hot topics

Top two shouldn't be difficult. Then three and four are Robinson and Ott (in that order for me, dunno what the rest of you think). After that it's trickier. Between Clemente, Reggie and maybe Rose for fifth?

You're forgetting Waner and Kaline and Gwynn

Here's the top 10, as I have it now:

1. Ruth
2. Aaron
3. Ott
4. Robinson
5. Rose
6. Waner
7. Kaline
8. R. Jackson
9. Clemente
10. Gwynn

Sam Crawford or King Kelly might possibly push into the top 10 at the expense of Gwynn and Clemente.

Evans, Winfield, Heilmann, J. Jackson, Slaughter are bunched next: I won't try to sort them out now.

Keeler, Flick, and Thompson bring up the rear, with Thompson a rather distant last.

Wow, this is easy after center field!

Preliminary ballot:

1. Babe Ruth
2. Hank Aaron
3. Mel Ott
4. Frank Robinson
5. Paul Waner
6. Pete Rose
7. Joe Jackson
8. Sam Crawford
9. Reggie Jackson
10. Elmer Flick
11. Al Kaline
12. Tony Gwynn
13. Roberto Clemente
14. Enos Slaughter
15. King Kelly
16. Willie Keeler
17. Harry Heilmann
18. Dwight Evans
19. Dave Winfield
20. Sam Thompson

King is the only one I'm not sure about. All, except Thompson, are solid HoMers IMO.

Is Rose going to be very polarizing? It seems that one's peak/career preference has a bigger effect on one's view of Rose than just about anyone else. I remember that Dan's original salary estimator, which was very peak heavy and didn't value in-season durability, had Rose tied with Scott Rolen!

Chris Cobb and Tib,
Just eyeballing it, F Robby and Ott look very close to me, but I would lean to Frank - mostly due to strength of league (or timelining if you prefer). Why are you guys leaning the other way?
Thanks

Is Rose going to be very polarizing? It seems that one's peak/career preference has a bigger effect on one's view of Rose than just about anyone else.

Rose was a fine player in his prime, not only a career compiler. Elmer Flick and Joe Jackson are prime candidates more than peak candidates. With those caveats, Flick and Jackson may give Rose a run on that issue.

Joe Jackson is capable of giving Rose a run on both accounts, "peak/career" and polarizing.
Many "credit" Earl Averill and Larry Doby for some play outside the major leagues. Many "debit" Joe Jackson for some play inside the major leagues.
"A pennant is a pennant" - how much did Jackson help or hurt in 1920?
And what is a World Series?

Although the gap between Ott and Robinson is nowhere near as large as the gap, say, between Aaron and Ott, it's large enough that I don't see a competition difference closing it. We've got pretty good MLEs for Ott's NeL contemporaries, and Ott still comes out as the third best position player of the 1930s, trailing Gibson and (narrowly) Gehrig and leading Stearnes. Robinson comes out as the #4 man of his era, behind Mays, Aaron, and Mantle. It's not hard to see Robinson as preferable, though.

OK, I might as well start the Pete Rose stuff going. Normally, I don't pay any attention to this kind of thing. For example, I will be making no deductions of any kind for Joe Jackson, and could, just possibly, be talked into adding on credit for at least some of the seasons he might have played after 1920, although that would be one tough argument. But Pete Rose is different, because I am certain that he was betting on baseball games involving the team he was managing - I mean, he's admitted it, if I read the story right. He then says that, since he only bet for the Reds, rather than against them, he's innocent of fixing games.

The problem is that the above is not true. Let's say you're Pete Rose and you're managing the Reds, and you're betting, and you're down $10,000 to Bobby the Bookie. Bobby comes to you and says, "Pete, you're into me for ten grand. But I'll let that go if you do this: The next game your ace is due to start, you don't bet anything. I get to bet against the Reds, but you get to keep your nose clean on that score. And then, the day of the game, you decide you need to hold your ace back a day, to keep him fresh or something, and then you start some long relief man on no notice, including to him. I'll make the ten grand back, and you're off the hook." Well, Pete can do that, and no one will really notice, and he hasn't bet against the Reds, has he? He can also bench, say, Joe Morgan for that game, because Joe "needs rest" or something, just to make as sure as possible that the Reds lose. He can also decide, if the aforementioned long reliever proves effective, to pull him after five innings, so he "doesn't get too much more work than he's used to." Then he can go to a worse long reliever to insure Bobby's bet.

But wait, it gets worse. Pete can also bet FOR the Reds a couple of days in advance, when his ace is due to start the game BEFORE the one he's betting on. Then he can hold his ace back a day, to the game he bet on. And that would constitute making money by cheating, while managing a team that was hardly trying to cheat him on his salary, which is Jackson's reason/excuse (I call it a reason, and think that it's inexcusable that Charles Comiskey is in the Hall of Fame. But your opinion may differ.).

That's the kind of thing that bothers me. Rose admitted to betting on Reds' games, but showed no sign at all of being able to recognize these two opportunities, much less others that I, who do not bet on sports, would never think of, but that career bookies sure will. That is, even if Rose truly was stupid enough to miss the implications, I guarantee you that his bookies were not. Since his bookies were no doubt quite eager to educate Pete in these things, I consider Rose to still be in denial about what he did, which means he still does not want the public to know.

This all started for me when Bill James published his long essay about how little evidence the original report had on Pete Rose. In particular, Bill noted that the so-called "betting slips" could not be anything of the kind. And he's right. There's no real evidence. If he had wanted, Rose could have sued organized baseball with me as his lawyer and won that one, if that were all the evidence baseball had (I am not a lawyer). And yet, Pete Rose accepted a deal that banished him from baseball forever. He's trying a sympathy ploy now (not that it's working), but the sympathy ploy is based on not recognizing the opportunities he had to manipulate games when a manager.

That's what bothers me. Rose is still representing himself as innocent by concealing things. And the more he does that, the guiltier he looks to me. On the other hand, it's all circumstantial. Part of the original deal was that baseball would never reveal what they really had on Rose. So we don't know. But I am absolutely sure that Pete Rose was not made to accept banishment from baseball on the basis of what was revealed. There has to be something more there, and it has to be a LOT more. It is absolutely NOT just that he made a couple of bets on the Reds out of a misplaced sense of loyalty or something.

So the question is whether it is constitutional, in any way, for a HoM voter (meaning me) to take any of this into account. I'm perfectly happy to abide by what the group decides, but I do want to raise the question, because, right now, I am so sure that there is a LOT that Rose doesn't want us to know that I am inclined to just discard every season he was managing and also playing. That, of course, would reduce his career value. I'm sorry to be the one who raised the issue, but since it would affect my vote, I figured I'd better do it now.

- Brock

Slaughter missed his age 27,28, and 29 seasons for WW2. He bracketed those seasons with 37 and 29 win share seasons, I guess the Charlie Keller contingent will argue that he is missing about 85 Wins Shares.

Ruth gains a little WW1 shortened season credit - I doubt he needs it :)

Quite few guys that have strike credit, I guess Dewey Evans being the biggest benefactor.

Early Thoughts:

1. Ruth



2. Aaron

3. Ott

4. Robinson

5. Waner
6.
7.
8. Gwynn
9. Crawford
10. Flick
11. Kaline
12. Clemente
13. Slaughter
14. Heilmann
15. Reggie Jackson
16. Rose
17. Dwight Evans
18. Keeler
19 / 20. Winfield and Thompson (easily last, just need to sort them out.)

I don't know where to place King Kelly. And I am still deciding how to treat Joe Jackson. What are other thinking about? Do you give any reductions for 1919 since the point of the season is to try to win the World Series? What about 1920? Some of the Chicago White Sox were suspended at the end of a very close race because of their involvement with gamblers.

Thanks for the info.

Just reading the names in the order of my rather arbitrary comprehensive offensive stat (so that this is offense only, and with no era adjustments either):

Ruth
Aaron
Ott
Robinson (So I am on board with that 2-3-4 order as in posts 2, 4, and 11)
Reggie
Rose
Crawford
Kaline
Shoeless (with full credit for 1919 and 1920 - which I might not give him)
Heilmann
Waner
Gwynn
Slaughter (includes some war credit)
Flick
Clemente
Winfield
Keeler
Evans
(There are a number of non-HoMers above Keeler and Evans, including Staub, Reggie Smith, and Jack Clark)
(and a bunch more before we get down to the last name - for instance, Colavito, Belle, Parker, and Klein)

Thompson

King Kelly is unranked by this - he doesn't easily fit into the system. And offense isn't everything resulting in some players moving up (notably Clemente) and some down (perhaps Heilmann).

===========

In fact, I really already have my ballot prepared - or I will have it prepared if only I can decide what to do with Kelly. See post #2 here.

My comment on mulder & scully's sample ballot is that I think he has Reggie too low.

Is Irvin just SOL?

I think we're going to try to splice Irvin into the LF results. At least that's what Joe says he wants although he has to do something to make that happen, like set up a special thread for it.

