Baseball for the Thinking Fan

Login | Register | Feedback

btf_logo
You are here > Home > Hall of Merit > Discussion
Hall of Merit
— A Look at Baseball's All-Time Best

Tuesday, January 13, 2004

1918 Ballot Discussion

Sorry it’s late guys . . . the 1918 entry class is just a tad below the 1917 class . . .

1918:
NIXEY CALLAHAN
MIKE DONLIN
MATTY MCINTYRE
PAT MORAN
JAKE STAHL
BILLY SULLIVAN
BUSTER BROWN
CY MORGAN
BARNEY PELTY
JACK POWELL

Joey Numbaz (Scruff) Posted: January 13, 2004 at 02:05 PM | 178 comment(s) Login to Bookmark
  Related News:

Reader Comments and Retorts

Go to end of page

Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.

Page 2 of 2 pages  < 1 2
   101. John (You Can Call Me Grandma) Murphy Posted: January 16, 2004 at 09:49 PM (#520819)
Both amateur (sp?) and minor leagues have championships today, but I aint giving them any credit.

I wouldn't either, because we know that they are not the cream-of the-crop for today's game (unlike the situation during the 1860s).
   102. RobC Posted: January 16, 2004 at 10:07 PM (#520820)
From the constitution:

All major league players are eligible for the Hall of Merit.

The is the clause, amongst others, that bans my previously planned 1980's vote for Oh. It also, by my reading, bans votes for players (other than negro leaguers specifically excepted elsewhere in the constitution) who didnt, at the very least, play a little in the NA.

Anyone who looks back will note that I did give significant credit to Joe Start and others for pre-NA play. But, the argument above was based on the pennant is a pennant argument. If I accept that (and I do about 90% or so), it only applies to professional and major league pennants, not pennants for amateur play in the NY metropolitan area. Even if they were the best teams, they weren't both professional and major. The sentence you snipped provides a solution (not surpisingly, requiring a time machine) but they chose not to organize the NA until the 70s.
   103. John (You Can Call Me Grandma) Murphy Posted: January 16, 2004 at 10:32 PM (#520821)
Also eligible are all ?excluded? players, most notably Negro Leaguers, and pre-MLB players.

Here's the next sentence from the constitution. :-)

BTW, I wish Oh (or possibly Mexican League players) was eligible, too.
   104. Marc Posted: January 16, 2004 at 10:49 PM (#520822)
Your point is well taken. Starting at an earlier date would solve the problem. But in the absence of timelining there would be no problem to solve. All players would be judged equally based on their value toward their team's success in whatever year it happened to be.

BTW, I think the only player of any note (of the very early days) at all who did NOT play in the NA was Jim Creighton. The other notables from the '60s--H. Wright, Pearce, Chapman, Reach, Spalding, G. Wright, Brainard, et al--all survived to play at least a few years of NA ball.

OTOH, the distinction between the NA and the previous championships is to me a shade of gray rather than a black and white difference, just as the evolution from NA to NL, the addition of the AL, etc., all represented gray scale evolutions. IOW, I would hard pressed not to consider the Atlantics and the Cincy Red Stockings not to be "major" teams, regardless of whether they were Major teams.

But again, as a practical matter, Creighton is the only name I could come up with who even fails a black and white test.
   105. RobC Posted: January 17, 2004 at 07:05 AM (#520823)
John,

I some how missed that next sentence, at least the pre-MLB part. However, the way it is written seems to allow Japanese players to fit in as "excluded". It doesnt make negro leaguers and pre-MLBers the only "excluded" players. Maybe there is another line in the constitution that excludes Japanese players. If there is, it should probably be amended.

We both seem to have problems with next sentences tody.

Marc,

The absense of timelining is not an option, because "timelining", or "league leveling" (as it really has nothing to do with a timeline - only quality of league play) is necessary otherwise the UA, the FL, the Sally league, the ACC, and the Highview Babe Ruth League (which, when I was in it, was dominated by a future major leaguer - fortunately he was on my team) all are equal. I know that you reject most of those, but while I accept the "a pennant is a pennant" argument to a certain point, I think we have to draw a line. You do too, you dont count all those as equal to the NL and AL - for me its adjust for the value of the league. This works out nice and fair for all except a few early players, and I gave a few bonus points to the those guys to help make sure the best of them got in. And they did. Now, there is no problem in adjusting for WW2 or an imbalance between the AL and NL, or "dilution due to expansion" or Martians entering baseball (2047). Your method screws up those adjustments forever in order to correct a small problem for a small group of players, who were as good as they were because no one outside the NY metro area was playing the game. Give those guys a boost if you want and then lets move on.
   106. Brad G. Posted: January 17, 2004 at 03:09 PM (#520824)
RobC says: <the Highview Babe Ruth League (which, when I was in it, was dominated by a future major leaguer - fortunately he was on my team)>

