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

Monday, May 24, 2004

1927 Ballot Discussion

The top new candidates (thanks DanG and Chris Cobb!):

WS W3 Rookie Name-Pos (Died)
287 73.1 1907 Ed Konetchy-1b (1947)
227 52.6 1908 Dode Paskert-CF (1959)
205 50.7 1910 Hippo Vaughn-P (1966)
189 48.4 1908 Slim Sallee-P (1950)
180 42.2 1910 Duffy Lewis-LF (1979)
161 44.9 1911 Ray Caldwell-P (1967)
155 38.0 1909 Dots Miller-1b/2b (1923)
143 37.9 1911 Lefty Tyler-P (1953)
106 28.8 1914 Braggo Roth-RF (1936)

Negro Leaguers
HF%—from expert voting in _Cool Papas and Double Duties_
BJ—Bill James positonal ranking in _NBJHBA_
MVP—Sum of Bill James’ best player and best pitcher awards and John Holway’s MVP and best pitcher awards
All-Star—number of times player designated as seasonal all-star by John Holway

HF% Career  Name-pos (born)          BJ    MVP All-Star
68% 1901-25 Pete Hill-CF/LF (1880)  #4 lf   2     5*
08% 1909-21 Frank Wickware-P (1888)         2     1*
00% 1914-21 Horace Jenkins-OF       (??)    1     2*

Players Passing Away in 1926

HoMers
Age Elected
75 1914 Cal McVey-C/1B
50 1924 Eddie Plank-P

Candidates
Age Eligible
66 1901 Bill Hutchison-P
64 1899 George Pinkney-3B
63 1900 Danny Richardson-2B
60 1903 Lou Bierbauer-2B

Joey Numbaz (Scruff) Posted: May 24, 2004 at 10:12 AM | 387 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 3 of 4 pages  < 1 2 3 4 > 
   201. Michael Bass Posted: May 28, 2004 at 01:20 AM (#648454)
I'm with Marc on this one. As a career guy, catchers get hammered even more than normal by the stats I generally consider. As such, they get a bump up my ballot depending on the relative amount they played at catcher.
   202. Jeff M Posted: May 28, 2004 at 01:32 AM (#648470)
If they are so defensively valuable, why don't they accumulate 15+ fielding WS/FRAR?

Imperfect systems? Fielding is notoriously hard to evaluate, and particularly so for catchers. Two quotes from the Win Shares book regarding evaluation of catchers defense:

"At some positions, like this one, we are starved for data, and the best we can hope for is what might be considered a well-organized fishing expedition."

and

"If you want to argue that [sacrifice hits allowed] is an oblique and speculative measure of a catcher's defensive quickness, you are of course correct. We're doing the best we can with very imperfect data. I would have a hard time arguing that fielding bunts is 10% of a catcher's defensive responsibility - but on the other hand, I would have a hard time arguing that throwing out baserunners is 50% of a catcher's defensive responsibility, or that 'not making errors' is 30% of a catcher's defensive responsibility, or that 'avoiding passed balls' is 10% of a catcher's defensive responsibility. This is what we've got to work with. If you've got a better idea, speak up."
   203. David Concepcion de la Desviacion Estandar (Dan R) Posted: May 28, 2004 at 02:01 AM (#648514)
I guess my question is, do we have any data on how much catcher defense affects the game at all? I can't imagine it is really comparable to 2B/SS/CF.

WARP, at least, measures wins above a replacement at a given position...if Cochrane was so damn good, why did he only have two years where he was nine wins better than a replacement catcher? I'm much more impressed by Bench (45.9 WARP3 his best 4 seasons), and by Berra's longevity.

Is it possible that there isn't as much difference between catchers as there is among other position players? I'd think that the few that could really hit--Piazza, for example--would be racking up insane WARP3 in the 13-14 range, if a replacement catcher is really so bad.
   204. Michael Bass Posted: May 28, 2004 at 02:05 AM (#648521)
Isn't at least part of it that catcher play less games in a season than other position players? Catchers play, what, 10% fewer games generally than other full time players? And they obviously have much shorter careers (productive careers at least) on the whole. Less opportunity to acquire WARP or Win Shares or whatever non-rate metric you want to use.
   205. David Concepcion de la Desviacion Estandar (Dan R) Posted: May 28, 2004 at 02:46 AM (#648563)
shouldn't that be held against them, though? A guy provides less value over replacement in 130 games than 160.
Is it maybe that replacement value is just not being set low enough for catchers? For example, over Bresnahan's career, BP credits him with 434 batting runs above a generic replacement player and just 444 above a replacement catcher. Berra actually has fewer career BRARP than BRAR. I would expect a gap of at least 100 runs over a career, possibly more.
   206. Chris Cobb Posted: May 28, 2004 at 03:04 AM (#648572)
It's obvious, isn't it, that catcher defense affects the game tremendously, much more than any other fielding position. A catcher has _many_ more chances than any other position player does. The fact that most are not obviously difficult, and that it is absolutely essential that a catcher be skilled enough to handle those chances routinely, obscures the way in which catcher defense affects the game.