Prelim

1. Ruth
2. Aaron

3. F. Robby
4. Ott--this could go either way
5. R. Jackson--better road BA than Yaz, grossly under-rated

6. Gwynn
7. Crawford
8. J. Jackson
9. Waner--these 4 are very close

10. Clemente
11. Kaline--these 2 could go either way
12. Heilmann
13. Rose--these 4 are even closer yet (than 6 through 9)

14. K. Kelly
15. Slaughter--with WWII credit
16. Winfield

17. Thompson
18. Flick

19. Dw. Evans--not PHoM
20. Keeler--not PHoM

Reggie Jackson is practically Dick Allen to some people. A boastful black man who came along at the same time as Muhummad Ali, or I mean who peaked in his athletic career at about the same time. Add to that the old-fashioned prejudices about power hitters who hit for a low BA, or were perceived to do so. We now know, of course, that playing in Oakland had a lot to do with his BA. After Morgan and Bench, I have Reggie as the best player of the '70s.

Sam Crawford is surely the worst oversight of the BBWAA.

I'm probably going to have Clemente below many of you. I saw him play. He looked great and made great plays. But the comp there is Gwynn, Gwynn is so much better it's ridiculous. Clemente is closer to his other comp, Parker, than he is to Gwynn.

I used to have it Morgan, Bench, Rose and Reggie. But I suppose if the disregard for Reggie is prejudiced, maybe I'm prejudiced where Rose is concerned. He was a legitimately great player. To me the comp in terms of some of his values is Molitor. But how much better than Molitor was he? That tells you something. Other than one year, he was better than Carew at their peaks. But OTOH he was no Joe Morgan. Anyway, placing Pete on this ballot was difficult. He just kept sliding and sliding.

One of the most over-rated players in history would have to be Willie Keeler...well, after Tommy McCarthy.

Bobby comes to you and says, "Pete, you're into me for ten grand. But I'll let that go if you do this: The next game your ace is due to start, you don't bet anything. I get to bet against the Reds, but you get to keep your nose clean on that score. And then, the day of the game, you decide you need to hold your ace back a day, to keep him fresh or something, and then you start some long relief man on no notice, including to him.

Just FYI, most baseball bets say something like
"Both pitchers must start for all wagers to be official."

Not disagreeing with the overall point at all.

Mike says, Just FYI, most baseball bets say something like
"Both pitchers must start for all wagers to be official."

Thanks, Mike. Heh. That should tell everyone all they need to know about what I know about sports gambling. Approximately nothing. What I do understand is that a manager can do a lot more to manipulate a game than any one player can. But that's baseball knowledge, not gambling.

Prelim ballot

Some look to be non-controversial, others I'm going to generate a reaction

1) Ruth (top 2, #1 if Gibson doesn't get a full catcher bonus)
2) Aaron (14th overall)
3) Ott (24th)
4) Robinson (27th)
5) Rose
6) Kaline - looks a little bit ahead of the consensus, career voter bias
7) Paul Waner
8) Sam Crawford
9) Tony Gwynn
10) Enos Slaughter - with WWII credit, obviously. Not far ahead of the next guy so if Dan R calculates Slaughter's credit better than I did he may slip down.
11) Roberto Clemente - best glove on the list
12) Reggie Jackson
13) Harry Heilmann - bad glove
14) Dwight Evans
15) Willie Keeler
16) Elmer Flick
17) Dave Winfield - good baserunner, worst glove on the list
18) King Kelly
19) Sam Thompson
20) Joe Jackson - no credit for 1919 and 1920. Full credit for the rest of his career but it's only 9 years. Makes the PHoM but just barely.

others I'm going to generate a reaction

Only the Jacksons. I'm not really going to question what anyone does with Joe, but Reggie behind Gwynn and Slaughter is too low.

I'll defend Gwynn ahead of Reggie but I doubt I'm the only one with that ranking. Slaughter has a lot of glove credit. Slaughter, Clemente and Jackson are basically tied and Enos is getting 3 years of credit. I will probably go Clemente, Jackson, Slaughter on the final ballot. Reggie has no baserunning extra credit and he's a slight negative with the glove. If you look at post #11 you'll see Jackson behind Heilmann and Flick!

I admit I don't know what to do with Reggie. When I combine peak - best 3 straight years, prime - best any 7, seasonal rate, and career totals, I get a giant knot of Gwynn through Reggie. Here is how the group roughly compares. Ruth breaks the system and comes out at about 1100 on a 1000 point scale. Only War Credit Ted (with action grip) breaks the 1000 point barrier. Aaron is a little over 800. Ott is a little under 700. Robinson is about 650. Waner is a little under 600. Gwynn through Reggie are all between 565 and 520. Rose is a hair over 500. Evans just a little under 500. Keeler is about 470. Winfield and Thompson are about 440. Kelly is a good question.
Even though Winfield is last here, he doesn't compare to players like Nellie Fox or Jake Beckley for poor choices/bottom of the barrel.

So while Reggie is low on the prelim, there is not much difference between Gwynn and the CandyBar Man. So I am looking forward to some reading what others have to say about various players.

Take another look at Elmer Flick in the context of his times. He did average a 149 OPS+ for his career. He does well with both DanR WARP and WS. 8 years top 10 in OBP, 7 in SLG, 9 in triples, 6 in homers, 6 in stolen bases, 7 in walks. He is a very good "prime" candidate. He is just short the bulk part of his career.

If someone played in the 1930's up to WWII, went to war in 1943, and was soon killed in action, would he be given any extra credit or war credit? Or is war credit limited to those who survived the war?

I ask this thinking of Clemente - he was killed in a very noble mission, it was obviously not his fault and it was clear that he would have played afterwards had it not been for the earthquake. So any extra credit there? And if not, what is the justification for giving so many other players credit for games not played, and not Clemente?

If Reggie Smith happens to get elected in 2009, will be consider him to be a right fielder?

Diz - chart for you, Win Shares for each of these guys from age 37 on, maybe it will help you decide how much credit you would give Clemente:

Brock, just to be clear, Rose's gambling has nothing to do with this election. It's unconstitutional to hold it against him, unless you think he was throwing (not betting on) games as a player.

For Jackson, I think it's within the Constitution to dock him for 1919, although I am not sure if I will or not (leaning no). I think 1920 is a lot more questionable on those grounds although the argument could be made.

If someone played in the 1930's up to WWII, went to war in 1943, and was soon killed in action, would he be given any extra credit or war credit? Or is war credit limited to those who survived the war?

I ask this thinking of Clemente - he was killed in a very noble mission, it was obviously not his fault and it was clear that he would have played afterwards had it not been for the earthquake. So any extra credit there? And if not, what is the justification for giving so many other players credit for games not played, and not Clemente?

I think the argument is/was (reading the old war credit thread) that things like race and being of draft age at the time of major wars were factors determined at birth outside of your control. It's true the earthquake and plane crash were outside of Clemente's control, but they were also not a definitive piece of destiny. If we start extending credit for every event that shortens someone's career we'll end up just scaling everybody to 20 year careers regardless of what they did. However it is worth noting that Clemente, at age 37, was still putting up OPS+ of 137. Even with a quick decline given his age (though there certainly didn't seem to be signs of it) you could probably expect 2 more years at least without stretching things too much I would think.

Joe says, and he is the inventor of this HoM concept, if I've read correctly, "Brock, just to be clear, Rose's gambling has nothing to do with this election. It's unconstitutional to hold it against him, unless you think he was throwing (not betting on) games as a player.

For Jackson, I think it's within the Constitution to dock him for 1919, although I am not sure if I will or not (leaning no). I think 1920 is a lot more questionable on those grounds although the argument could be made."

Thanks, Joe. As you can doubtless gather, I think Rose is guilty of a lot more than Jackson is, although I have no proof of either man's activities. However, I think Rose manipulated games as a manager, not a player ( he couldn't have done it alone, and I've never even smelled a whiff of a conspiracy among the Reds), so I completely understand that it's not germane to this voting. As for Jackson, all I am willing to dock him for is the 1919 Series, where he is a small part of a large conspiracy that turned, as it must if you don't have the manager in tow, on the pitchers. I don't know of any evidence at all that implicates him in any other manipulated games, even in the 1919 and 1920 seasons. If anyone knows better, please post up. Thanks, Brock

Oh, shoot. In all the anxiety about raising the Rose issue, I forgot about this one: Can I give King Kelly credit for all the early rules that he exposed and that then got changed? That is, can I give him credit for improving the game by his actions as a player? He wasn't trying to do that - he was trying to win games, and he found out he was good at rules lawyering - but the effect was to speed up the process of getting rid of, or improving the language of, a lot of the early rules. Also, can I give him credit for trying out things like setting up in unusual defensive positions, which allowed baseball to try those experiments out in game situations? It's sort of like Babe Ruth, except that Ruth doesn't need credit for experimenting; the value of the homers he hit gives him all the credit he needs. Kelly tried out a lot of things that were smaller, and a lot that failed, too; but still, the fact that he tried them all out meant that they weren't there later to gum up the works. I'm sure that has value; I just want to know if it's the sort of value that counts here. Thanks again, Brock

My rationale for eliminating 1919 and 1920. The "pennant is a pennant" stipulation seems to have a corrolary for me that if you throw the pennant that you shouldn't get any more credit for that year than someone who didn't play at all. That takes care of 1919. For 1920 I posit that Jackson wasn't banned yet only because the wheels of justice took time to finally ban him. If they had been able to process things sooner he would have been banned for life immediately after the World Series. I am retroactively eliminating him from the game at the earliest opportunity.