I give up... who?
   107. RobC Posted: January 17, 2004 at 04:34 PM (#520825)
Huh?
   108. Marc Posted: January 17, 2004 at 05:05 PM (#520826)
RobC, to me a "timeline" and a "league adjustment" are different things:

A league adjustment recognizes differences between two or more leagues operating at the same time--NL and AA, principally, but the red herring of modern major vs. minor league play is covered in the same way. Players at any given time who are not playing the best competition available *at that time* are discounted.

The timeline makes similar adjustments but to leagues and players from different times, generally on the *assumption* that the quality of play has has continuously improved by "then" to "now." "Pool" (i.e. size of pool) arguments and standard deviations, etc., are used to "prove" this assumption. Now, I think the assumption is correct, though there is legitimate debate about how much improvement has occurred (how steep a timeline is appropriate).

To me the problem with the timeline is that it makes the debate one about "tools," not "value." Could Harry Wright have been as successful in 1920-1960 as he was in the 1870s? Of course not. Of course, if Ty Cobb or Willie Mays had been born in 1836 and had to invent basic baseball skills without any mentoring rather than simply master them with the help of experienced coaches, etc., well, they might not have been as successful as Harry Wright.

But even that is not a meaningful comparison, because it is about tools, not value. But to one of the most famous and successful baseball teams in history, the Cincinnati Red Stockings, Wright had immense value. The timeline takes that away from him on the basis of a lack of skills. It asks the wrong question.

But anyway, back to my main point, comparing players and teams and leagues at the same time and from different times are two different things, and I think it helps to have two different names for those operations.
   109. Howie Menckel Posted: January 17, 2004 at 06:22 PM (#520829)
FYI, this from DanG's 1917 ballot was quite notable, I thought. Something to think about as you evaluate the top of the backlog....

10) Keeler (9,ne,ne)? Similar career length and peak to the OF glut, seems to fit right in with them. For those who put a lot of importance on pennant race impact, consider the following: Keeler and Kelley were teammates on five pennant winners (1894-95-96-99-1900). They were exactly the same age. Keeler never had more win shares than Kelley in any of those five years. Here?s their WS, with Keeler?s first in each pair: 23/30, 23/27, 25/31, 29/30, 22/22. A difference of 3.6 WS per year. Kelley also topped another teammate, Jennings, in three of those five pennant years. Keeler tops Kelley in longevity, that?s about it, hanging on for about two extra bulk years. Also, despite his shorter career, Kelley also compiled more career defensive win shares than Keeler, 51.7 to 48.5.

Criticisms invited, but this sums up better than I have why Kelley is better than Keeler (though I would vote both into the HOM).
   110. Howie Menckel Posted: January 17, 2004 at 06:29 PM (#520830)
token Matty McIntyre mention: MM was one of the best players in baseball in 1908, when his team won the AL pennant.

Too bad he played in an OF with Ty Cobb and Sam Crawford!!
   111. Chris Cobb Posted: January 17, 2004 at 08:22 PM (#520831)
10) Keeler (9,ne,ne)? Similar career length and peak to the OF glut, seems to fit right in with them. For those who put a lot of importance on pennant race impact, consider the following: Keeler and Kelley were teammates on five pennant winners (1894-95-96-99-1900). They were exactly the same age. Keeler never had more win shares than Kelley in any of those five years. Here?s their WS, with Keeler?s first in each pair: 23/30, 23/27, 25/31, 29/30, 22/22. A difference of 3.6 WS per year. Kelley also topped another teammate, Jennings, in three of those five pennant years. Keeler tops Kelley in longevity, that?s about it, hanging on for about two extra bulk years. Also, despite his shorter career, Kelley also compiled more career defensive win shares than Keeler, 51.7 to 48.5.

Criticisms invited, but this sums up better than I have why Kelley is better than Keeler (though I would vote both into the HOM).


I agree that this is a mostly fair view of Keeler and Kelley, and I agree that Kelley should rank ahead of Keeler. I would qualify it a bit, however, by looking at the latter half of their careers more closely.

Keeler's career value is higher than Kelley's not just because Keeler hung on longer, but because after his peak he was season-by-season better than Kelley.