Put another way, defensive replacement level for catchers is higher than for any other position, so the _relative_ impact of catcher defense is low. But every catcher is more important defensively than any other position player.
   207. Jeff M Posted: May 28, 2004 at 03:07 AM (#648576)
shouldn't that be held against them, though? A guy provides less value over replacement in 130 games than 160.

But can catchers really play 160?

Is it maybe that replacement value is just not being set low enough for catchers?

I took a look at Piazza in this regard too, and he has only 30 more BRARP than BRAR -- that's only about 2 a year. That seems impossible, since Piazza is probably the best hitting catcher in history.

Bench has BRARP that is 10 less than BRAR. Doesn't that indicate that in Bench's era, replacement catchers hit better than replacement position players? Is that possible? I don't think of the catching position as having a huge pool of talent.
   208. David Concepcion de la Desviacion Estandar (Dan R) Posted: May 28, 2004 at 03:19 AM (#648581)
But...but...if they all can do it so well--including triple-A guys (replacement players)--then why does it matter?

Let me put it this way, assuming that, as you say Chris, C defense affects the game the most but defensive replacement level is higher. I'm going to put forth two real players and four imaginary players: Ozzie Smith, Charles Johnson, the best AAA SS, the best AAA C, a high school SS, and a high school C. Let's say that I have a defensively average team with an average pitching staff, allowing 700 runs. Now let's substitute my league-average shortstop for the three SS candidates. With Ozzie, I allow, say, 670 runs, with the triple-A SS, 750 runs, with the high school SS, 800 runs. Ozzie is 30 FRAA, 80 FRARP, and 130 FRAHS (Fielding Runs Above High School). Now, let's put back the league average shortstop, and take out our league average catcher. With Charles Johnson, I allow 685 runs, with the triple-A catcher, 730 runs, with the high school catcher, 900 runs. CJ is 15 FRAA, 45 FRARP, and 215 FRAHS.

Who's the more valuable defensive player? Ozzie. Why? Because the high school players don't matter. MLB games are won and lost based on how much better a team's players are than the available replacement level--not the absolute level of difficulty or impact. The fact that an absolute dolt could screw up a game more at one position than another doesn't matter if there is a surplus of players who will do fine. The *only* thing that matters at the major league level, as far as I'm concerned, is value above replacement.
   209. David Concepcion de la Desviacion Estandar (Dan R) Posted: May 28, 2004 at 04:22 AM (#648608)
Hm, well, the numbers seem to check out here. The 8 starting NL catchers in 1908 had a combined EqA of .247. Figuring back from Bresnahan's BRARP I get that they are using a catcher-league average EqA of .250. Similarly, Yogi in 1954 has BRARP equal to his BRAR, and sure enough, the league average EqA for catchers that year was .256 (leaguewide average is .260 by definition). So at least in these two randomly selected league-seasons, catchers just weren't as bad hitters as we thought.
However, there's a catch-22 here, if you'll excuse the pun--these are eight-team leagues, and these star catchers are responsible for one eighth of the league average for catchers. For example, if you take Bresnahan out of 1908 catchers, the league average for the position falls 10 points, to .237.
There are a bunch of ways to approach this problem. If you include AL catchers in the average (cutting Bresnahan's contribution to the league average in half), then the catcher average EqA is .241, which increases Bresnahan's BRARP from 58 to 61 (a small change). But perhaps a better way to go would be, rather than arbitrarily setting replacement at .230, look at the actual player pool. Using (.23/.26)*league position average EqA, you get a replacement catcher EqA of .213. But there were three catchers who had EqA's below .213, including the stunning Bill Bergen, who racked up a .188 (and that was a career high!). If you take out the two extreme outliers (Bergen at the low end, Bresnahan at the high one), catchers' EqA's in 1908 differed from each other by an average of .007. So assuming that Bergen was below a replacement-hitting catcher, let's take the second-lowest EqA out there (.206) and deduct .007 twice (one to fill the last starting catcher spot in place of Bergen, the next representing replacement level). That comes out to a .192 EqA for a replacement-hitting catcher in 1908, which is 28 EqR in 351 outs. Now Bresnahan is 70 BRARP (instead of BP's 58 BRARP and 53 BRAR), which is a lot more impressive, elevating him from 8.7 WARP1 to 9.9 (although going from WARP1 to WARP3 would still cut him down to 7.5 WARP3).
What do you all think of this approach? Does anyone have a spreadsheet with sufficient data (like, every player-season ever) to calculate BRARP this way?
   210. Chris Cobb Posted: May 28, 2004 at 04:32 AM (#648610)
The *only* thing that matters at the major league level, as far as I'm concerned, is value above replacement.