I have Reggie Smith as a CF but I'm not sold on it. As a RF he would be between Winfield and Kelly.

I forgot to rank Gavy Cravath - he's in my PHoM and would be just ahead of Joe Jackson. That's pretty much the in-out line.

If I remember right about discounting Joe Jackson's 1919 and 1920, it was similar to this: The goal or point to a baseball season is to win the World Series and players are supposed to put their best efforts toward that end. For 1919, the argument was since the Jackson participated in throwing the World Series, it invalidates his entire contribution to the season. For 1920, Jackson and the others were suspended for the end of the season. The White Sox were in a close pennant race that they ended up losing to the Indians. They were not able to put their best efforts toward winning the Series since they were suspended.

I have been actually having to work, so this post may have become redundant since I started it.

Okay, my prelim I guess...

Top 4 for me -- competition adjustments and Ott's home field advantage just nudge Robinson over Ott in my opinion:

1) Ruth
2) Aaron
3) Robinson
4) Ott

I haven't really ranked the rest yet. I've got them ranked within (roughly) the eras they played.

The recent-ish players:
Rose (probably deserves some infield credit)
RJackson
Kaline
Clemente (part of me wants to put Clemente and Kaline ahead of Reggie due to defense -- I still might)
Gwynn
Evans
Winfield

19th century and deadball:
JJackson
Crawford
Flick
Thompson
Keeler
Kelly (I don't have him ranked below Keeler and Thompson, and he's probably above Flick and maybe Crawford due to positional versatility -- I just haven't got a good hold on him yet)

Between the wars:
Waner
Slaughter
Heilmann (Big Poison and Country could catch the ball -- Slug sure could rake, though... I might play with this some, I don't know, hitting conditions were easier for Heilmann, so he's probably the last of the group)

I have Gwynn #6 and now I'm thnking that's too high.

OPS+.....AB + BB.....yrs w/100 games.....median # games in those yrs

Gwynn 133 10,078 16 134.5
Reggie 140 11,239 19 143
Crawford 143 10,330 17 149
Waner 133 10,550 16 148
Kaline 134 11,393 20 137 the second least durable modern player

I'd have to say Gwynn is pretty over-rated. Reggie, meanwhile.... Only Ruth, Aaron and F. Robby have both a higher OPS+ and more PA. And he was a pretty good defender in his younger days, even played some CF.

I don't care if you add 3 full years at 650 PA to Slaughter...he's still around a 122 OPS+ player and that gets him up to 10,750 PA. That is not better than a 140 at 11,200. I mean, seriously. What is this prejudice against Reggie Jackson?

Early preliminary ballots are leaning Gwynn over Clemente, 5 to 2, though one of those doesn't include defense. Despite being similar types of players in many respects--high average, relatively low walks, extra-base power but not sluggers, speed, good defense (at least for the young Gwynn), they are rather difficult to compare, because their career shapes are so different.

Clemente's counting stats are padded at the front because he started for a number of years for a terrible Pirates team, when he really should have been seasoning in the minors, but his rate stats are suppressed by this same feature. His counting stats are cut short, however, at the other end of the career by his untimely death.

Gwynn has only a couple of part-time seasons before he became a star at age 24 (one year earlier than Clemente's breakthrough year at age 25), but he has a long decline phase, with four seasons of gradually reduced playing time after his last big year, which at age 37, came at the same point in his career as Clemente's final season.

It may be instructive, therefore, to compare their primes. Gwynn's prime is obviously age 24 to age 37. Clemente's actual prime was age 25 to 37, though we can't know if he had another excellent season in him. To put them on equal footing, let's look at age 24-37, so that Gwynn gets credit for establishing himself at a younger age.

Here are their basic stat lines, with all seasons adjusted to 162 games (for work stoppages and 154-game seasons)

Gwynn 2025 g, 8789 PA, 136 OPS+
Clemente 1935 g, 8212 PA, 141 OPS+

Now to really work this out, you'd have to add in baserunning and defense, with the baserunning advantage going to Gwynn, I think, and the defense advantage going to Clemente for sure.
Dan R's WAR looks at everything, of course, and his results, age 24-37, for Clemente and Gwynn are

Gwynn 74.6
Clemente 67.7

Gwynn leads by 6.9. 6.2 of that advantage comes from their age 24 seaasons, when Gwynn was great and Clemente was barely above replacement level.

Dan R's salary calculator, however, is more friendly to Clemente, who had more great seasons than Gwynn did.

Gwynn $113,761,050
Clemente $116,331,605

Based on this study, a career voter clearly needs to favor Gwynn, whose decline phase was surely more valuable than Clemente's apprentice phase. But prime and peak voters could surely go either way. What should be confidently ruled out, I think, is sunnyday2's claim that Gwynn was so much better than Clemente that "it's ridiculous," and that Clemente is closer in value to Dave Parker than he is to Tony Gwynn.

For the record, here are Parker's 24-37 totals:

Parker 1939 g, 8124 PA, 126 OPS+, 37.8 WAR, $57,845,423

Gwynn 2025 g, 8789 PA, 136 OPS+, 74.6 WAR, $113,761,050
Clemente 1935 g, 8212 PA, 141 OPS, 67.7 WAR, $116,331,605

Dave Parker, who had a great run in the late 1970s, is nevertheless not worthy of being the same conversation with either Clemente or Gwynn.

Having just defended Clemente, I should concur with Sunnyday2 on Jackson: I have him ahead of both. Not by much, but ahead.

Early preliminary ballots are leaning Gwynn over Clemente, 5 to 2, though one of those doesn't include defense.

Assuming I stay with what I posted on the LF ballot thread, my 11-14 will be Waner, Gwynn, Clemente, Slaughter.

I think you're using the old salary estimator, Chris Cobb, not the new one, which values in-season durability and is less peak-friendly. (It's $380,000*SFrac + $1,200,000*(WARP^1.5)).

I used the salary numbers in what I thought was the latest version of the Dan R WARP, which I downloaded a couple of months back. Does that version not have up-to-date salary estimator numbers?

Dan R's WAR looks at everything, of course, and his results, age 24-37, for Clemente and Gwynn are

Gwynn 74.6
Clemente 67.7

WARP1 has the totals for those years as:

Gwynn: 109.6
Clemente: 121.2

I wonder what causes the difference.

Well, one problem is my spreadsheet management skills.

Looking again, I find I grabbed Gwynn's W2/year rather than his WAR2 for his 24-37 seasons. Using the correct numbers, his total is

Gwynn 69.2

I'm looking into other factors now.

So, we're looking at BP rating Clemente at 11.6 wins better than Gwynn during their primes, while Dan R's WARP 2 puts Gwynn 1.5 wins ahead.

I'd say the most obvious difference is in fielding evaluations: WARP likes Gwynn less, and Clemente more, than Dan R does.

BP WARP has Gwynn at 2.0 wins above average for this period.
Dan R WAR2 has Gwynn at 6.9 wins above average for this period.
+4.9 wins for Gwynn

BP WARP has Clemente at 13.8 wins above average for this period.
Dan R WAR2 has Clemente at 9.7 wins above average for this period.
-4.1 wins for Clemente

Net +9 wins for Gwynn

Dan is better at answering questions like this, so I won't dig further, but that looks like it's probably the key difference.

Yes, I changed salary estimators after the last version of my WARP. The underlying numbers haven't changed since the version of the spreadsheet you have, only the equation I use to convert WARP scores into my personal rankings, since the group convinced me of the importance of in-season durability.

Remember that for pre-1987 seasons, my published FWAA are nothing more than an average of BP FRAA and Fielding WS, albeit doctored to fit an empirically valid standard deviation. For Clemente, the difference betwen BP's +14 wins and my +10 is probably just the regression to the mean I apply because the stdev of BP FRAA in the outfield is a bit too big; for Gwynn's post-1987 seasons, it clearly means BP FRAA is less favorable than Dial Zone Rating and Fielding WS are. I can always refer to DRA/TotalZone/SFR in individual cases, but I haven't yet incorporated them into my published FWAA.

whoisalhedges wrote: Top 4 for me -- competition adjustments and Ott's home field advantage just nudge Robinson over Ott in my opinion

Ott's home field advantage wasn't particularly large. He hit 7% better at home than on the road, per Total Baseball I. Frank Robinson hit 10% better at home.
Frank Robinson's splits

sunnyday2 posts up the following useful chart:
Gwynn 133 10,078 16 134.5
Reggie 140 11,239 19 143
Crawford 143 10,330 17 149
Waner 133 10,550 16 148
Kaline 134 11,393 20 137 the second least durable modern player

If you're talking about these five, based on this chart, I'd say to give the advantage to Crawford. He's from an earlier era than the others, when bodies could play fewer games than in later times, and so his games played don't look so hot, but they were much more impressive in his time than anyone else's on this list. In short, Sam Crawford holds the title for most games played by a right fielder (50% of games in right) until 1944, when Waner and Ott finally pass him. The other guys don't hold for anything like that long, and Sam has the highest OPS+. That is, among these five guys, Crawford should be credited with the highest OPS+ and ALSO the longest career, relative to his times. I know I'm ranking him first in this class. - Brock

1. Ruth (Easy.)

2. Aaron
3. Ott (These two are actually reasonably close, but Aaron clearly ahead.)

4. F Robinson
5. J Jackson (Full credit.)
6. S Crawford
7. Rose
8. Kelly (Not at all sure about his placement, but I’m thinking I like him more than most. May move down, perhaps significantly, though.)
9. Waner
10. R Jackson
11. Slaughter (With WWII credit.)
12. Flick
13. Gwynn
14. Heilmann
15. Clemente
16. Kaline (He and Clemente are really, really close.)