Kelley's WS, season-adjusted, after his eight great years:

21, 17, 17, 17, 11, 8, 8

Keeler's WS, season adjusted, after his first eight years as regular.
   112. Marc Posted: January 17, 2004 at 09:54 PM (#520832)
>Kelley has a very high peak and a rather lackluster decline

Does this also describe Hugh Duffy?
   113. Chris Cobb Posted: January 17, 2004 at 11:00 PM (#520833)
>Kelley has a very high peak and a rather lackluster decline

Does this also describe Hugh Duffy?


Yes, except that his peak wasn't quite as high and his decline was abrupt rather than lackluster. Kelley had four averge and slightly below average seasons and three part-timer seasons after his eight great years. Duffy's line after his great eight year run looks like this in season-adjusted win shares:

27, 18, 6, 9, 3, 1

He has one very good season, one mediocre season, and after that he was a part-time to bit-part player. It's the weakness of his later career that puts Duffy a notch below both Kelley and Keeler in my view.
   114. Dag Nabbit is part of the zombie horde Posted: January 17, 2004 at 11:42 PM (#520834)
<i>Posted 11:20 a.m., January 13, 2004 (#11) - JoeDimino (e-mail)
   115. Howie Menckel Posted: January 18, 2004 at 12:52 AM (#520835)
This might be good on an 'off-year vote.'

Pecking order of leading current eligibles, going backwards from the 1917 election. Lists where player ranked that year among those not elected, Stovey takes up most of the high 'missing' spots of recent years.
   116. Howie Menckel Posted: January 18, 2004 at 01:02 AM (#520836)
Career votes-points leaders
   117. Paul Wendt Posted: January 18, 2004 at 01:44 AM (#520837)
RobC #135 on "league leveling"
   118. Chris Cobb Posted: January 18, 2004 at 05:24 AM (#520838)
1918 Prelim Ballot

Rethinking, rethinking. I?ve taken advantage of the lull in new arrivals to do a little qualitative thinking about what makes a HoMer and about how eras should be compared (props again to Andrew Siegel for giving me the impetus to do this.) On the prelim ballot here, I?ve ranked all the players whom I believe I could be convinced belong in the Hall. They?re divided into three groups.

Ought to be elected: These are players whose induction I am ready to argue for.
   119. Brad G. Posted: January 18, 2004 at 04:23 PM (#520839)
RobC- Who was the future major leaguer you played with in B.R. league?
   120. RobC Posted: January 18, 2004 at 05:11 PM (#520840)
Brad,

http://www.baseball-reference.com/d/dixonst01.shtml

Probably wont be receiving any HoM votes.
   121. Howie Menckel Posted: January 18, 2004 at 05:31 PM (#520841)
Wow, no offense, Rob, but that's one of the worst careers in major league history!
   122. OCF Posted: January 18, 2004 at 05:53 PM (#520842)
Worse even than Allan Travers? Of course, Travers IS eligible for us to vote for now. I trust that most of you know the story.
   123. RobC Posted: January 18, 2004 at 06:22 PM (#520843)
Its better than my baseball career. By a lot. I was about that bad at the Babe Ruth level.
   124. Marc Posted: January 18, 2004 at 08:58 PM (#520844)
Well, Dixon's record is even worse than Travers. Not only that, but the management that saw fit to pitch Dixon was a lot worse, too. There was a compelling reason for Travers to pitch that day. Dixon...?
   125. RobC Posted: January 18, 2004 at 09:34 PM (#520845)
He was a 3 time (I think) AAA all-star. He just didnt do it at the ML level.
   126. John (You Can Call Me Grandma) Murphy Posted: January 18, 2004 at 09:40 PM (#520846)
He was a 3 time (I think) AAA all-star. He just didnt do it at the ML level.

Since he only had 5 innings to prove himself, I don't think we can conclude necessarily that he didn't have the goods to stay in the majors.

He didn't really help his cause, though. P.U.!
   127. Chris Cobb Posted: January 19, 2004 at 02:27 AM (#520847)
Since serious discussion is a bit slow over this weekend, I'll venture to post this bit:

Just noticed that Rob Neyer's Friday column over on ESPN (which includes a bunch of info from MGL) mentions in passing questions in which considering quality of competition _within_ a season may be important to finding the right answer. That list includes the following:

"Were some old-time pitchers even better than their stats, because they pitched a disproportionate number of games against the better teams?"

