Prelim:

No new comers this week so two that fell off have returned.

1. Willie Keeler (3,2,x)

Also my personal HoM inductee this year, barring any changes this week.

2. Bob Caruthers (4,4,9)

This year should provide a bit of a breather--no offense to Mr. Powell and even Mr. Donlin, one of the great coulda-been-somebodies of baseball history. At his best, he was perhaps Fred Clarke or Joe Kelley, not quite Elmer Flick. But at his worst, he made Dick Allen and Albert Belle look like a couple of choir boys.

But anyway, about that breather. The trouble is I've still got to figure out what to do about WARP. Some of the discussion has helped--e.g. Paul Wendt's post showing that the new WARP scores actually correlate better with WS. I'm comfortable knowing how and how much to rely on WS, so maybe that's a good thing. But at a minimum, I remain convinced that as far as I do consult with WARP, adjWARP1 (WARP1 adjusted for season length) is the number to use, not the published WARP3.

And, secondly, I'm pretty sure that rate stats, generally, are like ephedra. More apt to be abused than other substances. So I am wary of rate stats.

The more I've looked at different methods, the more I think the PA are really useful, because they integrate peaks and careers nicely, at least at a conceptual level. Unfortunately, I could not find the PA formulas anywhere on the Internet so I can't calculate them myself.

So, no, no breather this year, probably more shifting around among my top 15.

Re: Jack Powell. I'll just re-post the comment I made earlier in the pitcher's thread about his run support & how that impacted his W/L record.

Comment #1:

No surprise - his support stunk. From 1901-onward he started 369 games & had an RSI of 91.49. The only pitchers I know of with worse RSI & more starts are: Dazzy Vance (90.47), Bobo Newsom (89.84), Mark Langston 90.56), & Steve Rogers (88.99). He's just ahead Tom Candiotti (91.51). To be fair, I haven't RSI'd everyone over 369 starts, but I've gotten most of them (I figure I got 15-20 20th century pitchers to go). Year by year for JP:

1901..110

There will be some revisions in pitcher WARP3 quite soon.

revised scores; recalculation rather than change in design

Joe,

Sounds like a good idea, but we'd run into the wall of my computer illiteracy fairly quickly? . . . Um, what's text formatting? I'm not sure what that is let alone how to do it. I've got a brother who does IT for a living, so I could ask him, but since he's got a 2-year-old son it may take a while before we could sit down & thresh it out.

I do actually have some of the RSI stuff up on a blogspot website (see homepage), though navigating it will probably be a major pain in the butt. I've got the season-by-season RSIs for almost everyone in the HoF & Bill James's top 100 pitchers listed there. I was providing links to each individual pitcher in the career leaders section but for some reason (no idea why) after I certain point in time, when I put the links in, they wouldn't show up on the website. My guess is that it's related to a change I made in archiving, but I really have no idea. And then I just got sick of typing in a bunch of info that I already knew, so that's why I haven't touched in a a month-plus. (And in that time, I've done another 40ish pitchers & figured out a W/L system that I like. Ah well).

If I can figure out the computer-ese & have the time I'd be willing to set something up here. If anyone's masochistic enough to do it for me I'd be willing to e-mail them my Big Giant Excel File of RSI.

Ixnay on the new candidates! None of them deserve any ballot space with all the returning candidates available.

Prelim ballot

I have just realized that the top 10 guys on the ballot this year will likely get in due to the talent scarcity between voting years 1924 and 1932, so I am trying to make sure I get their order right. I'm sue most people know this but the people that finish in the top 10 or 12 in this year's voting will almost all certainly go in. So the placement of guys like grant, duffy, pike, waddell (off my list), ryan (also just off) and beckley is more important than ever.

1. Wee Willie ? top 10 in hits 13 straight years, top 6 in runs 12 of those 13 years. ~3000 hits, subjective opinions solid. Only no brainer on the ballot for me (if I had the option of a small hall).

For reasons mentioned above, this is definitely prelim, but I wanted to get on here to counter the momentum that Wee Willie seems to be gathering (in the two prelims above) ;-)

1. Dickey Pearce (3 last year)--no need to worry about WARP too much with Dickey (no ability to worry about WARP with Dickey), I like that

Since there really aren't any new candidates that will make my ballot, I decided to focus on a question that has been bothering me for a couple of elections now. Frank Grant vs. Dickey Pearce. I have both Grant and Pearce on my ballot, but Grant has been comfortably ahead of Pearce since Grant came on the ballot. I really don't understand that. I realize that some of you have problems with pre-1871 baseball or undocumented play, or that some of you may have a severe timeline.(I really don't timeline. I'm one of the penant-is-a-penant crowd), but even so I can't see having Grant over Pearce. I realize that most of their careers are undocumented, but here is a comparison.

Grant - 18 year career

I have both Grant and Pearce on my ballot, but Grant has been comfortably ahead of Pearce since Grant came on the ballot.

Sorry if this sentence isn't clear, I meant that Grant is comfortably ahead of Pearce in the voting totals, not on my ballot.

My 1918 prelim starts like this:

Joe, that sounds reasonable. Might take a while to get going (with computers I usually have to screw things up 10 different ways before getting it right, even if the task is simple & I've been given clear instructions). I assume if I do screw something up there'd be some way to go back & fix the pre-tags, right?

I'm still not seeing what's special about Keeler.

He'll be back at the bottom of my ballot for the next election, so I'm not a huge fan of him, either. But I can't see the fourth best leftfielder of his time (Kelley) above him, also. Rightfielders are the most underrepresented so far in the HoM and need a few more representatives.

11) Pearce?hard to place. Hard to spell.

BTW, I always smile when I read this. :-)

1) Waddell?so much to say, so little time.

All right, Jim, I'll call you on this. This is the Monday of a discussion week; there's plenty of time. You were the only voter to have Waddell as high as 4th last week (there were two voters who had him 5th), and from last week to this week you moved him up one notch by putting him ahead of Kelley.