17. Winfield

18. Dw Evans (Not pHOM, but pretty close.)
19. Keeler (Not pHOM, not particularly close.)
20. S Thompson (Not pHOM, and soooo far from close.)

Oops, there shouldn't be a gap between Winfield and Evans--that's not a huge separation, despite being the place where my pHOM line hits.

By the way, has anyone figured out how we're voting on pitchers yet? We seem to be about to run out of position players to put it off with....

2. Aaron
3. Ott (These two are actually reasonably close, but Aaron clearly ahead.)

Surprisingly the OPS+ are razor-close:

Aaron 156
Ott 155
F. Robby 154

But of course Aaron has an extra 2,500+ PA over both. So I don't really see Aaron and Ott being close. Ott and R. Robby OTOH are very close. 1 OPS+ point for Ott, 300 PA for Robby, though obviously in longer seasons (8 games X 18 years or so = >150 extra games, so I'm guessing Ott's adjPA are greater.

But then Robby played in an integrated league. I just don't see how Ott is close to Aaron, nor how he is ahead of Robby.

OTOH nobody has suggested they're not 2-3-4, so that's a good thing. I don't see any justification for anybody else in the top 4. No, forget justification. Not even a rationalization. This of course from the lowest consensus score in LF and CF, and who had Mantle #1. Still....

By the way, has anyone figured out how we're voting on pitchers yet? We seem to be about to run out of position players to put it off with....

Don't we have a "what are we going to do next?" thread somewhere? I was looking for it so I could bump it with this question, but I don't remember its exact name.

Not that there's any question who's #1, but here's my hypothetical take on voted awards had the current system of such awards been in place: AL Rookie of the Year 1915; AL Cy Young 1916; AL MVP 1918, 1920, 1921, 1923, 1926. The statheads would have argued that he got robbed in the MVP votes of 1919 (Joe Jackson) and 1924 (Walter Johnson). The 1928 MVP probably shouldn't have been Cochrane, but Ruth is only one among several other possibilities. And while we understand what the voters did in 1927 (Gehrig), 1929 (Foxx), 1930 (Simmons or Grove), and 1931 (Grove), Ruth was a solid MVP candidate in all of those years.

I just don't see how Ott is close to Aaron, nor how he is ahead of Robby.

Well, I didn't mean to say he's that close to Aaron. Their peaks aren't too far off one another, but Aaron pulls away on career (heck, his career is one long extended prime, really). I meant to say they're closer than I thought they were, I guess. Got carried away.

As for Ott over Robinson, perhaps I'm not penalizing Ott enough for league strength/integration issues, but I have his peak/prime as a good notch ahead. (Not being remotely a career voter, I'm paying essentially no attention to career OPS+ rates.) Basically, Ott pulls ahead for me on third-through-ninth-best years, and the league adjustments aren't close to making up that difference. (In my still-peak-oriented system, these are very important seasons.)

But I do agree about these four clearly being the top four.

Busy times meant that I've missed most of the positional threads. Let me join in the RF discussion since I have done the work here. This might be the easiest in that you have no Negro League players to try and fold into the rankings.

I still use the same system I used in the HOM yearly elections. In scoring it takes the top candidate, assigns value of 1, then decreases from there.

Methodology in brief: The system used for my ranking entails a little bit of everything including WS, WARP, OPS+/ERA+, positional adjustments, edits for minor league, war, NeL credit, even some contemporary opinion. Oh, and Dan R’s salary estimator made me re-examine folks as well. So once that info is assembled I try and make other changes to metrics when deemed fit, weighting the various measures. My hope by including all this material is to get the most complete picture, a worthy player from all angles. The results of this work tend to favor prime/peak players over career types but that is not 100% tried and true.

Here's where they stand - forgive formatting I forget how to table results:


I'm giving full credit to Shoeless Joe, some WWII credit for Slaughter. I'm wondering about Waner - seems he spent 3 years in the PCL, winning a batting title. Should credit be given in that case?

I'm wondering about Waner - seems he spent 3 years in the PCL, winning a batting title. Should credit be given in that case?

If you use your own numbers as you did in #52 and already put him 5th, then there's no way he could possibly catch Robinson no matter what you gave him in minor league credit. For you, a moot point.

Not being remotely a career voter, I'm paying essentially no attention to career OPS+ rates

I don't disagree with this, it was just a quick and dirty. The following is of more interest:

OPS+ best to worst ?100g

Ott 155/179-78-75-73-71-70-66-65-59-53-52-50-49-49-39-38-37-33
FRob 154/200-189-72-69-64-64-58-54-53-53-52-51-48-43-39-32-31-29-16

Ott 12 Robby 4 Even 3. But as a peak voter I kinda like those big 2 from F. As a GM or a manager, I think Robby's 200-189 gives me a bigger chance at a pennant than Ott's small leads in yeard 3-9.

From Post 28 from Joe and Brock:

(Joe) I think 1920 is a lot more questionable on those grounds although the argument could be made."
....
(Brock)I don't know of any evidence at all that implicates him in any other manipulated games, even in the 1919 and 1920 seasons. If anyone knows better, please post up.

Took me a day or two to dig this up, Asinof's Eight Men Out - Third Section, The Exposure - Last 4-5 pages of the Chapter have several damning paragraphs about Cicotte and Jackson and the others.

After a paragraph describing a 2-1 lead with Faber pitching which was lost late when a flyball went over Jackson's head and Risberg's relay throw missed Schalk and Faber backing up the play Asinof writes
"The pattern was repeated periodically throughout the summer. The ballplayer accepted it as a matter of course."

Is this evidence? I'd say yes, at least the best we have. After all in none of these situations are their actual corpses or smoking guns.

I think you would be more than justified in giving Jackson (Cicotte, Risberg ect.) ZERO credit for 1920.


BTW - Risberg led the league's SS in errors, though Galloway had a worse fielding %, Jackson led the league's OFers in Errors. Weaver was last in the league in Range Factor, and Fielding %, made 12 more errors than any other 3b - only one player played third more than he did though. Cicotte was tied for 2nd most errors by a pitcher.


(You might notice, I didn't cite a page number(s), why? Of just the guys on this list how many different editions of this book do you think we own? Has to be at least five, if I was setting an over/under I'd set it at 6.5 and let you pick a side)

Here's Ott vs. Robinson, as I see them, after penalizing Ott's war seasons and correcting my published data for the years when they played multiple positions:

Ott



3-year peak: 25.2
7-year prime: 56.6
Career: 111.4
Salary: $350,075,166; above Rickey, Ripken, and F-Robinson, below Morgan, Schmidt, and Gehrig


Robinson



3-year peak: 25.7
7-year prime: 53.0
Career: 106.2
Salary: $316,560,930, behind Ott, Ripken, and Rickey, ahead of Foxx, DiMaggio, and Vaughan

Double-checking the defensive stats, TotalZone has Robinson at +28 (compared to around +50 here), and in the seasons that DRA is available for Ott, it has him at +62 (compared to +54 here), so they're in the ballpark. On baserunning, Dan Fox's EqBRR has Robinson at +12, against +22 or so here, so again, close enough for comfort.

Robinson had more baserunning and fielding value than Ott: he accumulated 8 wins above average in those categories, versus Ott's 2.5. But that gap is roughly made up by the fact that the positions Ott played had more defensive value: he had 250 games at third base when it was still a key fielding position, whereas Robinson had 275 at first when the position was at its deepest, and 350 more at DH. This can be seen in the Replc column, where an replacement player at Ott's mix of positions would have been .9 wins below average per 162 games, wherease one at Robinson's mix of positions would have been .6 wins below average per 162 games, or an advantage of about 5 wins for Ott.

What breaks the tie, then, is that my system simply doesn't agree with OPS+ that the two were roughly equal as hitters: it sees Ott as meaningfully better. Based on my BWAA, Robinson's OPS+ "should" be 151, while Ott's "should" be 158. Discrepancies of more than a few points over a long career are extremely rare. Here are the explanations, as I see it:

1. Ott was *extremely* good at staying out of double plays. I don't have records for the first part of his career, but from 1933 onwards, he hit into just 82 double plays, whereas a league-average player would have hit into 162 in his plate appearances. That's worth about 4 wins, or 2 points of OPS+ (giving him 0 credit for anything before 1933, when I assume he was league-average). By contrast, Robinson had a slight DP propensity, hitting into 27 more than a league-average player would have in his opportunities. That costs Robinson the equivalent of one point of OPS+.

2. Ott's OPS was fairly OBP-heavy--his OPS+ breaks down into a 121 OBP+ and a 134 SLG+. By contrast, Robinson's figures are 118 and 135.5. Since we know OBP is underweighted in OPS+, I suspect this gap is worth something on the order of 1.5 points of "fair" OPS+.

3. Robinson drew a lot of IBB. If only a league-average percentage of his walks were intentional, his "fair" OPS+ would have gone down half a point more. (We have no information on Ott's IBB rate, so we assume it's league average).