The only thing that matters to what? I'm going to venture away from baseball and into semantics for a minute, because it seems to me that in discussing catchers you keep switching the question that you are asking.

You asked at the outset if it might not be the case that catcher is a historically weak position.

What does a historically weak position mean? Does it mean that catchers typically accumulate less value than other players? or does it mean that catchers are generally less meritorious than other players?

If it means the first thing, then the answer is obviously yes. Individual catchers do accumulate less value, because the physical demands of the position limit playing time in any given season, undermine a catcher's hitting, and shorten his career. But does that mean tha catchers are less meritorious, when the fact is that if you took any other player and placed upon him the demands that are placed upon a catcher, he would play fewer games, hit less well, and have a shorter career?

Then you asked how much catcher defense affects the game. I made a case showing how it affects the game. It then turns out that what you were interested in was not the general importance of catcher defense, but the defensive value of particular major-league catchers as measured above replacement level.

Value above replacement is surely important, but it should not be an absoute guide to merit, because it is a relative value, not an absolute one. By ignoring the importance of catcher defense in general and judging the value of catchers only for their value above replacement, you are in effect penalizing catchers in your evaluation, for being, as a group, better defensive players than players at any other position.

If you want, from the outset, to say that value above replacement is the only thing that matters at the major-league level, then there is nothing to discuss. WARP measures value above replacement, and it shows clearly that catchers will have less of it than players at any other position.

But does that make "catcher" a weak position or make individual catchers less meritorious? Bringing the discussion back to value above replacement gets you nowhere with the question of merit, because it provides insufficient evidence to answer the question that you originally posed. Catchers have less value as individuals because they have taken on the uniqe physical demands of the position, in order to help their teams. Don't they deserve to be evaluated on a standard that takes account of this?
   211. John (You Can Call Me Grandma) Murphy Posted: May 28, 2004 at 04:37 AM (#648612)
I guess what I want to know is, is it possible that catcher is just a historically weak position?

If you are saying the position of catcher is as historically weak as the other eight positions, than I agree with you. IOW, there is no such thing as a historically weak position. They all have the same value to the team.

shouldn't that be held against them, though? A guy provides less value over replacement in 130 games than 160.

Why should someone be penalized for taking on a position which is unquestionably tougher on the body than any other position (we'll leave pitching out of the discussion for now)? How many games would Hank Aaron or Pete Rose have played if they had donned the "tools of ignorance" instead? A helluva lot less, that's for sure.
   212. David Concepcion de la Desviacion Estandar (Dan R) Posted: May 28, 2004 at 04:39 AM (#648613)
Hm...let me give you a reductio ad absurdum case. Let's say that 50% of football is quarterback play (it's obviously not, but just for the sake of argument), with the remaining 50% divided up among the 21 other positions. Now let's say that every QB in history is exactly equal to every other in every way.
We're electing players to the football HoF. I feel like my HoF would have no QB's, and yours would have all QB's. Is this right?

Now returning to the realm of the reasonable, if catchers impact the game so much, wouldn't marginal differences in ability lead to large differences in runs/wins?
   213. Jeff M Posted: May 28, 2004 at 04:40 AM (#648615)
What do you all think of this approach? Does anyone have a spreadsheet with sufficient data (like, every player-season ever) to calculate BRARP this way?

I like the approach, although as I've said many times before, I tend to ignore the Warp1 to Warp2 adjustment and I just adjust Warp1 for season-length. I then apply my own discount factors for the "lesser" leagues like the AA and FL. Otherwise I feel like I'm timelining.