How about the long version? We've got time to listen.

My 1918 Prelim... I moved some pitchers up. My ballot was not kind enough to pitchers.

1. Hughie Jennings (2)

Jake Beckley may not have been a star of the first magnitude, but he sure as hell wasn't mediocre either.

There is no way you could argue otherwise. Aptly said, Joe.

Jake Beckley may not have been a star of the first magnitude, but he sure as hell wasn't mediocre either.

There is no way you could argue otherwise. Aptly said, Joe.

Re:RF

Joe, a couple of elections ago, you took exception to something that I wrote about A-Rod. While I was using it as an example and not necessarily using it within my ranking system, there was an underlying truth, nevertheless.

Let's really be extreme and ridiculous: every team now has a player who has the talent of Rodriguez at short for 2004. Not only that, each 29 clones (plus the original) have the identical park-adjusted offensive and fielding stats. Not only that, they all hit ten times better than Barry Bonds and fielded ten times better than Ozzie Smith.

Now the question: have they helped their teams? The answer is no. They haven't hurt their teams, either, but they canceled each other out when they played against each other. WARP or Win Shares will tell you different, but they are wrong.

Is this the way to go when ranking players? Well, it's one way. I don't say it it's the only one. I'm not 100% wedded to it myself, though I do lean in that direction.

One thing that I have learned while working with this project is that there isn't necessarily one way to rank these players. It isn't as easy as I thought it was going to be initially.

Just something to think about.

In Defence of Keeler:

Jesse Burkett is Keeler's most similar player and while Burkett is quantifiably the better player, there is not a large gulf between them. Keeler is even Burkett's third most similar player. Similarity scores leave a lot to be desired but their career totals by most measures are close and the shape of their careers are not too different either.

Yet Burkett was elected overwhelmingly on his first real shot (1912) and Keeler is being left off several ballots. And Keeler is the one in the underrepresented position. I can see placing Keeler in the middle of the ballot if you favour peak over career, but I just don't see the same people who put burkett 1 or 2 on their ballot in 1912 (36 of 42 voters) leaving keeler off entirely.

Does anyone see an overreaction that militates against Hall of Famers with good traditional counting stats (see, for example, McGinnity and Keeler)?

Now the question: have they helped their teams? The answer is no. They haven't hurt their teams, either, but they canceled each other out when they played against each other. WARP or Win Shares will tell you different, but they are wrong.

John, I think your reasoning here just doesn't make sense: you're carrying position-limited comparisons way too far. Of course these hypothetical A-Rod supermen helped their teams! They were, we can assume, the best player on their teams, and _if_ any one of those teams had failed to start that player, the team would have suffered a lot. No one of them helped their team _more_ than any other one of them did, but they all helped their teams more than any other player on their teams. Just because they all played the same position doesn't mean their value can be compared _only_ to the other players at that position. Shortstops don't play shortstops. Pitchers face hitters, and defenses face hitters. If one team happened to have Nomar Garciaparra, who could also play a mean shortstop and hit a bit, but had Marquis Grissom pencilled in for centerfield, that one team might decide to switch A-Rod Superman to centerfield and played Garciaparra at short. Would that A-Rod Superman centerfielder be tremendously more valuable than the other A-Rods, because his value would no longer be cancelled out by the others? Of course not. Yes, he would be the best centerfielder in the league, but that doesn't make him a more valuable player than the other A-Rods.

>>"Those who like a looong run of mediocrity (i.e. Jake Beckley) should like Smilin' Mick; a long run of mediocrity is vastly more special from a pitcher"

>That's a ludicrous statement. A long run of mediocrity is Tom Brookens. Or Doug Glanville. Jake Beckley may not have been a star of the first magnitude, but he sure as hell wasn't mediocre either.

My dictionary says mediocre means "middling, ordinary, neither very good nor very bad, average." I think that describes a guy who earned 18 WS or more 5 times in 20 years. Tom Brookens earned 84 WS in 12 years, I can't imagine he was ever anywhere near 18; 10 maybe. He was above average acc. to TB once. Glanville was much better, above average twice in his first 4 years; the jury is still out on him.

But Brookens never came close to being mediocre, he clearly was not that good. I don't disagree with your eval. of Jake Beckley but with your understanding of the word mediocre. Now "ludicrous," that would be "causing laughter because absurd." That would be more like Wid Conroy. I was gonna say Willie Keeler #1, but that would be more "mediocre" than "ludicrous" at least based on the prelims so far. (Get it, an "average" ballot?)

>There is no way you could argue otherwise.

Way.

My 1918 Prelim... I moved some pitchers up. My ballot was not kind enough to pitchers.

1. Hughie Jennings (2)

Huh. 86 minutes between post and double-post, easily the longest lag I've seen. Weird.

Does anyone have a sense of how many HoMers should be playing (on average) at any given time?

I've calculated how many HoMers are at least semi-regulars (appeared in at least half of the team's games or had 100 IP) in any given year.

The totals run from 8(1971) and 9 (1872 and 1877) at the low end, fluctuating through the low teens in th 1870s, going through a plateau in the 1880s (19 or 20 each year from 1881 to 1889, with up to 3 being in the AA) to a high of 25 (1890, 11 NL, 14 PL), 23 (1892), and 22 (1891, 19 NL, 3 AA) before tapering off in the post-1893 era where lots of guys are not eligible yet.

There are still about a half dozen eligible/likely players who played in 1890 (Bennett, Grant, Caruthers, Thompson, Duffy, Van Haltren. Also Beckley and Browning.)

When we get to 1990 and look back, will there be about 30 active regulars or semi-regulars in any given year, or will 1890 stick out like a sore thumb? If so, are the 1870s grossly under-represented (one third the players of the next generation)?