So the "nominal" OPS+ gap between the two is 1.5 points. Add on 3 points for double plays, 1.5 for OBP-heaviness, and 0.5 for Robinson's intentional walks, and you're at 6.5 points, almost equal to the 7-point gap implicit in my BWAA. Now, if baseball-reference corrected for that and showed a 158 OPS+ for Ott and a 151 for Robinson, I doubt anyone in the group would choose Frank.

Hey guys. I've been too busy to participate in all the newer elections, but now that things have cleared up, I'd like to rejoin.

I'm still using the same system I used in the last yearly election: OPS+, with adjustments based on Dan's baserunning and defense numbers, compared to a positional baseline. I balance peak and career with a JAWS-like formula.

So, I have them ranked:

1-Babe
2-Hank
3-Mel Ott
4-Frank Robinson
5-Paul Waner
6-Sam Crawford
7-Al Kaline
8-Joe Jackson
9-King Kelly
10-Pete Rose
11-Roberto Clemente
12-Tony Gwynn
13-Harry Heilmann
14-Reggie Jackson
15-Elmer Flick
16-Enos Slaughter
17-Dwight Evans
18-Sam Thompson
19-Dave Winfield
20-Willie Keeler

Crawford to Joe Jackson and Clemente to Reggie are somewhat tightly bunched, and I'm still not clear on the ordering of the last three.

We have no information on Ott's IBB rate, so we assume it's league average

That's not a fair assumption, though obviously it doesn't make a big difference.

I'm still not clear on the ordering of the last three.

My suggestion would be to put Thompson last.

DL from MN:
6) Kaline - looks a little bit ahead of the consensus, career voter bias
7) Paul Waner
8) Sam Crawford
9) Tony Gwynn
10) Enos Slaughter - with WWII credit, obviously. Not far ahead of the next guy so if Dan R calculates Slaughter's credit better than I did he may slip down.
11) Roberto Clemente - best glove on the list
12) Reggie Jackson
13) Harry Heilmann - bad glove

Here at rightfield "career voter bias" cannot explain much. This position at the Hall of Merit is chock full of long careers including long careers fielding RF. On might call Harry Heilmann's career only "middling" long among Hall of Fame members and strong candidates. Maybe the same for Willie Keeler and Mike Kelly. Elmer Flick and Sam Thompson have the only "shortish" careers on this ballot.

More to the point, the six players just behind Kaline on your ballot DL all played a long time.

Ranked by full seasons fielding RF, the ten leaders on this ballot are all in the first 13 all-time, with Harry Hooper and two very recent players Sammy Sosa and Paul O'Neill. Passing over the very recent and the active players, Harry Hooper "represents" Willie Davis, Doc Cramer, Brett Butler, Dummy Hoy, and Amos Otis among the career leaders in full seasons fielding centerfield.
(Why? If you aren't a good enough batter to be a strong Hall of Merit candidate, you don't hit well enough at age 35 to enjoy a long career at rightfield.)

I'll get back to career fielding time at RF and in the outfield, and to career playing time in all roles.
For now here is the candidates career time at centerfield.

Full seasons fielding CF, at least 2.0 seasons
6.7 Dawson (#50 at RF, 1871-2006)
3.1 Kaline (#9)
3.1 Crawford (#16)

Among the top 50 by full seasons fielding RF, from Waner to Dawson, these three and Sam Rice with the 3.9 season are the only ones with even two full seasons fielding CF!

Here’s a measure that I thought some folks might find interesting. It’s one I call peak rate, and it’s one of three measures that contribute to my ranking system. Basically this is the player’s highest rate of wins per 162 games over five consecutive seasons. If a player has a higher rate over more than five consecutive seasons than he does over five consecutive, I use the longer period’s rate. (Doesn’t happen often: of the players listed here, only Reggie Jackson does better over six consecutive years than over five.)

The following table has peak rates, denominated in Dan R’s WAR2, for all the right field candidates, with the peak seasons listed.

Rate Player (years)
12.23 Ruth (1920-24)
8.22 Aaron (1959-63)
7.92 J. Jackson (1911-15)
7.83 Ott (1934-38)
7.50 Robinson (1966-70)
7.26 Heilmann (1923-27)
7.23 Clemente (1966-70)
6.69 Kaline (1964-68)
6.60 Flick (1903-07)
6.60 R. Jackson (1969-74)
6.52 Waner (1926-30)
6.14 Keeler (1895-99)
6.11 Gwynn (1984-88)
5.74 Thompson (1891-95)
5.68 Slaughter (1939-42, 46)
5.48 Evans (1981-85)
5.45 Rose (1968-72)
5.42 Crawford (1907-11)
5.40 Winfield (1976-80)

One interesting thing that this metric shows is that Clemente and Kaline can be seen as peak candidates. If rate is separated from playing time, they do very well. It is easy to see why people who watched them play thought they were great: among post-integration players only the inner-circlers Aaron and Robinson were better.

It also points out just how great Joe Jackson was, once upon a time. He was at his best at the very outset of his career—he and Paul Waner are the only players who peak rate begins in their first full major league season—so there’s no guarantee that he would have gone on to have the full career of Ott or Robinson, both of whom peaked late, but it’s hard to see how he wouldn’t have at least locked up the #5 spot.

Crawford does surprisingly poorly by this measure. He had a lot of great years, but they were always interspersed with quite ordinary ones, a problem that also afflicts Gwynn, whose great seasons are scattered through his career. It’s a limitation of this individual metric not to spot the merits of those players, but that’s why one measure is never enough.

Sam Thompson's low performance shows just how little he brings to the table: short career, and his peak performance rate is bottom third also: everyone below him had a huge career and most played in the post-war era when it was a bit tougher to put up high rates, even when adjustment for standard deviation has been made. He's not necessarily a mistake pick, but he really ought to be resoundingly last in this ballot.

Ruth laps the field, of course.

Food for thought, perhaps.

Peak rate is intriguing. I know that Kaline, Robinson, and Clemente hit their five-year peaks in series, not in parallel, but I see the point. (Robinson was my favorite player and I've known since 1975 was underway that he was still a devastating batter then, when he broke the color line for managers, with a .237 BA in a DH league.)

My reaction in more detail.
One interesting thing that this metric shows is that Clemente and Kaline can be seen as peak candidates. If rate is separated from playing time, they do very well. It is easy to see why people who watched them play thought they were great: among post-integration players only the inner-circlers Aaron and Robinson were better.

As a schoolboy fan who arrived in the mid-sixties, I was there for Roberto Clemente at his peak and I know that Al Kaline peaked in the fifties ("you should have seen him when he was really good"). It was sad that the Tigers didn't win until Kaline was over the hill, in 1968 and 1972.

I have many times named Kaline and Vada Pinson, with some exaggeration, as the players (long-time stars, not pitchers) who were at their best in their first seasons --the anti-Clementes. Now I still know that --except that Kaline's best was 1955 and 1955-59, beginning in year two-- but it is remarkable that AlKaline may have been better, when he was able to play, ten years later when I could see him on TV! When the Tigers won in 1968 and 1972, he played only 102 and 106 games, only 74 and 84 in the outfield --but he hit OPS+ 146 and 149, better than his 1955-59 average. (Kaline played 736 games in '55-59; despite expansion, only 646 in '64-68.)

OK, I understand. But was he really better by "rate" in '64-68 than in '63-67?

It also points out just how great Joe Jackson was, once upon a time. He was at his best at the very outset of his career—he and Paul Waner are the only players who peak rate begins in their first full major league season—so there’s no guarantee that he would have gone on to have the full career of Ott or Robinson, both of whom peaked late, but it’s hard to see how he wouldn’t have at least locked up the #5 spot.

As for Kaline who peaked in the late fifties, "everyone" knows that Frank Robinson peaked in the early sixties. In '67 he missed the last(?) month after the Jack-- beanball and in '68 he was a shell of himself with double vision.
This peak rate (how high) and rate peak (when) provide a different perspective, on occasion radically different.

Chris, your "peak rate" (how high?) and "rate peak" (when?), if I may coin another term, provide a different perspective, on occasion radically different.

Not that this would ever affect any ranking of him, but has anyone else noticed that FRAA/FWS REALLY love the young Babe Ruth? From 1921-1923, he was +1.6, +1.0 (in a shortened season), and +2.5 wins in Dan's spreadsheet. Now, as I understand it contemporaries thought he was a decent if unspectacular defender when young, but I can't remember anyone claiming that he was basically Roberto Clemente for a few years. Does the electorate believe that Ruth really could have been as good a fielder as these numbers suggest?

Maybe everyone was too busy dropping their jaws at the sight of his hitting to notice his glove. Still, what do the other stats (if available) say, Dan?

I pretty much have my ballot set, except for one thing - I'm still no closer to understanding King Kelly. Can anyone help me out by giving some arguments about him, pro or con? Especially in contrast to assorted mid-ballot candidates like Flick, Kaline, Slaughter, and so on?

The peak rates are all pretty plausible except a couple things.