As for the data, the Lahman database has all the stats from which to calculate EqA, but it isn't divided by position, so it would be some work to filter the data for just catchers. But it could be done.
   214. David Concepcion de la Desviacion Estandar (Dan R) Posted: May 28, 2004 at 04:40 AM (#648617)
I just don't see why that the SD of WS or WARP for catchers, on a per-game or per-season basis, would be lower than for other positions. But is it?
   215. David Concepcion de la Desviacion Estandar (Dan R) Posted: May 28, 2004 at 04:47 AM (#648622)
Jeff M--I think you need some adjustment to WARP1, although I do think BP's is a little severe. Were Nap Lajoie's 6 best years (107.8 WARP1 after adjusting for season length) really 30% greater than Barry Bonds's 6 best (82.9)?
   216. Jeff M Posted: May 28, 2004 at 05:18 AM (#648632)
Dan, in addition to the size of the BP adjustment, I have a conceptual problem with the adjustment, at least as far as the HoM project goes.

I personally don't favor timelining. Lajoie's 6 best years, relative to his contemporaries, may be better than Bonds' 6 best years relative to his contemporaries, even though Bonds is unquestionably a better athlete and player. So I'm not willing to discount Lajoie's performance 20% simply because he played the game at a different time.

I only discount when the player played in a clearly inferior league relative to the other available leagues. So the FL, the UA and the first few and last few years of the AA get a discount. I also discount Negro League MLEs for their speculative nature.

Just my approach. We've had some discussions here in the past between the members of the timelining camp and the members of the non-timelining camp, and there's no consensus (I don't think).
   217. DavidFoss Posted: May 28, 2004 at 06:02 AM (#648641)
Here is the pre-NA data I have for Lip Pike:

===========================

LIP PIKE

1866 -- Philadephia Athletic (23-2)
-- Pos -- 3B-OF-2B (16 games)
-- Competition -- Philly/NJ/NYC+/Wash/WilkesBarre
-- Hitting -- 3rd on team with 6.25 R/G (McBride 6.4, Hayhurst 6.31 (13G), Dockney 5.85)
1867-1 -- Irvington NJ (16-7)
-- Pos -- 3B (6 games)
-- Competition -- NYC+/NJ/PA/Bal/Wash/Philly
-- Hitting -- 19 Runs in 6 games (3.17 R/G) (Wolters 4.05, 3.42, Campbell 3.31)
1867-2 -- New York Mutual (23-6-1)
-- Pos -- OF-3B-2B-1B (21 games)
-- Competition -- NYC+/NJ/Philly/Wash/Bal
-- Hitting -- 2nd on team with 3.90 R/G (Waterman 3.93, CHunt 3.88, Hatfield 3.70)
1868 -- New York Mutual (31-10)
-- Pos -- OF (25 games)
-- Competition -- NYC+/Alb/Wash/Philly/Cincy/
-- Hitting
---- 7th on team with 2.4 R/G (Devyr 3.02, Swandell 2.98, McMahon 2.97)
---- 4th on team with 3.28 H/G (McMahon 3.58, Swandell 3.46, Devyr 3.36)
---- 3rd on team with 4.36 TB/G (McMahon 4.51, Swandell 4.46, Flanly 4.26)
1869 -- 2B
-- Team Record -- (40-6-2) (Pro: 15-6-1)
-- Competition -- NYC+/NJ/Bal/Wash/Cincy/Alb/Philly
-- Hitting (all games):
---- Runs -- 3rd on team with 4.02 R/G (Start 4.41, CChapman 4.10, Pike 4.02)
---- Hits -- 5th on team with 3.65 H/G (Start 4.41, CChapman 4.10, CSmith 4.00)
---- TB -- 2nd on team with 6.77 TB/G (Start 7.41, CChapman 6.52, Ferguson 5.83)
1870 -- 2B
-- Team Record -- (41-17) (Pro: 20-16)
-- Competition -- NYC+/Rockford/Philly/Bal/Wash/Cincy/Cle/Chi
-- Hitting (pro only):
---- Hits -- 4th on team with 2.33 H/G (Start 2.75, Pearce 2.50, JChapman 2.39)
---- TB -- 1st on team with 4.25 TB/G (Start 4.14, JChapman 3.31, Pearce 2.94)
   218. DavidFoss Posted: May 28, 2004 at 06:03 AM (#648642)
Here is the pre-NA data I have for LEVI MEYERLE:

========================================

LEVI MEYERLE

1867 -- Philadelphia Geary (19-6)
-- Pos -- ??
-- Competition -- Philly/Greater Philly
-- Hitting -- 1st on team with 4.4 R/G (Halbach 4.2, Hamberger 4.07, Hopkins 3.89)
1868 -- Philadephia Geary (7-6)
-- Pos -- P
-- Competition -- Philly/Wash/Camden
-- Hitting -- 1st on team with 4.1 R/G (Opdycke 3.89, Halbach 3.81, Heubel 3.56)
1869 -- Philadelphia Athletic (45-8) (Pro:15-7)
-- Pos -- OF-C-P (34 games)
-- Competition -- Philly/NJ/Cincy/NYC+/Bos/Bal/Wash/PA/Alb)
-- Hitting (all games)
---- 9th of 10 on team with 3.97 R/G (Reach 5.39, Cuthbert 5.13, McBride/Sensenderfer 5.00)
---- 9th of 10 on team with 3.76 R/G (McBride 5.35, Reach 5.26, Sensenderfer 4.56)
---- 9th of 10 on team with 6.03 TB/G (Reach 9.11, McBride 8.18, Cuthbert 7.70)
1870 -- Chicago (65-8) (Pro: 22-7)
-- Pos -- 3B-P (3B only in the pro games)
-- Competition -- Chi/StL/NO/Mem/Mil/Cle/Rock/Syr/Bos/Bal/Wash/NYC+/Phi/Ind/Cin/Lou
---- Hitting (Pro games only!) -- Documented the At bats!
---- 4th on team with 116 AB (McAtee 139, Wood 139, Cuthber 135)
---- 8th on team with .362 BA (Flynn .481, Wood .417, Cuthbert .415)
---- 3rd on team with .612 SLG (Treacy .687, Pinkham .641, Cuthbert .607)
   219. DavidFoss Posted: May 28, 2004 at 06:16 AM (#648647)
OK... fact dumping mode off... commenatary on...

Looks like Pike really did have better pre-NA numbers. He broke in with one of the top teams (Phi-Athletic) and then bounced around a bit but stayed with the top teams (Irvington excepted). Looked like he had an off year in '68.

Meyerle played very well for Geary, but my opinion is that Geary was a not a top club. He was a big fish in a small pond. Jumped up to Athletic in '69 and struggled. Improved a bit in 1870... but I wouldn't have predicted his big NA- bustout.
   220. David Concepcion de la Desviacion Estandar (Dan R) Posted: May 28, 2004 at 06:30 AM (#648650)
Yeah. What do you all make of Meyerle? I haven't heard anyone talk about him, but his NA hitting seems comparable to that of Pike/McVey and just a bit behind Barnes.
   221. DavidFoss Posted: May 28, 2004 at 07:29 AM (#648659)
Meyerle has never gotten much support.

McVey has 2543 MLB PA's, Pike has 2032 MLB PA's and Meyerle has 1453 MLB PA's. The data I showed above doesn't give Meyerle as much pre-NA numbers as Pike. Perhaps comparable to McVey there.

Meyerle does have a 164-155-152 advantage in OPS+ over Pike and McVey respectively... but the career is short.

McVey got inducted 10-15 year ago... Pike's been clawing his way up the ballot, but whether he is ever inducted in quite uncertain. A McVey/Pike/Meyerle ranking seems reasonable put Meyerle's support is zilch. I think Meyerle just fell off the radar and there was never any desire to resurrect his candidacy as the ballot grew thinner.
   222. Chris Cobb Posted: May 28, 2004 at 01:54 PM (#648731)
Jim Sp, thanks for your analysis of OPS+! Very helpful for me.


DavidFoss, thanks for the additional pre-NA data. As to resurrecting Meyerle's candidacy -- he just didn't play enough. There was _a lot_ of debate about McVey's career length and total games played as he neared election as he neared election, and Meyerle's is significantly shorter. By my estimate, he has about 250 season-adjusted career win shares, which is not enough to get him close to my ballot. Maybe he should be in my top 40, though. Have to think about that. As for an early candidate who has truly fallen off the radar who might well be better than Meyerle, there's John Clapp to consider.