I'm trying to decide whether I should start giving an 1890 demerit to players who are not one of the 25 best players in the Majors that year. Or maybe add an 1870s bonus to Pike, C. Jones, Bobby Mathews.

Just thinking out loud . . .

The following is (c) Howie Menckel. It took a long while to find it (1915 results thread), so I moved it here for easier access. Maybe Howie is updating, but I did take the liberty of adding Stovey, Young and Chesbro. (Er, I mean Clarke. Sorry Hap.)

By Howie's reckoning, then, we have 10 players who started from 1859-1872; 14 from 1878-1882; and 9 from 1888-1894, though it's too soon to know whether 1888-94 will end up being a useful category or not.

But along with positional balance, there's chronological balance. No way are the '80s gonna end up getting shorted. And I'm guessing that in 1917 we had Kelley, Keeler, Flick, Collins and McGinnity among the top 6 runners-up. The '90s are lookin' pretty good. To those who say "enough" from the pioneer era, why?

MattB just made my point for me, and better, too. And re. timeline, a ratio is two numbers. 8 ML teams in 1880, 24 in 1890.

My view is that Powell and Williams are the two worthwhile newcomers. Donlin is Tip O'Neill II : Return Of The Son Of Tip O'Neill. To be honest, Donlin and O'Neill are probably good enough to be considered along with my 22-30 guys.

That said, Donlin was probably a better player than Powell or Williams, but he's more of a known quantity. Powell is a worthy player, nice long career, but he's not as good on any measure of actual greatness as any of the McCormick/Mullane/Welch trio. If Welch can't make my ballot, Powell is miles away.

Clarence Williams in a capsule...

* Played on a large number of the very best Negro League teams between 1885 and 1910 or so (Cuban Giants, X-Giants, Philly Giants, etc.).

* Still playing in 1910 as the starting catcher on the Brooklyn Royal Giants, a sign of very high quality (would have been in his mid-forties).

* Highest-paid player on the Cuban Giants in 1886.

* One of the first players to come out of the blackball factory that was Harrisburg, PA.

* Well known as a defensive catcher, mostly as a game-caller, but even better-known as a showman.

* Behind the plate for some remarkable games (Dan McLellan's perfect game for the X-Giants in '03, Foster's World Championship clincher in '03, famous Billy Whyte game vs. Detroit in 1887)

* Hit .256 for Harrisburg in the Eastern Interstates League in 1890 (playing 3B and C - Frank Grant hit .325 for Harrisburg, Sol White .358 for Cuban Giants) After 1890, though, Williams and Sol White tend to travel together.

My verdict on Williams is that he was a defensive catcher with a long career, but nothing special as a hitter. The sort of player that successful teams find irreplaceable; someone who knows how to work with a good pitcher. But offensively the little evidence we have isn't encouraging. I wouldn't consider him a candidate.

Now the question: have they helped their teams? The answer is no. They haven't hurt their teams, either, but they canceled each other out when they played against each other. WARP or Win Shares will tell you different, but they are wrong.

John, I think your reasoning here just doesn't make sense: you're carrying position-limited comparisons way too far. Of course these hypothetical A-Rod supermen helped their teams! They were, we can assume, the best player on their teams, and _if_ any one of those teams had failed to start that player, the team would have suffered a lot. No one of them helped their team _more_ than any other one of them did, but they all helped their teams more than any other player on their teams. Just because they all played the same position doesn't mean their value can be compared _only_ to the other players at that position. Shortstops don't play shortstops. Pitchers face hitters, and defenses face hitters. If one team happened to have Nomar Garciaparra, who could also play a mean shortstop and hit a bit, but had Marquis Grissom pencilled in for centerfield, that one team might decide to switch A-Rod Superman to centerfield and played Garciaparra at short. Would that A-Rod Superman centerfielder be tremendously more valuable than the other A-Rods, because his value would no longer be cancelled out by the others? Of course not. Yes, he would be the best centerfielder in the league, but that doesn't make him a more valuable player than the other A-Rods.

Huh. 86 minutes between post and double-post, easily the longest lag I've seen. Weird.

Chris Cobb's got you beat: 118 minutes.

Marc,

Does anyone see an overreaction that militates against Hall of Famers with good traditional counting stats (see, for example, McGinnity and Keeler)?

To his credit, Marc has given good reasons for his skepticism about Keeler. I'm not sure why others have been bashing Wee Willie so much. Perhaps a combination of not realizing how much the extreme deadball era hurt Keeler's traditional stats in the latter part of his career, and the legacy of Palmer and Thorn's Total Baseball Rating, in which Keeler's rating is unreasonably low, due to being compared with league average performance, as opposed to "replacement level" performance.

By the way, I think that Keeler is a no-brainer HoMer.

Jim Spencer - good argument. I have no large disagreements with it, just three things, each of which is small in itself. I don't lean as far towards pitchers as you do. In six of the seven years you cite, Waddell's RA+ was lower than his ERA+ - not by much, but it makes a small difference. One of those 7 years was the 1902 AL: clearly a stronger league than the 1901 AL, but I don't think quite to full competitive strength yet. Add it up, and I have Waddell just behind McGinnity, with a few outfielders ahead of them.

And the strikeouts are spectacular. Note that bbref's similar pitchers include some modern names: Mike Cuellar, Kevin Brown, even Tom Seaver. He was not typical for his times.

1918 prelim.

1. Elmer Fudd- gweat peak

Does anyone have a sense of how many HoMers should be playing (on average) at any given time?

If the average HOMer is a regular for 15 years or so, then 200 HOMers playing for 15 season is 3000 player-seasons. There are about 2400 team seasons, so this works out to about 1.25 HOMer seasons per team.

About 10 per season in the 8-team 70's would be about right.

One of those 7 years was the 1902 AL: clearly a stronger league than the 1901 AL, but I don't think quite to full competitive strength yet. Add it up, and I have Waddell just behind McGinnity, with a few outfielders ahead of them.

