article reader than the author, but the latter's also possible), but having a more organized place for this might make the process more effective. This site is certainly saber-friendly, and might be a good venue for it.

Comments most welcome and encouraged.

Edit functions don't seem to take here. If it's of interest, found the websites for the pro-Tim Raines argument referenced above that worked:

http://bleacherreport.com/articles/1004993-2012-mlb-hall-of-fame-ballot-are-they-hall-worthy-or-not

and the Ron Santo one:

http://bleacherreport.com/articles/970827-mlb-ron-santos-hall-selection-yet-another-example-of-flawed-bogus-process

I'm "b bachslunch" over there.

I think the real thing that was wrong with your #9 was that you assumed earnestness on the part of someone who was being ironic.

Cumulative-WAR arguments are kind of shoddy for a lot of reasons -- they erase nuance, and hand-wave problems with WAR, for instance. The best argument for Smoltz -- and I imagine it won't really be necessary to present it too many times -- is that he's better than a lot of guys who are already in the Hall of Fame. But honestly? I doubt he'll need it. Maddux-Glavine-Smoltz were legendary in their time, and I suspect that Smoltz's bounce to the bullpen will ultimately be seen as a point in his favor, rather than the other way around.

#117 Basically I'd suggest that arguments be framed as arguing that a player is fully qualified as opposed to marginally qualified. Don't argue that he's better than the weaker member of the hall, but rather that he can be sensibly compared to those who everybody accepts is a fully qualified HOFer. I'd actually suggest arguing even more strongly. I generally don't support a player if I can identify 10 players who are clearly better, have been up for vote and are not in.

As to how to frame the argument, use good stats that have been adjusted for context. Need not be WAR or adjusted batting wins or OPS+. EQA is also not hard to find and is a generally good stat. SNWL is an excellent point to start from with starting pitching. WPA really only makes any sense for relief pitchers (though I think you can invoke it sensibly when arguing for extra credit for a given player on a team that substantially over or under-performed his pythags). Be prepared to explain why (for instance) you aren't interested in a player's rbi.

Now if you're arguing the case with somebody who isn't a stathead be aware that they'll assume you're cherry picking your argument (you'll hear a lot of things like "figures don't lie but liars can figure"). Meaning that you should be prepared to explain the foundation of your argument -- why you have to adjust for context and how to do it.

It helps if you understand what type of candidacy you're advocating. There are 3 primary types of HOF cases. Career (basically bulk WAR, or perhaps bulk oWAR and a letter grade evaluation of his defense), prime (basically career value but counting only those years where a player played like a star. Number of years matters too) or peak (nobody agrees on the precise meaning. I like Dale Stephenson's definition. Best 5 years. A lot of people want the years to be consecutive which is why for instance Sandy Koufax will be seen as having a stronger peak by some people)

Well there is a 4th type. The narrative case. For instance Johnny Evers' case was basically presented as: The 1906-1910 Cubs were the most dominant team of all time. Evers was their best position player (once you factor in playing time at any rate). Their best pitcher (Brown) was not the best pitcher in the league so Evers has to have been great. What's more, Evers won an MVP for a surprise pennant winner. Lou Brock and Jim Rice are semi-narrative cases too. It's really tough to defend them by any kind of value case.

You'll almost never hear a stathead offer a narrative case, and when arguing against one (see Jack Morris) they'll attempt to run tracers (fact checking) on the claims of the case.

Ichiro arguments are interesting in that they're part trivia, part narrative, part statistical (but supported by the least widely accepted aspect of the statistical case -- he has no statistical case without either accepting some form of Japanese credit or taking his defensive stats as BBRef gives them). In a sense the arguments aren't important because everybody accepts that he's going in as soon as he's up for vote, but they are very useful in that you get to see the assumptions laid out.

WPA really only makes any sense for relief pitchers (though I think you can invoke it sensibly when arguing for extra credit for a given player on a team that substantially over or under-performed his pythags).

WPA never makes sense when discussing a player's value, ever, period. End of discussion. Use WPA/LI.

#118 It's a bad argument. Or a silly definition of elite. First of all 200 extra games of high level matters. Second, a hall of fame case for anybody from the 60s using raw stats is lame. Even when comparing him to somebody who played at a similar time. Third, Boyer and Santo's careers don't precisely align. Boyer was done before deadest ball II, Santo played through it. And while everybody thinks of Wrigley as a great hitter's park (it certainly was for Santo -- he undeniably took extreme advantage of it), Busch Stadium generally was a better place to hit.

Still, put it all together and Boyer played in an offensive context where a league average position player could be expected to hit .269/.335/.412 and Santo played in one where a league average position player could be expected to hit .268/.332/.399. Which is why Santo comes out with an OPS+ of 125 and Boyer comes out with 116.

Even then, since OPS+ under-values OBP (conceptually OPS+ is OBP*1.2+SLG, park adjusted and then put into a scale where 100 is average. When Santo and Boyer were active OBP*1.6+SLG works better), slightly understating Santo's offensive advantage. What this boils down to is that even though Boyer has a fairly large advantage in the stuff not covered in the batting counter stats (baserunning, reaching on error and DP avoidance -- ~56 runs) Santo ends up with a huge offensive edge over Boyer. 123 batting runs in a context where fewer runs are required per win.