Kaline was better in the '60s than the '50s?
Crawford??? You're kidding, right? Way too low. That's gotta be a period adjustment talking.
And Heilmann higher than one might think.
And Keeler.
Reggie maybe a little low

Current Prelim

1. Ruth
2. Aaron
3. Ott
4. Robinson
5. Waner
6. Crawford
7. Gwynn
8. Kaline
9. R. Jackson
10. Clemente
11. Slaughter
12. Heilman
13. Rose (This is the one mos tlikely to adjust...upwards)
14. J. Jackson
15. Evans
16. Kelly
17. Winfield
18. Flick
19. Keeler
20. Thompson (The only non-PHoM on the list)

I wonder if Clemente lost much development due to the Dodgers' handling of him as an early blue-chip prospeoct who they desparately "hid" from other clubs by not playing him when other scouts were around. They supposedly also set up a whispering campaign of sorts to label him poorly - all in the unsuccessful effort to keep teams like Pittsburg from finding out about this guy that the Dodgers had no room for in the OF when he was a teenager.

Yes, there was a lengthy discussion thread once about Ruth's early-20's fielding. FRAA, Fielding WS, and DRA all agree that Ruth's 1923 was one of the finest defensive seasons ever by a corner outfielder--he simply had an ungodly, Ashburn-esque putout total. The '21 was pretty nifty too. I searched online and indeed found quite a number of contemporary references to Ruth's glove being very highly regarded at the start of his career as a position player, before he became a real liability around 1928: see this page.

I would note that it's not inconsistent with a typical aging pattern for a fielder to be excellent in his mid-20's and poor in his 30's, that Ruth didn't start playing the outfield full time until 1920 (meaning he may have needed a year or two to learn the subtleties of the position), and that we are, after all, talking about one of the greatest athletes in the history of the world. The reason we are stunned that Ruth's early-career fielding numbers look so good is that we remember him as being fat. But it's very possible that he was quite a lithe athlete in his first years with the Yanks--his triples totals (16 in '21, 13 in '23, never more than 9 after that) are one indication that that might be the case.

I'm a firm believer in athletic g, the idea that some guys are just gifted with their bodies and will probably be good at whatever you ask them to do if you give them enough practice. Ruth was a brilliant pitcher and the best hitter ever; why couldn't he have been a good fielder from age 26-28 as well? Note that most of the all-time greats we remember for their hitting (Ted Williams, Bonds, Musial, Aaron, Mantle etc.) were at least above-average fielders in their youths; they just didn't sustain it over the course of their very lengthy careers. While I think there is certainly a case to be reasonably skeptical that Ruth in 1923 was Babe Ruth and Darin Erstad all rolled up into one--there were probably more balls hit to him that year in particular than the team's K rate, handedness, FB tendency, and ballpark can tell us--I have no problem thinking he was a well-above-average, possibly even Gold Glove-caliber defender during his first years with New York.

That's interesting. I still don't think he was +2.5 wins in 1923, but I agree that it is not crazy to say he was a pretty good fielder as a youth.

Also, if you go to his wikipedia page, there are a few pictures of the early Ruth. While he was very thin in the 1910's, by the early 1920's he was substantially bigger; the 1921 Ruth was definitely portly. From the comments Dan linked to and his mention of putout totals, it appears Ruth's arm constituted an enormous portion of his defensive value, so perhaps he was able to compensate for his large build by throwing the ball so well (which would, of course, certainly be consistent with what his previous position.)

Damn, no edit feature. I guess my last silly comment, confusing putouts with assists, will stand for eternity.

Responding to Paul:

re Kaline: OK, I understand. But was he really better by "rate" in '64-68 than in '63-67?

The difference between 63-67 and 64-68 is not large:

63-67 rate is 6.25
64-68 rate is 6.69

The later rate is higher for two reasons:

1) Kaline's rate in 1968 is considerably higher than his rate in 1963 (6.6 vs. 4.6). (Kaline's "decline" in 1968 was largely a false image created by the Year of the Pitcher and his being injured for a significant part of the season.)
2) Kaline's lesser playing time in 1968 vs. 1963 puts more weight, within the five-year window on his great seasons in 1966-67 (7.1 and 8.9 by rate, respectively)

The effect of seasons with lesser playing time on the peak rate stat is one of its limitations: Frank Robinson's late rate peak is also partly because his reduced playing time in a couple of seasons after the monster year in 1966 gives more weight to that season in the 1966-70 window than it has in the 1962-66 window. However, the more drastic effect in Robinson's case is that his 1963 and 1965 seasons were subpar years. Both of those seasons, by WAR2, have rates below 5.0, whereas Robinson's lowest rate in 1966-70 is his 5.7 in 1968 (he doesn't, by this measure, look like "a shell of his former self"--guess that perception was created mostly by the Year of the Pitcher, just as it was in Kaline's case).

Responding to sunnyday2:

"Plausibility" is not really the issue here: these rates are what you get if you sum WAR2 over five consecutive seasons and divide by playing time. As to points on which you are dubious:

Kaline was better in the '60s than the '50s?

In terms of rate, clearly.

His OPS+ from 1955-59: 162, 139, 120, 130, 151
His OPS+ from 1964-68: 134. 142. 161, 176, 146

The later run is obviously stronger. It's not part of Kaline lore because a) he was missing a lot of playing time and b) the depression of raw stats because of the era's low offense made it look like he had declined.

Crawford??? You're kidding, right? Way too low. That's gotta be a period adjustment talking.

Well, Crawford's rate peak was 1907-11 in the AL. Flick's peak is the 1903-07 AL, and Jackson's is the 1911-15 AL. If there's a problem with a period adjustment, it's hard to see how it could miss them and hit Crawford.

Now, Crawford's rate is definitely trimmed a bit by adjusting for standard deviation, but even so, he has a number of ordinary years mixed in with the great ones.

1904 -- 114 OPS+ -- nothing special for a right fielder
1906 -- 131 OPS+ -- very good, but not a great year.
1910 -- 143 OPS+ -- very good, but not a great year.
1912 -- 130 OPS+ -- very good, but not a great year.

He never put up five years running where he was consistently above 150 OPS+, and his fielding and his baserunning were basically average, so his peak is being carried entirely by his bat.

Compare Flick's 5-year rate peak to Crawford's by OPS+

Flick: 136, 159, 166, 155, 153
Crawford: 160, 159, 153, 130, 163

Pretty similar. Now figure that Crawford is average defensively and on the bases, while Flick, between baserunning and defense, is about 1 win above average per year.

That's the difference between their rate peaks:

Flick 6.60
Crawford 5.42

For the record, Crawford's rate without a standard deviation adjustment (Dan R's WAR 1) is 5.48, so the period adjustment's effect is minimal.

Correction to the post above in review of Sam Crawford's lesser seasons:

1910 -- 130 OPS+ -- very good, but not a great year.
1912 -- 143 OPS+ -- very good, but not a great year.

Ruth's very early defense was very good. His range rankings as a pitcher in Boston really are far above the league norms. In 1918, the Sox started playing Ruth a lot in the outfield, and he didn't do well in either that year or 1919, with range factors right around the league. The reason was probably the ballpark. The Sox played him as a left fielder / center fielder, with zero games in right field. As you all know, playing left in Fenway will not increase your fielding rankings, because there is so little territory to cover, and so a low number of real chances to catch flies. It's the weaker corner position in that ballpark, where the worst of your corner outfielders generally goes. Why was he there? Well, he'd probably played very little outfield before (as a semi-pro player, he was a pitcher / catcher, if I remember right). But that's not the main reason. The main reason was that the Sox had Harry Hooper in right, and Hooper did in fact have a Clemente-type reputation as a corner outfielder. It's no disgrace to get exiled to left field because of Hooper.

In 1920, he moved to New York, where it is clear that manager Miller Huggins does not know what, exactly, he has. Ruth played 85 games in right, 20 in center(!), and 36 in left. The center field games indicate that Huggins had not ruled out the possibility that Ruth was fast and good enough to play there. The emphasis on right field is the same thing as the Hooper. Unlike in Fenway, right is the small outfield corner in Yankee. The Yanks had Bob Meusel in left, who had a very good glove rep and a monstrous arm rep. Ruth mostly played right because of Meusel's skills.

In 1921, which is one of his very best year in all arenas of play, Ruth is a left fielder (134 games) who played a little center (18), too. That is, he moved established left fielder Bob Meusel to right field for that year! Yes, that's worth an exclamation point. In 1922, Ruth got hurt, but still was the starting left fielder, with some games in right and a few in center (ten, to be exact, which suggests that Huggins has largely given up on Ruth as a center fielder).

Then things just got weird. Huggins starts "platooning" Ruth and Meusel in the corners. I don't mean that he benches either player, but he shifts - a lot - between who is playing left and who is playing right. In 1923, the year where Ruth has truly outstanding range factors, Ruth played 73 game in right, 7 in center, and 68 in left. In 1924, when he's not much worse, he played 99/7/50, right to left. That's the end of the center field experiment, but the "platooning" continues on - forever. Of all the candidates here in right field who played full 154-game schedules, Ruth is very likely the only one to NEVER play as many as 100 games in right (he reached 99 twice). His range factors keep drifting downwards, and so his playing time in left field drifts downward, too. This takes time. As late as Ruth's age-35 year, he was probably still a left fielder, if you count how many games he played where. It's only the late fat Ruth who ends up being a right fielder.