Valid criticism, but in Waddell's defense IIRC he led the NL in ERA+ in 1900 AKA the Year of Only Eight Teams. Went 8-13 that year too. Ouch. Don't know what his run support was that year but I'm guessing it sucked.

One of those 7 years was the 1902 AL: clearly a stronger league than the 1901 AL, but I don't think quite to full competitive strength yet.

Actually, if you believe BP's DERA "adjustments", Rube flowed with the tide, moving from the strong NL of 1901 to the strong AL of 1902-5. The AL did an excellent job of raiding a large number of quality players from the NL during 1901-03.

Here is an All-Star team of players that moved from the NL of 1901 to the AL of 1902.

Ca McGuire/Schreckengost

Eric:

>(1) his record was greatly helped by playing on good hitting teams, (2) his record was greatly helped by having good defense behind him (3) how well a pitcher hits doesn't really matter that much, (4) there is evidence that the AA was weaker relative to the NL than is commonly believed, (5) competitive imbalance at the individual level was at a historically high level during his best years. Note thatpoint #5 severely affects most sabermetric analyses (e.g. WARP), as well as skewing traditional statistics. Please, people, use some common sense here.

What does point #5 mean? For that matter, what does #4 mean? What evidence are you holding back from the rest of us?

I also don't necessarily accept #1, 2 or 3, either. It's not like Bob's ERA+ is 97 or anything. Maybe Chris can remind me what his adjusted record would be. And maybe Chris can also remind me whether, when we adjust Bob's record, we deprive him of his own hitting?

OTOH, I love Elmer Fudd and his "gweat peak." And while Jake Beckley clearly was "average" (which is synonymous with "mediocre), it is "ludicrous" to say that Willie Keeler was "mediocre." Clearly he was above "average."

And I think Bennett and Bresnahan are pretty comp, though Bennett had a lot more value as a catcher. Check my post (about 10 days ago, I think) on the Catcher thread.

OCF: One of those 7 years was the 1902 AL: clearly a stronger league than the 1901 AL, but I don't think quite to full competitive strength yet. Add it up, and I have Waddell just behind McGinnity, with a few outfielders ahead of them.

Chris J.: Valid criticism, but in Waddell's defense IIRC he led the NL in ERA+ in 1900 AKA the Year of Only Eight Teams. Went 8-13 that year too. Ouch. Don't know what his run support was that year but I'm guessing it sucked.

Probably his run support did suck, but there's another factor here. 1900 may represent the extreme instance of a larger tendency already mentioned in Waddell's career. His RA was much worse, proportionately, than his ERA. There aren't league-wide rankings for RA+, but just comparing Waddell to his own team in 1900 highlights the issue.

Pitcher -- ERA -- RA -- RA/ERA -- RA-ERA

Guys, disregard post #33. I don't even know what the hell I was trying to say. Just toss it off to lack of sleep, an undigested potato or something. :-)

John, yes, those undigested potatoes are murder. Hope you're better. ;-)

What does point #5 mean? For that matter, what does #4 mean? What evidence are you holding back from the rest of us?

I also don't necessarily accept #1, 2 or 3, either.

Marc- Fair enough. On the other hand, I don't accept the premise nor the conclusion of your comment on Caruthers: unique is another way to say uniquely valuable .

I certainly owe the group a more detailed explanation of my comments, to show that they are aimed toward serious discussion of the merits (or lack thereof) of Bob Caruthers, rather than being a barrage of unsubstantiated attacks. While I haven't had time yet to write up a long version of my arguments, I'll say more here about point #5: competitive imbalance at the individual level was at a historically high level during his best years.

Everybody who uses advanced sabermetric methods to compare players from different decades ought to read the Stephan Jay Gould essay "Why nobody hits .400 anymore", reprinted in Triumph and Tragedy in Mudville (W.W. Norton & Co., New York (2003)). The graph of the standard deviations of batting average is illustrative. The highest standard deviation ever is in 1886. The decade-average "coefficient of variation" of batting average is 19.25 in the 1870s, 18.45 in the 1880s (something like 23 in 1886!), 15.60 in the 1890s, 14.97 in the 1900s, 13.97 in the 1910s, 12.70 in the 1920s, and then remains between 12 and 13 in subsequent decades. Now, the baseball talent of major leaguers is not actually shaped like a bell curve, and batting average does not correlate perfectly with run creation; nonetheless, the coefficient of variation of batting average is a good proxy for the variation among individual players in run creation.

What this means in English is that a good player in the 1880s had an environment

Re:

>What this means in English is that a good player in the 1880s had an environment

<i>If the average HOMer is a regular for 15 years or so, then 200 HOMers playing for 15 season is 3000 player-seasons. There are about 2400 team seasons, so this works out to about 1.25 HOMer seasons per team.

About 10 per season in the 8-team 70's would be about right.

I was surprised and disappointed that Sol White only made 6 ballots in 1917.

Those who downgrade Frank Grant claim that he wasn't on the list of enough experts as a top Negro League player.

Sol White was named 35th on the SABR Top Negro League Player list, which makes him the first eligible player to make the list. So he doesn't have the "experts don't love him" knock. I was expecting White to finish tied with or above Grant.

He was comparable to Grant in recorded stats, played longer, and was considered the best player on his team wherever he went.

Why no love for Sol?

Sol White was named 35th on the SABR Top Negro League Player list, which makes him the first eligible player to make the list. So he doesn't have the "experts don't love him" knock. I was expecting White to finish tied with or above Grant.

He was comparable to Grant in recorded stats, played longer, and was considered the best player on his team wherever he went.

Why no love for Sol?

White's great career length appears to have been a product more of his managerial role than his playing ability. My understanding is also that he was mostly playing first base in the 1900s while he managed. He doesn't have, overall, Grant's reputation as a defensive star. He doesn't have Grant's reputation as the best black player of the nineteenth century, or even Grant's reputation as "the black Dunlap." I'd welcome a more exact account of his career and his reputation that provides evidence that he was an outstanding player late in his career, that he was a defensive star, etc.