Dan Rosenheck -- Although I'm advocating for using a different standard for assessing catchers, I think your statistical work, e.g. examining catcher EQA and Bresnahan's 1908 season, is admirable. It's likely that there's some missing catcher value to be found.

As to the reductio ad absurdum argument, my HoM would not have all quarterbacks, because I'm willing to assign different standards of merit to different positions that have different demands -- I don't see them all as directly comparable. I try to build those different standards into a single system by means of positional adjustments so that I don't have to rely on strict positional quotas, but those adjustments are meant to ensure that the top players at each position all end up with values similar enough to ensure that my top 15 always contains a mix of infielders, outfielders, and pitchers.
   223. Carl Goetz Posted: May 28, 2004 at 02:20 PM (#648761)
Does anyone have their WS book in front of them? I'm leaving town for the weekend straight from work tonight and I forgot to bring my WS book with me. If anyone has them, please send year-by-year WS for Joe Jackson, Eddie Cicotte, Ed Konetchy, Dode Paskert, and Hippo Vaughn to carlgoetz5@yahoo.com. I appreciate any help any of you can give me. Thans.
   224. karlmagnus Posted: May 28, 2004 at 02:23 PM (#648763)
David, once again thank you very much. Meyerle down a bit, but only a little. Pike up a bit. I now think all 4 of Meyerle, Pike, Wright and Pearce are very close to each other, with McVey and G. Wright significantly ahead. Wright and Pearce should get in as Pioneers, but Pike and Meyerle have a much better ML record.

It probably is Pearce/Pike/Meyerle/H.Wright, but none of them are top 5 and there shouldn't be more than 10 ballot spots between top and bottom.
   225. stephen Posted: May 28, 2004 at 03:08 PM (#648849)
I was looking over the credentials of Konetchy in relation to Beckley. and I was wondering if someone who supports Beckley could give me anargument of why he desrves to be on the ballot instead of Konetchy. The only real advantage I see using Win Shares is the career number, which was accumulated over more seasons. Konetchy has a slightly better peak and rate per 162 games.

Really, I'm unimpressed by both of these 1st base candidates, but I'm getting the feeling Beckley has much more support than Konethcy and I can't really figure out why. they seem about even, and I might give Konetchy a slight edge because of his slightly better peak value.

Thanks for the help.
   226. DavidFoss Posted: May 28, 2004 at 03:30 PM (#648882)
It probably is Pearce/Pike/Meyerle/H.Wright, but none of them are top 5 and there shouldn't be more than 10 ballot spots between top and bottom.


Yeah, my ranking is Pike/Pearce/Meyerle/HWright but your mileage may vary (and indeed it does).

As for the ballots in between, its hard quantify that. These are thin ballots if the in-out line is murky, the out/outer/outest lines are murkier. I have them 2nd/7th/gone/gone at the moment.
   227. karlmagnus Posted: May 28, 2004 at 03:35 PM (#648887)
Beckley had 2930 hits, playing slightly short seasons. That's a very elite number, and is why he's in the HOF. Konetchy at 2150 is not an implausible candidate -- I have him level with Sheckard and Magee, the latter of whom has just been elected (which I wouldn't have done.) But if we fail to elect Beckley eventually, we'll have been led seriously astray by misguided sabrmetrics, IMHO.

If 300 wins is less impressive for a pre-1900 candidate, as people argue against Welch, then almost 3,000 hits must be correspondngly more impressive. I believe that if Beckley had got the extra 70 hits he'd have been only the second player to do so, after Cap Anson.
   228. DavidFoss Posted: May 28, 2004 at 03:43 PM (#648902)
Another fun source for pre-NA fans out there is "Early Innings: A Documentary History of Baseball, 1825-1908". It doesn't have statistical numbers, records, standings like Marshall Wright's NABBP book does, but it has a bunch of interesting articles from the time period. Its a collection of the primary source material interspersed with short comments by the editor(s).

-- Its got a box score from the 1858 All-Star Game that John Murphy keeps mentioning. HWright batted 9th and played "middle field' (cf?). Batting third and playing SS for Brooklyn was "Pidgeon" and seeing at how many other Atlantics are on the roster, I'm thinking that might be Dickey Pearce, but I'll never know for sure.