Which leads me to wonder if Rob Neyer has been looking in on the HoM threads, and if the forthcoming Neyer/James _Book of Pitchers_ will contain information that will be useful for us, supplementing Chris J.'s excellent work.
   128. KJOK Posted: January 19, 2004 at 06:52 AM (#520848)
Chris J. wrote
   129. KJOK Posted: January 19, 2004 at 08:00 AM (#520849)
Not trying to steal Chris J.'s thunder, but I think his idea about Support Neutral W/L being very important to look at is excellent, so I've posted the file for 1876-1912 at the Hall of Merit Yahoo egroup:

http://groups.yahoo.com/group/HallofMerit/files/Support Neutral W_L 1876_1/
   130. KJOK Posted: January 19, 2004 at 08:03 AM (#520850)
for some reason the link didn't past correctly - trying again:

http://groups.yahoo.com/group/HallofMerit/files/Support Neutral W_L 1876_1912.xls
   131. Chris Cobb Posted: January 19, 2004 at 05:34 PM (#520853)
Dickey Pearce is kind of on the fringe, I don't think his contributions as a player after 1870 are really 'significant', but an argument can be made that a player at his age playing like he did was a signifcant contribution. I guess you could compare Wright to Pearce also, but like I said, these guys are extremely borderline as to whether or not they were really intended to be covered by this project.

Just wanted to throw that out there . . .

I still think it would have been a good idea to set up a Centennial Commission or something to honor a couple of players like H. Wright and Pearce, but that didn't go over so well.


I have been, and am, a supporter of the Centennial Commission idea, but since it looks like we will not have one, I think we should give full weight to the pre-1871 performance of any player who played in the NA. I've learned enough about the early game from this project to feel that a Hall of Merit that doesn't honor at least Dickey Pearce from among the stars of the 1860s will be somewhat diminished by that omission.
   132. Dag Nabbit is part of the zombie horde Posted: January 19, 2004 at 05:39 PM (#520854)
Joe - my first e-mail crapped out. Don't know what went wrong but it came back undeliverable. So I tried it a 2nd time & I guess it worked because I haven't gotten a failed send message back yet. Let me know if anything went wrong.
   133. EricC Posted: January 19, 2004 at 05:52 PM (#520855)
[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.


Bob Caruthers had a career ERA+ of 123 and OPS+ of 135.
   134. Chris Cobb Posted: January 19, 2004 at 07:26 PM (#520857)
Good job, Eric, of showing the small size of the relative contribution of Caruthers' hitting to his value as a pitcher. I think that argument ought to bring Caruthers down out of the top tier of available candidates.

To assess the additonal value of Caruthers' outfield play accurately, however, you need to look at the 366 games in the context of Caruthers' seasonal play. A season's worth of Noodles Hahn plus a half season's worth of Jimmy Ryan in the same season is a lot more valuable than a season of Noodles Hahn.

Caruthers' case will ultimately rest on the fact that he was a great two-way player, not on the fact that he was a great-hitting pitcher.
   135. OCF Posted: January 19, 2004 at 07:33 PM (#520858)
Two all-star teams:

C: Bennett
   136. jimd Posted: January 19, 2004 at 07:42 PM (#520859)
<i>Caruthers: 123 + (135-78)/9 = 129.
   137. MattB Posted: January 19, 2004 at 08:32 PM (#520860)
EricC wrote:

<i>Caruthers: 123 + (135-78)/9 = 129.
   138. EricC Posted: January 19, 2004 at 08:49 PM (#520861)
And if you look at it long enough, it almost looks like math!

Insulting my competence in math or insinuating that I would EVER post a specious argument to this group is the final straw. I resign.
   139. MattB Posted: January 19, 2004 at 09:15 PM (#520862)
Wow.

My goal is never to win by driving my opponents away! I just want to convince them of the rightness of my argument.

Adding OPS+ and ERA+ is mathematically inaccurate, but perhaps is acceptable as a back-of-the-envelope comparison.

Nonetheless, I have to say that adding them is such a way that hitting is worth approximately 5% as much as pitching (irrespective of PAs or IPs) is simply specious. It would be intellectually improper to fail to point that out.
   140. John (You Can Call Me Grandma) Murphy Posted: January 19, 2004 at 09:17 PM (#520863)
Hold on, Eric. I don't think Matt was intentionally insulting you (though he could have handled his post a little better). Caruthers happens to be the guy he's pushing (I guess mine would be Dickey Pearce), so anything critical towards the Parisian One stings a little bit more than for some of the other candidates on his ballot.

I've had my dander up a few times myself, but I don't think anyone was trying to diss me on purpose. Passions sometimes have the best of us here.