There hasn't been any thorough discussion of his statistics in comparison to Grant's, but my interpretation of them is that, while they may be comparable, they are not, on the whole, equal. White doesn't have any high minors play comparable to Grant's IL stint, and nobody has shown a basis for assessing the strength of his leagues or for comparing his statisics to those of contemporary major league players. I suspect we could do better here. I'm certainly open to arguments based on a careful analysis of White's statistics, but what I've seen so far doesn't justify ranking him with Grant.

In interpeting White's ranking in the SABR list, it's unclear to me how much of White's positioning has to do with his instrumental role in the early African-American professional game and how much has to do with his on-the-field skills. If the grounds for his ranking were clear, I would give somewhat more weight to it. I have similar questions, actually, about Rube Foster, though expect that when I set out to study his record there will be a pretty large body of evidence of his accomplishments, compared to the African-American stars of the nineteenth century.

To sum up, I haven't seen any thoroughly documented arguments making White's case, and what I've found myself hasn't convinced me that White's case is as strong as Grant's.

Is there any systematic reason why Rube Waddell's RA+ should be worse than his ERA+? Some of you might think that's counterintuitive - shouldn't a strikeout pitcher be less dependent on defense? But the gap between ERA and RA isn't defense in general, it's some narrower slices of that.

The effect really isn't very large and doesn't require heroic explanation. Waddell's RA+'s are very, very good - just not quite his ERA+'s.

I can think of two possible explanations. Both of them are rooted in the idea that it takes several acts to score a run. An error wouldn't be a run all by itself, although it would often be an out and a baserunner. Rather than focusing on the error, let's look at what happened before and after the error, particularly after.

Everyone stole bases. We know in our time that different pitchers can vary quite dramatically in their ability to hold runners and prevent the SB. It's really a skill separate from pitching, although there is an association in which some of the worst pitchers at holding runners were extreme strikeout pitchers like Ryan or Gooden. I don't know whether Waddell might have been bad at holding runners, but it wouldn't surprise me if he were. The only problem with this is that the evidence of it probably wouldn't show that much in SB but rather in CS or SB% - statistics that we don't have. A pitcher who allowed more baserunning would be a little more likely to allow a runner who reached on error to score later in the inning.

The other explanation is the emotional one. We know that Waddell was tempermental and distractable. Perhaps he reacted just a little worse to having an error committed behind him than the average pitcher, and lost a little more of his concentration than average.

I was working on Mike Donlin's statistics last night. I didn't know that much about him but the gaps and holes in his career cried out that there has to be a story. Bill James devoted nearly three full pages of the NBJHBA to telling that story. What other player that we've ever considered missed most of a season because he was in jail?

What recent sports figures did his tale remind me of? The name that kept coming to mind was Dennis Rodman. Donlin was raised by foster parents (as was Rodman). Donlin was a drunk, probably an alcoholic. He was frequently at odds with umpires and was ejected from more than his share of games. He was flashy, flamboyant, arrogant on the field. ("Turkey" was for the way he carried himself.) He loved parties. He loved actresses. He loved show business and retired in his prime to go into show business, although he tried coming back to baseball later. But for all that baggage, when he played he was a very effective player. How much of that sounds like Rodman to you?

Strictly as a player, he's a big hitter but very short on career playing time. My personal rating for him is to put him ahead of George Stone (whose career was even shorter) but behind my favorite leadoff hitters, Thomas and Hartsel. Thomas appeared on three ballots (including mine) last year and Hartsel on just one (not mine). Stone has never been on anyone's ballot. Is Donlin worth a "personality" boycott? After all, Ban Johnson boycotted him, letting it be known that he was unwelcome in the "clean" AL after his assault conviction. Maybe, but he doesn't make the ballot anyway.

James puts Donlin as the LF on his all-decade team for the 1900's, which is a little surprising (couldn't you move Flick from RF to LF?), but he also targets Donlin with one-line shots under "drinking men" and "a better ballplayer than a human being."

There was a link to this chart from tangotiger's Primate Studies.

OCF #65

Paul re:

"Marc #42, based on OCF

Paul re:

"Marc #42, based on OCF

Damn, the "Stop" button didn't work.

(3) how well a pitcher hits doesn't really matter that much,

Marc took on a couple of other points, but this was the one that stood out for me. Why not? He had over 200 PAs during his peak years, and topped 400 in 1887. I think that's enough to matter. It was enough for him to accumulate the offensive value to push his WARP into the territory it is. I know you have issues with WARP, but I don't think they relate in this case.

In the 1980s, a pitcher's hitting may not have mattered much. But I think in the case of Caruthers, it does.

I also want to note that I do realize there are good reasons not to vote for him, even if I don't believe them. I just think Caruthers might be getting shortchanged by voters underestimating the impact of his hitting.

"Marc #42, based on OCF

"Who was the worst player to date with 15 seaons as a regular player?"

Ossie Bluege?

One of his "Most Similars," Mark McLemore, may be worse, though.

Ossie Bluege?

He was actually a good third baseman, though he looks inferior to the post-Mathews versions from that position.

Ed Kranepool might be the worst player to date with 12 seaons as a regular player from the last century.

Schreckengost - not from NL1901

You're right. I misinterpreted teamID 'BOS' in the database to mean Boston NL instead of Boston AL in that instance. My bad. To rotate at catcher with McGuire, they've still got Ed McFarland (or John Warner or Mike Kahoe).

Well, of course, no one with a 15 year career as a regular is going to be bad.

If you want, though, the Indians' Charlie Jamieson might have been worse in terms of overall failure to excel.

A mediocre left fielder, his OPS+ fell between 83 and 112 for 12 seasons. He was like 3 Travis Lees stacked end to end.