-- Its got a newpaper article advertising a private game organized by Pearce and Creighton where the proceeds go to them showing evidence of how players could get money for playing early on.
-- Its got Creighton's obituary in the paper which warns of the dangers of too much exercise!
-- Its got minutes from the 1866 NABBP meeting where the NABBP unaminously voted to prohibit games against teams that contained "one or more colored persons"
-- The box score from the 1867 game where Barnes/Spaulding's Rockford team upset GWright's Washington Nationals on their famous road trip.
-- Several articles on rules and the roles that each position played in the field.

Anyhow, there are some out there who have expressed interest in information like this so I thought I would pass it along.
   229. karlmagnus Posted: May 28, 2004 at 03:48 PM (#648910)
David, I fear we differ. I would argue that 2nd/7th/gone/gone is too much differentiation between players very close in quality. The ballot is only thin at the top; lower down each ballot is by definition fatter than the last as more new players become eligible than we elect. If Pearce is 7th but you prefer Pike, it should probably be no broader a spread than say 7/6/13/16. I had them 24/22/15/9 (well done, precisely the wrong order!) but now have them 13/14/15/16.
   230. David Concepcion de la Desviacion Estandar (Dan R) Posted: May 28, 2004 at 03:48 PM (#648911)
Was it really harder to amass bunches of hits back then, or just HR's? Seems to me that all the great deadball players did it--Collins, Speaker, Cobb (duh), Wagner, Nap. Crawford, Keeler, and Beckley were 2900+.
   231. karlmagnus Posted: May 28, 2004 at 04:01 PM (#648931)
Dan makes my point. Beckley moves in a pretty glossy crowd. We'd all elect the other 7, so why not Jake?
   232. DavidFoss Posted: May 28, 2004 at 04:07 PM (#648948)
David, I fear we differ. I would argue that 2nd/7th/gone/gone is too much differentiation between players very close in quality. The ballot is only thin at the top; lower down each ballot is by definition fatter than the last as more new players become eligible than we elect. If Pearce is 7th but you prefer Pike, it should probably be no broader a spread than say 7/6/13/16. I had them 24/22/15/9 (well done, precisely the wrong order!) but now have them 13/14/15/16.


OK... fair enough. Its just that players from different eras are jumping in between these guys. I have Thompson and Sheckard below Pike, but above Pearce. I have the similarly abbreviated careers of CJones, McGraw, Cravath better than Meyerle. I have Browning better than Meyerle. I happened to place the OF glut guys above Meyerle as well (VanHaltren, Ryan, Tiernan)... after that I don't really track the players much.

Pearce is an oddity because he was a SS from a earlier era, but I found room for a lot of people between Pike & Meyerle. Pike has 40% more MLB plate appearances and also about 2 extra years more before the NA too (where he played better than Meyerle) Perhaps I'm overrating Pike, but its easy to sneak a lot of players in between him and Meyerle.

Of course, large ballot position differences between players that may not be that different is typical of a thin ballot like this.
   233. karlmagnus Posted: May 28, 2004 at 04:12 PM (#648956)
I'd argue that Pike to Meyerle is a very small gap indeed. Meyerle's rate stats are significantly better, plus he was a 3B. One can find arguments why say the 4 90s OF should be above all 4, or below all 4, but not in my view splitting them.

To each his own, I guess!
   234. KJOK Posted: May 28, 2004 at 04:21 PM (#648971)
MLB games are won and lost based on how much better a team's players are than the available replacement level--not the absolute level of difficulty or impact.
They are also won and lost based on how much better (or worse) a team's players are than average...

The *only* thing that matters at the major league level, as far as I'm concerned, is value above replacement.

Depends on the timeframe and what question your trying to answer. If you're star is injured and you have to play tomorrow, replacement level is probably what you're interested in.

If one of your starters is out for the year or leaving for free agency, you should probably be looking at value above average.

If you're looking for who the greatest players of all time were, you should probably be looking AT LEAST at value above average starter level...
   235. John (You Can Call Me Grandma) Murphy Posted: May 28, 2004 at 04:37 PM (#649010)
Meyerle played very well for Geary, but my opinion is that Geary was a not a top club.

It wasn't. He was a "minor leaguer" at this point.

He was a big fish in a small pond. Jumped up to Athletic in '69 and struggled. Improved a bit in 1870... but I wouldn't have predicted his big NA- bustout.