I for one think our group would be diminished if you moved on.
   141. MattB Posted: January 19, 2004 at 09:26 PM (#520864)
I agree with John. I don't want to be the cause of anyone's departure. I apologize for the snarkiness of my post.
   142. Howie Menckel Posted: January 19, 2004 at 09:30 PM (#520865)
Some of us argue in lawyer-like fashion, others don't like that approach. I have no complaints, and hope others don't get too wound up over a post here or post there..

on another topic, surely Jack Powell deserves at least this bit of recognition for winning 245 games. from baseballlibrary.com....

"Powell, a contemporary of Cy Young, was built along the same lines, and threw with a free and easy sidearm delivery. He was the workhorse of every staff on which he pitched, but in 16 seasons was never on a pennant winner. He was 24-15 in his second season, for Cleveland (NL), and was part of the wholesale transfer of players from Cleveland to St. Louis in 1899. Powell went 23-21 for the Cardinals that year. He was 22-17 with the Browns in 1902, and lost an AL-high 19 in 1911. He spent a couple of seasons playing in New York, but continued living in St. Louis to run a saloon with his brother-in-law and former batterymate, Jack O'Connor."
   143. Chris Cobb Posted: January 19, 2004 at 09:55 PM (#520867)
Eric,

This reader who attempts to do his math accurately agrees with jimd's criticism of your analysis as weighting pitching too heavily by leaving out defense.

This reader disagrees, however, with MattB's criticism of your analysis. Math is not the issue, but the extent to which the mathematics employed accurately models playing value. Your model is _much_ more accurate than the one that Matt uses to show the badness of yours.

Matt makes a huge modeling mistake, if I understand jimd's correction to your original model correctly, by mapping OPS+ onto innings pitched. His ratio shows, more or less, what Caruthers' value would have been like relative to Hahn's if he had batted for one third of his team's plate appearances! I don't think Matt is seriously advancing his model, but is attempting a reductio ad absurdum critique of your model. It doesn't really work, though, because it's so much more absurd.

Let me try to corroborate your point with some work that I have been doing, that I haven't gotten into concise enough form to post in full.

Matt is right that adding OPS+ to ERA+ is mathematically inaccurate, strictly speaking. To be much more accurate, one could undertake a similar calculation using Caruthers' runs created above average. If one did that, (which I have more or less done, in fact), one would find that the basic point that you made, and that jimd's modifications to your comparison basically confirmed, is correct. Given that the pitcher is only 1/9 of the offense when he plays and is 50% to 67% of the defense, the value of his contributions as a hitter, even if he is a very good hitter, are quite small in comparison to his contributions as a pitcher. They're not so small as to be unnoticeable, but they are small enough that they can't make up for any signficant difference in pitching quality.

Here's a bit of data. In 1886, Caruthers created 74 runs, according to the basic RC formula used on baseball-reference. Prorate that to his estimated at-bats while pitching, you get 39 runs. An average-hitting pitcher using Caruthers' outs would have created about 9 runs (calculated by adding up RC and outs for all pitchers that season throwing at least 27 innings, subtracting out play at other positions). So Caruthers added 30 runs over an average pitcher as a hitter than year. If you prorate team runs scored and runs allowed to Caruthers' IP and then subtract the 30 runs above average Caruthers created, the pythagorean method shows that Caruthers' teams win about 1.3 more games than they would have with the run-support contribution of an average pitcher. Since that's an addition in wins wholly attributable to Caruthers, I figure that his hitting as a pitcher earned him about 4 win shares more than an average-htting pitcher of the same pitching ability would have earned. Four win shares isn't shabby. But it isn't world-shaking, either.

And that, I take it, is your basic point, which is not specious at all!
   144. Marc Posted: January 19, 2004 at 10:10 PM (#520868)
> [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.

Let the record show that "how well a pitcher hits doesn't really matter" was not my comment.

But thank god no one has questioned my competence in math, because I would be absolutely powerless to defend myself on that one.
   145. Marc Posted: January 19, 2004 at 10:38 PM (#520869)
IOW, I didn't start this food fight. I hope Eric will hang in here.

I do like the fact that Matt pointed out that pitchers are part-time players. It has been noted that Caruthers or other good-hitting pitchers only hit when they're pitching (not true for Caruthers of course). And it's been pointed out that most catchers were part-timers way back when. It is sometimes forgotten, however, that pitchers only pitch when they're pitching, if you receive my meaning.

Also re . Caruthers, note that his hitting created about 4 WS per year. This is roughly the same as 4 pitching wins. If Caruthers and Hahn are roughly comparable as pitchers (Hahn +8 on ERA+ and Caruthers +800 IP) then 4 hitting WS per year are significant. And that is before equalizing for shorter seasons (if you do).
   146. Marc Posted: January 19, 2004 at 11:00 PM (#520870)
And finally, this is why I have pretty much given up using rate stats. Sure, I look at them. They help fill out a player's profile. But in rating and ranking a player, they shed more heat than light.