So his rep early suffers from being in the same outfields as two of the hottest corner gloves in the game. And then, as he ages, he drifts into more and more right field, because it's the least challenging of the two Yankee Stadium corners, and both he and Meusel have cannon arms. Overall, he has only three good years as an outfielder: 1921, 23, and 24, with 23 being the REALLY hot one. Here's a list of his young outfield seasons. I've listed the games played in left, center, and right, followed by Ruth's range factor and the league range factor, from Baseball Reference. You should be able to easily see the pattern, just from this sample:

Year Left Center Right Ruth Range League Range
1918 47 12 0 2.19 2.17
1919 111 0 0 2.13 2.16
1920 36 20 85 1.99 2.16
1921 134 18 0 2.40 2.21
1922 71 10 40 2.18 2.22
1923 68 7 73 2.69 2.26
1924 50 7 99 2.36 2.27

And it's all into right field and downhill on the range factors from then on.

My guess, based on contemporary accounts, and also on body build, is that Ruth, when young, was quite fast, but may have had trouble starting and stopping, because of his extremely high center of gravity (thin legs, barrel chest, huge arms make for much more mass above the belt than is usual). That is, he likely could not turn and throw quickly, although he could make some of that up with his arm. But his teams always had a Harry Hooper or Bob Meusel to compete with him for corner playing time,and the Hoopers and Meusels eventually won out, because they were even better than the young Ruth. What would mess with his defensive rep later was the constant "platooning." In fact, it's only the insistence on platooning that allows him to land here, in right field. His GOOD years on defense are much more weighted towards left field. I'm happy with that, because I see no point in throwing Ruth into the Williams / Musial mix in left, but the change is clearly a product of his worst defensive years. Unfortunately, he only has three good years, and over a decade of mediocre and poor ones. In Ruth's three good defensive years (21, 23, 24), he sums up to 252 games in left, 32 in center, and 172 in right.

One note that's a bit off topic. I don't know how many of you follow college football. If you do, you're probably aware that Oregon just beat #1 USC, basically by running a freshman named Jaquizz Rogers (uniform #1; his brother the wide receiver wears #8). Rogers is just 5' 7" tall and weighs 185 pounds, which is basically my size when my back allows me to be in shape. That is, the Beavers beat the Trojans behind Hack Wilson or Billy Hamilton at running back. If you want an idea of how fast and how quick such a body is, take a look at a replay of the game. It really did give me a feel for what that body type brings to athletics. Extremely quick changes in direction, due to a very low center of gravity and very strong legs. Hard to tackle, because he's hard to get under. Surprising power because there's all that strength in those legs. I would imagine that makes for a primo base stealer, especially if you're a second baseman and those powerful legs are driving spikes right at you on a close play. I also imagine it makes for less durability than other body types because it just isn't very big.

- Brock

I had heard, but do not recall the source, that Ruth was shuttled LF-RF so he would be given LESS ground to cover (small ballpark in LF, play Ruth in LF).

Brock,

The Yankees played their home games in the Polo Grounds until 1923, so Yankee Stadium's small right field wouldn't have an effect from 1920-1922.

Anyone know if there was a difference between the two outfield corners there?

My understanding is that Ruth was switched between the corners to keep him out of the sun field. Not because they didn't have sunscreen, but to preserve his eyes.

His range rankings as a pitcher in Boston really are far above the league norms.

As misleading as range factor can be for most positions, it's basically meaningless for pitchers in terms of evaluating their defensive prowess. Groundball pitchers field more balls, flyball pitchers don't. It tells you the type of pitcher, not his defensive ability.

I had heard, but do not recall the source, that Ruth was shuttled LF-RF so he would be given LESS ground to cover (small ballpark in LF, play Ruth in LF).

I believe this is correct.

Anyone know if there was a difference between the two outfield corners there?

The Polo Grounds was very short down both foul lines and got very deep in center. Not sure which field was bigger, or whether LF or RF would be considered big overall. See here.

I had heard, but do not recall the source, that Ruth was shuttled LF-RF so he would be given LESS ground to cover (small ballpark in LF, play Ruth in LF).

From 1923 - 1934, Ruth would usually play RF in Yankee Stadium, and LF in a majority of games on the road.

Then things just got weird. Huggins starts "platooning" Ruth and Meusel in the corners. I don't mean that he benches either player, but he shifts - a lot - between who is playing left and who is playing right. In 1923, the year where Ruth has truly outstanding range factors, Ruth played 73 game in right, 7 in center, and 68 in left. In 1924, when he's not much worse, he played 99/7/50, right to left. That's the end of the center field experiment, but the "platooning" continues on - forever. Of all the candidates here in right field who played full 154-game schedules, Ruth is very likely the only one to NEVER play as many as 100 games in right (he reached 99 twice). His range factors keep drifting downwards, and so his playing time in left field drifts downward, too. This takes time. As late as Ruth's age-35 year, he was probably still a left fielder, if you count how many games he played where. It's only the late fat Ruth who ends up being a right fielder.

Vaguely I feel that I "know" the Yankees "platooned" Ruth home and away.
--in order to protect him from the bigger field or the sun field? I don't even vaguely recall.
And I don't know or have time to look up how his record fits with the move into Yankee Stadium.
or where was the sunfield in


72. Blackadder Posted: September 26, 2008 at 10:35 AM (#2956630)
Damn, no edit feature. I guess my last silly comment, confusing putouts with assists, will stand for eternity.

Welcome to club.


Chris Cobb's reply on Frank Robinson
The effect of seasons with lesser playing time on the peak rate stat is one of its limitations: Frank Robinson's late rate peak is also partly because his reduced playing time in a couple of seasons after the monster year in 1966 gives more weight to that season in the 1966-70 window than it has in the 1962-66 window. However, the more drastic effect in Robinson's case is that his 1963 and 1965 seasons were subpar years. Both of those seasons, by WAR2, have rates below 5.0, whereas Robinson's lowest rate in 1966-70 is his 5.7 in 1968 (he doesn't, by this measure, look like "a shell of his former self"--guess that perception was created mostly by the Year of the Pitcher, just as it was in Kaline's case).

(I was thinking Jack Hamilton but he felled TOny Conigliaro, right?)
It was said that Robinson still suffered from some double vision next year, 1968, and that did become part of Orioles lore. Batting averages were down everywhere --and Paul Blair beside him doubled Frank Robinson's 40-point drop-- but I suppose the power outage was the main thing, 15 hr and 52 rbi. The entire deadball era was obscured some by the home runs Robinson, Yaz & Killebrew, and Frank Howard hit; there is a lot of room for decline inside of OPS+ 187, too.

Oregon STATE! Hello! Oregon STATE!

Chris, I keep forgetting that rates are (uh) rates. And Kaline's rates, probably more than anybody's, need to be seen in light of his extreme dis-durability. In his seasons of ?100 games, he averaged 137 games, less than anybody on this list except Kelly, Thompson and Keeler (doh). Oh, and Gwynn (134.5). Anyway, so here are Kaline's 2 peaks via WS.

1950s 31-26-20-23-27 = 127 1955-59 in 744 games
1960s 25-24-20-31-30 = 130 1963-67 in 690 games

So obviously his '60s rate had to be a lot better, so much so that surprisingly to me his total value in the '60s is even better. Surprisingly, maybe, because what little offensive black ink he's got came in 1955 (.340 BA and 200 hits), 1959 (.530 SA) and 1961 (41 2B). He led in IBB in 1959 and 1963. I just don't think of him as a major star after about '63 as he was overshadowed by guys like Yaz and F. Robby and Oliva, and even by teammate Willie Horton. Yet 2 of his 3 best OPS+ were 1966 (161) and 1967 (176) but in 142 and 131 games, and he never again played as many as just those 273 games in 2 consecutive seasons (the actual number is 264).

All in all I see Kaline much like Winfield. Easy HoFers with more than 400 WS, but who don't look as good as one expects on a year by year basis. OPS+ 134 and 130 in 11,393 and 12,219 AB + BB respectively. Kaline appears to have been a superlative fielder, however.

OREGON STATE?

Marc,
All circumstances considered you are expected to say something about the MINNESOTA TWINS! Only just now I learned from the Mets-Marlins radiocast that 6-1 Chicago turned into a Twins sweep.
OK, I understand waiting. I will be away for a few days and I hope to read it on Monday night.

My understanding is that Ruth was switched between the corners to keep him out of the sun field. Not because they didn't have sunscreen, but to preserve his eyes.

Someone probably gave that as an excuse, but in EVERY AL park except for League Park, RF was the 'sun' field.

Full Prelim Ballot

1. Ruth
2. Aaron

3. Ott
4. Robinson

5. Kaline
6. Waner
7. Rose
8. Crawford
9. R. Jackson
10. Clemente
11. Gwynn
12. Kelly
13. Heilmann
14. Slaughter
15. J. Jackson

16. Flick
17. Keeler
18. Evans
19. Winfield

20. Thompson

It looks like a career-heavy ballot, but the fact is that everybody 1-12 has both peak and career, so even though Heilmann, Joe Jackson, and Elmer Flick have very strong peaks, they don't exceed the peaks of players like Clemente and Gwynn enough to justify placing them on the top half of the ballot. Their peaks of these shorter career guys are good enough to put them ahead of the career guys without strong peaks--the Keeler/Evans/Winfield trio.

Hey, Dan R!

Is Kaline's (limited) time at first base in his 68-73 seasons accounted for in his numbers in your main spreadsheet? If his first-base time isn't accounted for, is there enough of it, in your judgment, to justify doing a manual adjustment?