15 seasons as a regular is a very tough standard. While Bluege, McLemore, and Jim Gantner all have at least 1750 games played over very long careers, I wouldn't say that any of them have 15 years as a regular. They all had seasons when they didn't play all that much. I think you're going to have to take a step up in quality from these players to get the 15 years.

Okay, maybe Ferguson and Bluege don't quite make 15 seasons.

If we need 15 seasons of over 100 games, I'm having trouble finding worse than Tommy Corcoran. He had a career OPS+ of 74, with a high of 92 (in the 1891 AA). Oddly, four of his "Most Similar" batters are Hall of Famers.

MattB:

The people on Corcoran's "most similar" lists are all top-notch defensive middle infielders, mostly shortstops. You have to have tremendous defensive value to last that long as a weak hitter ... don't you? Ah, but there's another name from Corcoran's similar list: Larry Bowa! Yes, he's got 15 seasons as a regular, 2247 games total. His career OPS+ is 71, lower than Corcoran's. (His high was 95, and he had to bat .305 to do that.) As for fielding, Prospectus has Bowa at 391 FRAR, -3 FRAA, as compared to Corcoran's 462 FRAR, 54 FRAA. WARP3: Bowa 47.1, Corcoran 48.6. Of course, you said it: "no one with a 15 year career as a regular is going to be bad."

Have we got our man?

15 years or more as regular and eligible for HOM:

Elected: Start, Sutton, White, Hines, Kelly, O'Rourke, Glasscock, Ward,

Basically the same as 1917, with 2 guys removed. None of the new guys make my top 30. There were a few 1 place shifts here and there. Will be discussed in greater depth on the real ballot. I think it goes 11 deep as far as guys that I will really like to see in the HoM. Not the most all inspiring group. I think the next 5 years (1919-1923) are going to be fun. There are a number of guys that it will be interesting where they slot in. Sure there are a few top of the ballot guys, but Sheckard, Wallace, Walsh, Tinker, Brown, Evers, Plank should be interesting. Plus, there are a few more guys that will get some votes.

1. Keeler

No mention? Burdock, Brown, Quinn, W.Robinson, Corcoran, Tenney

I'm not going to say that any of these don't have 15 years as regulars, but in several cases, you're pushing pretty hard at the edges of what you would mean by regular - seasons of barely over half the team's games, partial seasons where no one else on that team played any more at that position - that sort of thing. That just goes to reinforce your point.

How about Charlie Grimm?

Yes, I saw Bowa, but couldn't pull the trigger.

The problem was those 5 All-Star Game appearances and 2 gold gloves.

That seemed to much like "peak" to me to be considered the worst. He may still be the best choice, though.

>Posted 2:41 p.m., January 14, 2004 (#70) - Howie Menckel

I'm open to other definitions of a regular, but it has to be something that I can fairly easily quantify for use in a series of database queries. My definition allows each position to be self-defining. I determine the max fielding games played at each position in each league season, and a regular is someone who plays more that half of that. Utility regulars play more than half the league leader in G. Pitchers can also qualify with more than half the league leader in IP or GS.

I also have two "levels" of regular. The other is based on more than 3/4's of the maximum.

15 years or more as "full-time" regular and eligible for HOM:

Elected: Start, Sutton, White, Hines, O'Rourke, Ward,

<i>Little support: Gleason (gone Force, D.Farrell, McGuire)

If we need 15 seasons of over 100 games, I'm having trouble finding worse than Tommy Corcoran.

Looks like we have a winner. Special mention should go to catcher Malachi Kittridge, who played more than 50 games in his first 15 seasons, then 27 more in his 16th and final season. He had a carrer OPS+ of 56 and a high of 89.

Van Haltren payed a good CF. Kelley played a very good LF. They both tinkered with IF a bit.

Not sure why one is much higher than the other. I suspect it might be that George has no one thing that comes out and screams "PICK ME!", like Mazeroski's glove or Brock's speed or Ryan's strikeouts, but he sure seems to be a fine package.

I'd agree that Van Haltren is a fine package, and Andrew's argument that we may look back in 30 years and decide to vote him in is quite well-reasoned, but surely the obvious distinction between Kelley and Van Haltren is that Kelley has a great peak and VH doesn't?

A quick scan of my adjusted win share numbers shows Kelley at #3 among eligible position players in five-year consecutive win shares with 173. He's behind Jennings (196), and Pike (179), and he leads Flick (169) and Duffy (167) by a narrow margin. Van Haltren stands at (rough estimate) #22 at 148. And since he was a durable player, he ranks even farther down on a measure of peak rate. Kelley places at #4 on peak rate, behind Jennnings (again), McGraw, and Chance among eligibles. Van Haltren ranks no higher than #25 on this measure (I didn't make as exhaustive a survey here).

These two factors surely contribute the gap between the rankings of these two players with similar career values. The electorate has shown little support for short-career players with high peaks, but it has also tended to show only tepid support for long-career players without high peaks. Van Haltren has a higher peak than, say, Beckley, or Cross, but his peak is not outstanding by any stretch of the imagination.

>To that end, I am giving renewed consideration to Pearce,

A minor note: Joe McGinnity was voted for in 13 different positions on the ballot last year. Only 1st and third positions saw no Joe votes. There is a good chance this year (particularily given his 6 fourth place votes last year) that Joe will place on all 15 spots on the ballot (as well as missing the ballot). I know no strategic voting is allowed, but that would be interesting. I'm going to have him second.

I just discovered that I will be away all next week, so could someone do me a favour and post my ballot for me. It is on this thread at post #3.

Thanks.

I looked at every source I could find re. pre-NA ball early on here at HoM. Without getting out my notes, I thought the comparison of Pearce and Wright was more or less as follows:

1. Pearce is indeed more commonly mentioned as a player than Wright is. The Wright mentions more often than not were re. his leadership, management, etc.