IOW, what does Hahn's higher ERA+ really mean? Well, when he pitched he was a little more effective. A little. And how much did he pitch, compared to Caruthers or whomever? And what offensive and defensive contributions did he make? And how does all of that inform the ERA+?

Well, I just can't see starting with a measure like ERA+ and then trying to "inform" it. I can see starting with WS and WARP, both of which give Caruthers a 2-to-1 edge. Caruthers pitched in 100 more games and played in 3X as many games as Hahn. With WS and WARP, I don't have to "eyeball" the ERA+ vs. IP calculations, they're built in. Then I can look at ERA+ in order to add some coloration to the matter.

But relying on or starting an evaluation with ERA+ or OPS+ (or advocating for a player using ERA+ or OPS+) is rarely going to go anywhere. That at least is MHO after 20 years of HoM balloting.

This certainly is not to say that both WS and WARP are problem free. They're not. That's why I use both, in hopes that they balance out or, when they agree, I can take that as a providing some certainty.
   147. jimd Posted: January 20, 2004 at 02:06 AM (#520871)
And finally, this is why I have pretty much given up using rate stats.

Agreed. The rate stat gives you only one side of the value "rectangle". The other side is the quantity played. And for pitchers in particular, this varies a lot between those who are considered as "regulars".

Measured against "average", a pitcher with a 140 ERA+ who is 1/4 of the staff has the same value as a pitcher with a 120 ERA+ who is 1/2 of the staff. Measured against "replacement level", the latter pitcher is more valuable because of the very real difficulty in finding "average" pitchers to fill out the staff.
   148. Paul Wendt Posted: January 20, 2004 at 04:03 AM (#520872)
Measured against "average", a pitcher with a 140 ERA+ who is 1/4 of the staff has the same value as a pitcher with a 120 ERA+ who is 1/2 of the staff. Measured against "replacement level", the latter pitcher is more valuable because of the very real difficulty in finding "average" pitchers to fill out the staff.

Not quite. The difference between 140 and 120 is fewer runs than the difference between 120 and 100. ERA+ 140 corresponds to yielding runs at 71.4% of average rate; ERA+ 120 to 83.3% of average rate.
   149. Chris Cobb Posted: January 20, 2004 at 04:20 AM (#520873)
Also re . Caruthers, note that his hitting created about 4 WS per year. This is roughly the same as 4 pitching wins. If Caruthers and Hahn are roughly comparable as pitchers (Hahn +8 on ERA+ and Caruthers +800 IP) then 4 hitting WS per year are significant. And that is before equalizing for shorter seasons (if you do).

Not so. In my calculations, the pitcher's record actually changes by 1.3 wins if (under the conditions Caruthers was playing) he creates 30 runs above an average pitcher by hitting like an all-star outfielder. Since, in looking at the effect of the pitcher's contribution as hitter, the contribution of all other players on the team holds constant, the pitcher's hitting thus receives full credit for the 1.3 win increase. That equals 4 win shares. It would therefore be inaccurate to turn around and say that 4 win shares for a pitcher equals four wins in his record.

James hoped to create a system where pitchers' win shares would correspond roughly to their number of wins, but this is an eyeball sort of correspondence at best. The ratio between pitcher wins and win shares will vary a great deal depending upon the ratio between pitcher quality and team quality.
   150. Marc Posted: January 20, 2004 at 04:43 AM (#520874)
Chris,

>this is an eyeball sort of correspondence at best. The ratio between pitcher wins and win shares will vary a great deal

I meant my observation in this way. Caruthers' additional 4 WS per year in hitting value is roughly in an eyeball sort of way equal to 4 pitching wins. There was no "math" involved. I only meant elaborate on the model, not crank out a particular value.
   151. jimd Posted: January 20, 2004 at 05:00 AM (#520875)
Measured against "average", a pitcher with a 140 ERA+ who is 1/4 of the staff has the same value as a pitcher with a 120 ERA+ who is 1/2 of the staff.

ERA+ 140 corresponds to yielding runs at 71.4% of average rate; ERA+ 120 to 83.3% of average rate.