Yes, Boyer probably was the better fielder, but Santo wasn't a bad fielder. Boyer's simply not going to make much headway here.

That's the heart of the bulk career argument. Santo at his best was a much better player. His extended prime is 1566 games of a 137 OPS+ by a very good defensive player. Boyer only exceed a 137 OPS+ once. As for peak, Dale Stephenson has Santo with the 8th best offensive peak for a 3rd baseman -- and that's counting Dick Allen, Harmon Killebrew and Edgar Martinez among those with a better offensive peak. Boyer's at 27.

Now maybe a short career 3rd baseman with only the 8th best offensive peak doesn't meet your definition of a HOFer. His strongest case is actually extended prime. Mathews, Schmidt, Boggs and Brett are fairly clearly better. I'm not 100% certain that Brooks Robinson is, but I'm willing to stipulate it (mighty generous -- Oriole fan here. started following the team in the late 60s). Guys like Killebrew and Allen are complicated. Home Run Baker for a different reason (weaker competition). And there are a few guys now active who also belong in the discussion. Anyhow, Bill James gets Santo as the #6 third-baseman of all time, just ahead of Brooks. As I said, I can see an argument for reversing them, but that's about as far as I'd go (evaluating Allen as a first-baseman)

Now there are people whose HOF definition doesn't extend to (say) Gary Carter or Santo. But that's a more restrictive definition than is currently in use. He's clearly not an inner circle HOFer but he does match up well with any who aren't in the first cut.

Another way to look at it. Is there a clearly better third-baseman eligible who isn't in the hall? No. Are there any clearly better players not in (and eligible)? Not that I can see. There are players that you can argue for (and who I might well advocate), but that's not the same thing in my mind. (Another way to look at it, over the career length we're talking about, WAR's standard error is in the 3 to 4 win range. Whitaker and Grich have a higher WAR than Santo but not by enough to say they were clearly more valuable. And that's just bulk WAR)

Yeah there are Negro League players who belong in this discussion too. Nobody clearly better than Santo is not in (at least as best I can tell)

#121 Not at all clear to me. Yeah I know, type of team affects a player's WPA in very predictable ways and that a player doesn't get to choose his usage.

That said, WAR basically reconciles at a team's pythag rather than their actual winning percentage. Check that. In fact it's team offensive runs created (not James' RC, but the predicted runs scored given all of the known events) rather than actual runs scored (though runs allowed are used on the defensive side). So you've got a 13-14 run standard error on the offensive side and a standard error in the range of about 4 wins in the team total.

And the one thing we know that predictably affects pythags is relief pitching. Teams with really good bullpens tend to beat their pythags by about 2 runs.

All that to say that excellent pitching in high leverage situations has tangible value and straight WPA probably measures this better.

WPA/LI also annoys me in a sense. It's pretty clear to me that it's false precision. The initial estimate (ie straight WPA) isn't accurate to tenths. The LI calculation is pretty noisy too.

Even so, I accept that WPA/LI correlates better with how well a guy pitched than straight WPA and that no form of WPA has any predictive value (beyond that which is carried in the pitcher's basic stat line) and that as I said before, WPA is heavily affect by factors beyond a player's control. Straight WPA though is the best explanation of a relief pitcher's affect on a team's W/L record. It's of zero value in explaining how good he is.

WPA never makes sense when discussing a player's value, ever, period. End of discussion. Use WPA/LI.

absolutely agree.

If we're talking about postseason memories, Johnson has Pedro beat.

Well, only if selective people are doing the remembering.

I was discussing performance. Pedro's ERA, mostly vs DH-led teams, in the postseason, is superior to Johnson's somewhat-NL-based. The W-L record also helps. And as to Randy getting significantly better each round, wel that's great if your teammates carried you enough so you could produce better later on,but you're awful lucky still to be playing then later on,right? It's the !Reggie! syndrome, where people forget how poorly he played in non-WS playoffs.

I am going to add nothing new, but in every thread that this comes up I just have to say:

He's Pedro Freaking Martinez. This time three years from now he will be Pedro Freaking Martinez, Hall of Famer.

Damned right he is. John Smoltz and Curt Schilling were great pitchers. Rare pitchers. Hall of Fame pitchers, quite possibly first ballot Hall of Fame pitchers. John Smoltz had a fine peak. Schilling's peak was even a little better than that. And Curt Schilling's peak is only an average season in Pedro Martinez's career. That's how good Pedro Freaking Martinez is.

[whew--that was fun]

It will be an interesting case study to see if when Schilling goes on the ballot, he gets credit for his low unearned run total.

I don't think the BBWAA will pay any attention to this at all. It won't register with them any more than, say, FIP does.

Classic Pedro:
http://espn.go.com/page2/s/questions/pedro010509.html

It's pretty easy to explain why it's a skill to give up very few unearned runs. It's even easier with someone like Schilling who was historically good at it. I could see a writer or two using that idea for a column, and it picking up some steam. I can also see it being totally shut out.