2. The mentions of Pearce more often than not were in re. to his defining the SS role and his fielding prowess. The mentions of Wright as a player, though few, were mostly about his "scientific" hitting. They say he was a "great scientific" hitter rather than a "great hitter." Who today really knows what that means? But if I had to guess, I'd guess that Wright was a better hitter than Pearce, though Pearce had more defensive value.

3. Both played as early as the mid-'50s and as late as 1877, though in Wright's case he only played 3 games after 1874. Since Wright was born in '35 and Pearce in '36, Wright's productive career was essentially 20 years from age 20 to 39, Pearce's was more like 22 seasons from age 20 to 41.

4. Both of course are prominently associated with two of the best teams in America in that era, though it is also true that Wright's association with the Red Stockings is more as its leader and manager, though he was also its CF.

5. Pearce played 291 games in the NA and NL at OPS+ ~80, still almost exclusively at SS to age 41. (This is about the same as Davey Force's career OPS+ from age 22-37.) Wright played 180 games in the NA and NL (though just two in the latter) at OPS+ ~90, still in CF through age 39 (this is about the same as Dave Eggler's OPS+ for a whole career from age 20-34.)

6. The fact that neither was a star--or at least an offensive force--after age 35 can hardly be seen as a demerit. Who was in those days? Well, Joe Start. Cap Anson's seniority was 20 years off into the future. The fact that both still played regularly and in important defensive positions seems to be to their credit.

Put it all together and I see both of them as, along with Start, the very best players who started play before the Civil War, other than the tragic Jim Creighton. The best, in other words, of their generation. That counts with me. I'm not a big timeliner. But between the two, I rate Pearce more highly, but only a little more highly, because of the more frequent mentions of him as a player (they don't prove or even suggest he was better, they just add to the level of certainty) and the fact that he would clearly have had more defensive value. So Pearce has worked his way up toward the top of my ballot (he's in the no-brainer category--i.e. #s 1-4 area) while Harry is in the middle range (the probably deserving category somewhere from #5-10).

Could you plop them down in 1890 or 1910 and they would excel? Don't know. Don't care. What's that got to do with anything?

That would have to be a joint Anglo-Canadian venture...

David, OK, now you made me get out my notes.

First, Wright played for St. George's Dragonslayers (cricket) from 1850-59, with his father.

He began playing baseball in 1857 with the Knickerbockers. One report said he was the "best exponent of batting as a science." Another said he was the "finest, safest, best and least showy player in America." You decide what all that means.

In 1863 he first received money to play. He was presumably among the first half dozen or so to do so. Creighton is often said to be the first in 1860-61, but there are various such claims.

He remained with the Knickerbockers through 1863, the Knicks being a gentlemen's club, they were frankly not one of the most competitive. By this time the workingmen's clubs (Atlantics, Eckfords) were taking over.

He joined the Gothams in 1864, however, the Gothams being a more competitive outfit. He was said to have sat out the '65 season, however, and in '66 went to Cincy not for baseball but for cricket. In the fall of 1866, however, the entire cricket team jumped over to the Cincinnati Base Ball Club and the rest is history.

Asa Brainard was the Cincy pitcher, one of the "fireballer" pitchers of the day. It is possible that he was a more valuable member of the Red Stockings, I don't know, but no way would he rate ahead of Harry on my ballot. I'm not aware of Brainard prior to 1866-67 and as early as 1871 he was pretty much washed up (4 year ERA+ of 51 in the NA). So he had at best 4 effective years. Nobody has ever said he was as good as Spalding, and had less than half the career. Harry, meanwhile, played 19 years to Brainard's 8 or so. 100 years later I would say that Willie Mays:Jack Sanford::Harry Wright:Asa Brainard.

Brainard probably has an excellent case as a peak guy, but I'd rather go with the 1860s guys who were tops at their positions for many years, plus had long careers (Pearce, Start, and Wright). As much as I feel that decade needs to be represented just as much as all of the later ones, the lack of documentation forces me to be wary of the Reachs, Chapmans and Brainards (who may be deserving of the honor of induction).

If I were going to support a pitching candidate who was a star in the 1860s, I'd back George Zettlein, I think.

Re Andrew Siegel's post at #91: what a great post! Much to think about!

It's a shutout for Griffith over Joss! How can anyone possibly say Joss was better??

Top 10 finishes in ERA+ among all major league pitchers, assuming that I'm using the correct league factors:

Griffith: 1,8,9,10,10 (within a seven year span, around 13th the other 2 years)

re: Beckley & mediocrity . . .

Re-posting something I figured & posted about a month ago, here's where Beckley's OPS+s ranked among all starting first basemen every year he was a starter, w/ his OPS+ in paranthesis:

1888..5/16 (157)

Kelly Leak was a poor baserunner.

Well, it's been said 100 times. Now, 101.

They played against the best competition available anywhere on earth. They had no choice to play against any other. I'm not gonna penalize them for stuff they couldn't control.

And Chris, why not also adjust for Jake Beckley's competition--meaning the other 1Bs. If he had played in the '80s (i.e. compare him to the ABC boys) nobody would be voting for him. Different competition. Or, if you want to compare Beckley to all the other eligible players on the ballot, and not just the 1Bs, or even just all the eligible players from the '90s, how does he compare with Flick, Kelley, Keeler, Thompson, Pike, Duffy, VanHalen, Ryan, et al. Not very well.

When you set up your comparisons in the "right" way--e.g. Harry Wright vs. Billy Hamilton, Paul Hines, George Gore, Jimmy Ryan, Hugh Duffy, G. V. Haltren and Jake Beckley vs. Kitty Bransfield, Fred Tenney, Dan McGann, Jack Doyle and Jake Ganzel--and you base your ratings on those kinds of ratios, you can make your ballot come out any way you want.