Based on what Paul pointed out above, the pitcher with 120 ERA+ who is 1/2 of the staff actually saves more runs relative to "average" than the 140 ERA+ pitcher. He is actually comparable to a 150 ERA+ due to the non-linearity of ERA+. (Another reason I don't like working with ERA+.)
   152. Chris Cobb Posted: January 20, 2004 at 07:02 PM (#520877)
Does Griffith's W-L record look better than his expected record? If the answer is yes, I would raise his ranking a bit.

The WARP system sees him as being 9 wins better than his expected record for his career, one of the few pitchers among the top pitchers now under consideration who substantially exceeds his expected record. I think jimd has a post about this on the pitcher questions thread somewhere.

Since we don't have game data for Griffith, it's hard to know if WARP's calculations are 100% accurate, but they do match the anecdotal evidence.
   153. Marc Posted: January 20, 2004 at 09:09 PM (#520878)
Is it really relevant to ask "why" a pitcher allowed runs? Or, maybe I should say, does this open a can of worms? I always understood that all pitchers at least through the deadball era and perhaps even later, relaxed with a lead. Specific superstitions don't seem to be any different than the generalized belief that it is OK to relax with a large lead.
   154. Howie Menckel Posted: January 20, 2004 at 09:19 PM (#520879)
I've given Waddell a little break in my rankings for poor run support, but those who are slaves to WARPs and ERA+s alone are missing the boat on Griffith.
   155. Daryn Posted: January 20, 2004 at 10:28 PM (#520882)
howie,

griffith made it on to my ballot for the first time this year thanks to you.
   156. Chris Cobb Posted: January 20, 2004 at 11:18 PM (#520883)
Here's another view of the quality of Griffith's performance. I've been working on evaluating pitchers by comparing their record to the likely record of an average pitcher receiving the same offensive _and_ defensive support. I assume that an average pitcher throws an average number of innings, so in seasons in which the pitcher in question throws more than an average number of innings, I compare his wins from his innings about average to the wins of a replacement level pitcher with the same support for those innings (the wp of a replacement level pitcher is that of the average pitcher divided by 1.2.) This seems a better way to assign value to durability than either a) comparing the pitcher to an average pitcher working the same number of innings, which underrates durability or b) comparing the pitcher only to an average pitcher working an average number of innings, which overrates durability. This approach acknowledges that durability is a virtue because it means teams have to give fewer innings to replacement level pitchers, but it also acknowledges than any well-managed team will have somebody around replacement level to throw those innings, so the pitcher is only improving his team as much as he is exceeding replacement level.

Anyway.

Here are the % wins above average value, listed in order from highest to lowest, for all seasons in which Joe McGinnity, Clark Griffith, Rube Waddell, and Vic Willis were full-time starters.

McGinnity -- Griffith -- Waddell -- Willis
   157. Howie Menckel Posted: January 20, 2004 at 11:44 PM (#520884)
Joe,
   158. User unknown in local recipient table (Craig B) Posted: January 21, 2004 at 04:37 PM (#520886)
Griffith won a lot more games than almost all of his contemporaries

Among pitchers for the years 1886-1911 (roughly Griffith's era - that's five years before and five years after his career) he ranks ninth in wins (and 16th in winning percentage) in that era.
   159. Howie Menckel Posted: January 21, 2004 at 05:18 PM (#520887)
Craig, who does he trail? The quality of your point (which might be considerable) would depend on the answer.
   160. jimd Posted: January 21, 2004 at 06:23 PM (#520888)
Total Wins 1886-1911 (career wins, if different)

Young 511, Nichols 361, Mathewson 289 (373), Clarkson 264 (328),
   161. Howie Menckel Posted: January 21, 2004 at 06:53 PM (#520889)
Thanks, Jim.
   162. User unknown in local recipient table (Craig B) Posted: January 21, 2004 at 08:00 PM (#520890)
I wasn't making a "point". I was just providing a more accurate assessment of your statement, which I thought wasn't very accurate. Griffith's win totals, by themselves, are certainly typical of a good HoM candidate.
   163. EricC Posted: January 22, 2004 at 12:57 AM (#520891)
MattB-

Sorry that I overreacted to one hasty post of yours. I will continue to participate in this discussion, and in this and future elections.
   164. MattB Posted: January 22, 2004 at 03:06 AM (#520892)
Great!

Now if I could just convince you to put Bob Caruthers on your ballot . . .
   165. Chris Cobb Posted: January 22, 2004 at 03:37 AM (#520893)
Eric,

Glad you're back!
   166. Chris Cobb Posted: January 22, 2004 at 05:10 PM (#520897)
Chris, can you add Addie Joss to your list? I think he has a much better case than Griffith or McGinnity.