I ask because Kaline and Waner are extremely close in my system--separated by .6 wins--so even a small downgrade for Kaline would drop him below Waner. Waner has some first-base time himself (oddly enough, it is from early in his career) but Kaline has roughly twice as many career games at first base: 135 to 73.

Your thoughts?

I wonder if Clemente lost much development due to the Dodgers' handling of him as an early blue-chip prospect who they desparately "hid" from other clubs by not playing him when other scouts were around.

Stew Thornley's published research concludes that Clemente was not being "hidden" by the Dodgers.

-- MWE

And he couldn't have lost much development- he only played 1 year in the minors.

Ruth played RF in Yankee Stadium and Griffith Stadium, which also had a huge LF. In the other parks he played left. Well, looking at the numbers, he must have played right in another road park, but I don't know which one. Bob Meusel wasn't keeping him out of LF in 1920; Meusel played only 16 games there. Duffy Lewis was the Yankees' regular LF. And Ruth last played a majority of his games in 1926, at age 31, except his 2 months with the Braves.

Possibly Sportsman's Park as the other Ruth in RF stadium? Much bigger LF than RF.

1. Sorry, Sunnyday. I got the Ducks and the Beavers mixed up.

2. Thanks to everyone for the info on Ruth's defensive assignments. Where do you find all this stuff out? The idea that he was platooned by ballpark makes sense. One of the things that would have enabled that was that both Ruth and Meusel had good arms. Huggins didn't have to play Ruth in right. Also, yes, I missed the one Duffy Lewis year. On the other hand, given Lewis' defensive rep, I don't think it affects my point any. - Brock

Ah, no. Jackson and Allen had radically different personalities.

Fans don't nec know that. A lot of people hated Reggie Jackson.

But anyway the question is how a guy with a 140 OPS+ over 11,000+ PA and 444 WS ends up 12th on ballots behind guys @ 130 over fewer PA and 50 fewer WS. Maybe it's the low BA. I dunno.

My niece attended Oregon STATE and in fact worked on the football athletic training staff as an undergrad, so she is pumped.

Chris C. took issue with a couple of my posts which is fine. His ballot reinforces the central difficulty of this ballot. An obvious group at the top, an obvious group at the bottom (though he underrates Sam Thompson). But the central difficulty is that group from 5 to 15 in the middle that is tough to peg. Then there's the fact that Chris is more of a career voter than I am and so he's got Pete Rose 7th and Joe Jackson 15th. I obviously don't agree with either of those.

But the big question for me is:

5. Kaline? 134 OPS 11,000+ PA 443 WS w/3 yr peak 31-31-30
6. Waner? 133 OPS+ 10,000+ PA 423 w/36-34-32
9. Reggie 140 OPS+ 11,000+ PA 444 WS w/41-32-32

I mean, for the life of me I can't see Reggie behind these guys at all, but at worst how about 5-6-7? Well, of course, there's Rose at 7, and then I happen to agree that Chris is right to have Crawford up in this group at #8.

To me this next group is more like:

Reggie
Joe Jackson--but I understand people who have him lower
Crawford
Waner, Gwynn/Gwynn, Waner?
Rose--but I understand people who have him higher
and then Kaline

Kaline and Clemente seem pretty interchangeable, too, though I too like Kaline a little better. But Chris has them 5 and 10, a pretty big split. Well, like I said, difficult ballot in the middle. I don't strongly object to any of it except I don't understand Reggie down where he is.

Chris, the 1B time most definitely is not accounted for. But using better fielding estimates will probably make a bigger difference than just fixing the positional mix. For Waner, all we have in addition to FRAA/FWS is DRA, and DRA l-o-o-v-e-s Waner. It has him at at a quite monstrous +254 runs from 1926-36, before he turns below-average. Now, DRA has a very high standard deviation in that time period, but even a heavy dose of regression to the mean suggests Waner was a simply extraordinary outfielder. Even though I already give him about +11-+12 wins for that period myself, that might be selling him a bit short. By contrast, TotalZone simply confirms my current line on Kaline as a +7 win OF.

How good is Waner's fielding reputation? Good/great/otherworldly? If anecdotal evidence supports DRA's love affair with Waner, that'd definitely be a reason to put him ahead. Otherwise, I'll crunch a new set of numbers.

Kaline is probably going to drop from my prelim, both on the basis of additional information about his fielding value and about Waner's, and because I think his durability is a bit more of an issue than I have give it credit for so far.

Waner is staying put. I see his value as clearly, though not greatly superior to Jackson's for several reasons, none of which are reflected in the summary evidence sunnyday2 presents.

6. Waner? 133 OPS+ 10,000+ PA 423 w/36-34-32
9. Reggie 140 OPS+ 11,000+ PA 444 WS w/41-32-32

First, OPS+ overstates the difference between the two in offensive value.
WARP's EQA1 has it Jackson .306, Waner .310. EQA2 has it Jackson .307, Waner .305.
Dan R's WAR2 has it Jackson 59.3 BWAA, Waner 60.3. So when you start looking at the fine print in offensive value, Waner pretty much closes the gap offensively.

Second, win shares overstates the value of Jackson's top season, which came in the 1969 AL, which was the expansion version of a weaker league. That season needs to have some air let out of it.
WARP1-2 shaves off 1.5 wins; Dan R's WAR1-2 shaves off .5 win, so let's split the difference and call it 3 win shares. Jackson then at 38-32-32 is about even with Waner at 36-34-32 (assuming that Waner doesn't get a bump to 38-35-33 for converting from 154 to 162 game seasons)

Third, Waner's fielding value is substantially higher than Jackson's, by every measure. Once one pares down Jackson's apparent offensive advantages, Waner's defensive advantages put him decisively ahead. Jackson was a good outfielder in his youth, but he was a bad outfielder and then a DH later in his career. Waner was a great outfielder in his youth, and became a bad outfielder in his decline at about the same point in his career where Jackson became a DH.

I put more weight on fielding value than sunnyday2 does, so I don't expect us to agree on this point, but I have no second thoughts about ranking Waner as highly as I have.

The non-consecutive peak voter will probably see little difference between Kaline and Clemente (though the WS peak voter will underrate both). The consecutive peak voter will probably prefer Clemente, whose 1966-70 was a stronger set of seasons than any Kaline produced. For the career voter, Kaline is ahead of Clemente because he didn't have five years of below-average play at the beginning of his career.

The players occupying the 5-15 spots on this ballot are so close that small differences in ways of looking at merit will lead to substantially different orderings. I think, in the end, some ranking trends will be clear, but we won't have agreement.

At one time I planned to do something with this (write some useful and interesting prose accompaniment). Probably I won't get to it.

League Rank by OPS+


The table covers most of the RF candidates plus Sheffield and Sosa at the bottom.

Left to right:
- year of HOF induction
- baseball age at league top ten by OPS+
- sequence of annual league ranks by OPS+ spanning the first to the last top ten rank. Position represents baseball age with the colon (:) at age 18 (eg, Mel Ott alone was a top ten batter at baseball age 19, namely rank 8).
: T = rank ten
- (black ink, grey ink, hof monitor, hof standards)
: - = not in top ten
: . = military service or war work
- Name severely abbreviated
- span of years with league top ten rank; that is, the years spanned by the code sequence

The biggest surprise to me is Paul Waner's relative absence from high league rank

Mike E, thanks for the info - more solid than the anecdotal I had heard b4.

Dan R asks (post 96), "How good is Waner's fielding reputation? Good/great/otherworldly? If anecdotal evidence supports DRA's love affair with Waner, that'd definitely be a reason to put him ahead. Otherwise, I'll crunch a new set of numbers."

This one I know, if we're dealing with reputation only (I haven't crunched any numbers). I must have heard or read close to 50 accounts of Paul Waner's defense from contemporaries. They don't vary. At all. Here it is:

Paul Waner was a center fielder. The only reasons he didn't play there were:

1. His ballpark had a large right field that needed a second center fielder. (I don't know if this is true, but it is the unvarying reputation.)

2. He came up when the Pirates had Max Carey in center, and then Lloyd Waner came up, who was an even better center fielder than Paul was, or even Max. (This is absolutely true.)

If you just go by reputation, it's the same as Fred Clarke or Jimmy Sheckard. Bill James' Win Shares don't see all that value, and contemporaries tend to remember a player at his best (see Oscar Charleston), but Waner's contemporaries would have been anything but shocked to find him ranked with Sheckard and Clarke, or even any center fielder short of Speaker.

One reason that I'm putting in all the caveats here is that the big reason I remember the reputation so well is that it was usually accompanied by, "The two Waners and Kiki Cuyler were the greatest defensive outfield in baseball history." Well, that outfield only existed for a part of 1927. But then, that was a big pennant-winning year for the Bucs. But the fact that a half-year wonder comes along with the Paul defensive rep makes me nervous.

- Brock

1. His ballpark had a large right field that needed a second center fielder. (I don't know if this is true, but it is the unvarying reputation.)

In Pittsburgh, this was more true of left field than it was of right field. The grandstand in right cut the coverage area down considerably. Right field wasn't "small", mind you, as the grandstand angled out pretty quickly, but it wasn't as tough as was left field.

Andrew Clem's Web site is a good reference.

-- MWE