PS. John, there's a great article on Al Reach in the National Pastime No. 23 (don't you wish they put a date on it!). His career is fairly well documented. Among the 6 guys you name, he is a lot more Start, Pearce, Wright than he is Chapman or Brainard.

He moved into elite ball in 1861 with the Eckfords, and is another of the several rumored to have been the first paid player. Certainly he was one of the first. In 1864, when the Philadelphia Athletics decided they wanted to compete at an elite level, they went up to New York and bought Al Reach. He was said to be the first 2B to play off the base, "midway between the bases."

I've never voted for him, but after Start, Pearce and Wright, he was probably the #4 player (depending on how you regard Jim Creighton, I would say he was #4 in career value) to have started play before the end of the Civil War, at least among anybody I'm aware of.

Well, whoops. All that thought about designing how I wanted to figure out my HoM ballot, and I forgot to properly rank one of the players. And the worst thing about it is, it's the most-debated player we have who I didn't do properly... Bob Caruthers. Bob Caruthers is probably the best player in baseball in the mid-to-late 1880s. And I completely forgot to take that into account.

Revised ballot will jump Caruthers to #2 from #9 on my ballot, moving everyone else down a spot.

The SABR Bioproject entry for Harry Wright is the best on-line bio I've seen for him. Check the link above.

They played against the best competition available anywhere on earth.

On the one hand this is true. I don't find this as a sign that the players merit enshrinement, however. I see it as an indictment of the overall quality of play. I don't want to vote for a player because his competition was lackluster, I want to vote for a player because he had a great career. If you ignore the level of competition, you may as well vote for Kelly Leak.

They had no choice to play against any other. I'm not gonna penalize them for stuff they couldn't control.

If I don't think they were great, I'm going to penalize them for not being great.

Chris, I don't think you deliberately cooked up a method that was meant to be unfair to particular players. What I did say was that if your method is based on how a player compares to a peer group, then the composition of the peer group is important.

You compared Beckley to 16 contemporaneous 1Bs (though now you've expanded that to include contemporaneous OFers and that's good, though I note that he only ranks 6th out of 9 among that glut).

But you use a different method--you compare Wright and Pearce's league (not actually Pearce and Wright themselves, either, BTW) to later leagues, a much larger and more demanding peer group.

If you use the same method for Pearce and Wright as for Beckley--i.e. compare them to their contemporaries at similar positions--Pearce and Wright look pretty good. That's all I'm saying.

I don't mean to pick on Chris, but when all is said and done the timeline really bugs me, and here's why.

Right now in 2004 (in the real world and in Bill James' world), the timeline works against pretty much any player from before the 1920s and even some who came later.

But right now in 1918 (in the HoM world), it works in favor of Jake Beckley and against Dickey Pearce. Back in 1898 (in the HoM world), it worked in favor of Deacon White and against Dickey Pearce.

The only players in the HoM project against whom the timeline will ever work are the players who played in the earlier part of the period that was eligible here in 1898. 1845-1893 were eligible then, so those who were "discriminated" (i.e. timelined) against were those who played about 1845-1875. No other players will ever be discriminated against by the timeline--well, that's not quite true. Jake Beckley might be discriminated against some day, but *he had his day in court.* He had his shot.

The players from 1845-1875 never had *their day.*

<i>On the one hand this is true. I don't find this as a sign that the players merit enshrinement, however. I see it as an indictment of the overall quality of play. I don't want to vote for a player because his competition was lackluster, I want to vote for a player because he had a great career. If you ignore the level of competition, you may as well vote for Kelly Leak.,/i>

If someone had evidence that the top-line stars of the 1860s were inferior to the top-line stars of the 1890s, then that would be different. The level of competition during Wagner and Cobb's era was certainly inferior to our time, but that doesn't mean that the Flying Dutchman and the Georgia Peach wouldn't be MVP material today. It does mean that there were far fewer players of quality than today, which made it easier for the truly great players to dominate (and which is another reason why I hesitate including more players on my ballot from the 1860s).

Marc:

I have the National Pastime that you were referring to (what a great publication, BTW!) Reach was, without doubt in my mind, a very, very good player. I lean toward the notion that he was a great player, but I think his contemporaries were more impressed with Pearce, Start and Wright. I'd rather stay on the side of caution with him. Besides, he has a great argument as a Pioneer induction later on.

The players from 1845-1875 never had *their day.*

Good point, Marc.

BTW, I just received an e-mail from SABR that Deadball Stars is on its way to all 2003 members. Woo-hoo!

Chris, I don't think you deliberately cooked up a method that was meant to be unfair to particular players.

Mixing insomnia & message boards isn't always a good idea for me.

Don't forget undigested potatoes. :-)

Marc,

You're argument, which is a good one [well done!], is not, however, and argument against timelining. It is an argument against starting the voting in 1898, instead of 1871 or so. Your argument suffers from not being made 1.5 years ago, when it could have been persuasive enough to change the start date.

There is another argument though. Accepting the pennant is a pennant argument (at least within the confines of professional, major leagues) there werent any pennants prior to the NA. If the players of the 1860's had wanted to get elected to some hypothetical 21st century Hall of Merit, they should have formed a professional major league earlier than they did.

BTW, as an OT aside, I apply the same argument to "local" tv revenue sharing (local referring to superstations in this case). If Bud Selig wants a chunk of Atlanta's TBS money, he should buy a Milwaukee indy channel in the early 70's and turn it into a national cable tv superstation.

I am good at finding solutions to major problems, but people dont seem to like solutions that require a time machine. Tying this all back in, Im not going to stop timelining because we screwed up and started voting the wrong year.

There is another argument though. Accepting the pennant is a pennant argument (at least within the confines of professional, major leagues) there werent any pennants prior to the NA.

Didn't they have championships?

John,

Yes they did, but they werent professional, major leagues. Both amateur (sp?) and minor leagues have championships today, but I aint giving them any credit.