He's the next in line for completion. I should be able to add him to the chart in the next day or two.
   167. ronw Posted: January 22, 2004 at 05:27 PM (#520898)
No baby yet. Since I'm in a looking to the future mood, I just realized, if Kelley doesn't get in this year, next year should be an excellent 1890's high-offense vs. 1900's deadball in the LF battle of Joe Kelley vs. Jimmy Sheckard. Their WARP and WS numbers are very comparable, but generally obtained in vastly different environments. Sheckard even displaced Kelley as Brooklyn's LF.

On a side note, that 1900 Brooklyn-Baltimore lineup was amazing, even with Jennings and Kelley on the downswing of their careers. Nearly every starter (and even a reserve or two) is a serious HOM candidate. Generally the superteams (Brooklyn-Baltimore, Pittsburgh-Louisville, and St. Louis-Cleveland) look great, but the Brooklyn-Baltimore combo squad looks to be the best.
   168. Chris Cobb Posted: January 22, 2004 at 09:19 PM (#520899)
In answer to Joe's request, here's data on percentage above average value for post-1893 pitching candidates with Joss added as well. See post 195 above for where these numbers come from and exactly what they measure.

McGinnity -- Griffith -- Waddell -- Willis -- Joss
   169. karlmagnus Posted: January 23, 2004 at 12:36 AM (#520901)
What about Tommy McCarthy?
   170. Paul Wendt Posted: January 23, 2004 at 02:17 AM (#520904)
Hall of Fame members from 19th Century Base Ball
   171. Howie Menckel Posted: January 23, 2004 at 05:16 AM (#520906)
No problem, Craig B. My original statement was a little careless.
   172. User unknown in local recipient table (Craig B) Posted: January 23, 2004 at 02:53 PM (#520907)
Not a problem, Howie. Actually, I need to take what you are saying to heart and re-evaluate Griffith. My system is very tough on Griffith types (very good peak, very good career length, not unquestionably great on either measure) and on pitchers generally - I need to be scrupulously fair to those type of guys.
   173. DanG Posted: January 25, 2004 at 07:23 PM (#520909)
If you check the election threads, you'll find Griffin getting good support (~#16) his first five years on the ballot 1904-08. Then the strong 1909 class of newbies included Ryan and Van Haltren, who were exact contemporaries with longer careers. Griffin's support has never recovered, lost in the OF glut.

Reportedly, Griffin was sold to St. Louis after the 1898 season and chose to retire rather than report, ending his career prematurely. I think if he'd hung on and added another four years of bulk play, like Ryan did, our image of him would be much different.
   174. Marc Posted: January 25, 2004 at 10:37 PM (#520910)
I doubt seriously that there are any "pretty good" players who haven't been noticed. One of these days I want to compile a list of all the players who have ever received a vote here. I would guess there's easily 100 or more by now...? Or maybe somebody already has a consolidated vote list???
   175. Howie Menckel Posted: January 26, 2004 at 01:57 AM (#520911)
Marc,
   176. Marc Posted: January 26, 2004 at 03:09 AM (#520912)
Thanks Howie. Looks to me like there are 75 players who have received votes without getting elected--i.e eligible this year. That would be 71 listed on Howie's post plus (just eyeballing here) Thomas, White, H. Davis and Chance.

What would also be interesting would be to determine how many votes *to elect* each has gotten--i.e. how many times in the #1 or #2 slot, depending on how many we elected that year. Howie? (Better you than me for that particular study!)
   177. jimd Posted: January 27, 2004 at 09:01 PM (#520913)
Number of votes to elect, all elections prior to but not including 1918. (Some counts will change and Rube Waddell will be added to the list. I hope I didn't miss any votes.)

41 Charlie Bennett
   178. Marc Posted: January 27, 2004 at 10:04 PM (#520914)
Thanks, jimd, interesting. Obviously most of the votes to elect go to players who get elected. Among the also-rans, many of whom will someday be elected of course, but among the present-day also-rans, votes to elect seem almost to be randomly distributed, not so much tied to ballot position. In other words, the #3 guy gets there not with #1 and #2 votes but with #3 votes. Maybe. Just a theory.

More to the point, I'm not sure Thompson and Caruthers and Bennett's support means that they will get elected. My theory is that to get elected, a player needs to appear on 35-40 ballots or more. A few 1st and 2nd place votes don't cut it.
Page 2 of 2 pages  < 1 2

You must be Registered and Logged In to post comments.

 

 

<< Back to main

BBTF Partner

Support BBTF

donate

Thanks to
aleskel
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Syndicate

Page rendered in 0.7064 seconds
49 querie(s) executed