I was probably Meyerle's biggest supporter for a while at the beginning of this project. That changed once I analyzed his numbers using BP numbers for the NA. While impressive overall as a third baseman (he also played right field and second base), his fielding was sub par and he had a short career. His peak wasn't truly stellar to overcome his negatives.
   236. stephen Posted: May 28, 2004 at 04:57 PM (#649042)
Well, the first reason Beckley doesn't get to ride those others' coattails is because he's not as good as they were, and nobody seriously argues he was. Beckley has less hits than anyone else on that list, the lowest batting average, and the lowest OPS+. He's close to Crawford in batting average, but Crawford pretty much beats him everywhere else. The third worst BA on the list is Wagner .327, almost 20 points better. The only one close to his career OPS+ of 125 is Keeler's 127, but Keeler has a huge advantage in batting average (341-308). The third worst OPS+ on the list is Eddie Collins' 141, about 15 points better than Beckley.

OK, so he's close on hits to both Crawford and Keeler, all hovering around 2950. The next guy on the list is Lajoie with 3242, that's 300 hits away. Lajoie's lead over Beckley is the same as Beckley's lead over Lave Cross. I don't see how can you put Beckley on that list and call him a similar player. Maybe Crawford and Keeler, but he isn't even on the same planet as Collins, Speaker, Cobb, Wagner, and Lajoie. And head to head against his fellow just-short-of-300 club members, he rates behind them.

I find this to be an extremely unconvincing argument for Beckley.
   237. John (You Can Call Me Grandma) Murphy Posted: May 28, 2004 at 05:38 PM (#649110)
I'd argue that Pike to Meyerle is a very small gap indeed. Meyerle's rate stats are significantly better, plus he was a 3B.

Significantly better? I don't see it. Meyerle's OPS+ is a little better, but Pike had almost 600 more PA than Levi. Plus, Pike was a much better baserunner.

Meyerle did player a tougher position, so that's a plus (though he didn't play it well).

Pike was the best at his position(s) more times than Meyerle (and had a longer pre-NA career than Levi).

I'll take Pike.
   238. John (You Can Call Me Grandma) Murphy Posted: May 28, 2004 at 05:47 PM (#649127)
I now think all 4 of Meyerle, Pike, Wright and Pearce are very close to each other, with McVey and G. Wright significantly ahead.

I agree with you on the latter part, but Pearce's career length trumps the other three (and I don't see their peak easily beating Pearce either).
   239. John (You Can Call Me Grandma) Murphy Posted: May 28, 2004 at 05:52 PM (#649134)
-- Its got a box score from the 1858 All-Star Game that John Murphy keeps mentioning. HWright batted 9th and played "middle field' (cf?). Batting third and playing SS for Brooklyn was "Pidgeon" and seeing at how many other Atlantics are on the roster, I'm thinking that might be Dickey Pearce, but I'll never know for sure.

There was more than one game played that year, David. It's possible Pearce didn't play that game in question.

BTW, I really appreciate you digging up this info, David. Highly informative!
   240. TomH Posted: May 28, 2004 at 06:04 PM (#649157)
Jef M and Dan R, discussing catcher's defense, posted
I guess my question is, do we have any data on how much catcher defense affects the game at all? I can't imagine it is really comparable to 2B/SS/CF.
If they are so defensively valuable, why don't they accumulate 15+ fielding WS/FRAR?
((possible answer to quesiton)) Imperfect systems? Fielding is notoriously hard to evaluate, and particularly so for catchers. and a quote from Win Shares:
"At some positions, like this one, we are starved for data, and the best we can hope for is what might be considered a well-organized fishing expedition."
-----
I agree, but we might want to consider that catchers have historically been asked to help the pitcher by calling the game, framing the pitch, and other duties. Yes, it is very difficult to quanitfy this. But 20 years ago, very few people quantified fielding data very well for shortstops; we've made some big strides, haven't we?
I suggest that a low staff ERA is somewhat creditable to the catcher, as Craig Wright suggested, and as I have published elsewhere. I frimly believe that, in general, a catcher who playes a long time with the same team generates a benefit in run prevention. This is one legitimate (IMHO) reason to postulate a 'catcher bonus' to men who played many games behind the dish. Maybe in the next generation as we analyze play-by-play data we'll get better analysis to quantify this.
   241. OCF Posted: May 28, 2004 at 06:51 PM (#649228)
I suggest that a low staff ERA is somewhat creditable to the catcher, as Craig Wright suggested, and as I have published elsewhere. I frimly believe that, in general, a catcher who playes a long time with the same team generates a benefit in run prevention. This is one legitimate (IMHO) reason to postulate a 'catcher bonus' to men who played many games behind the dish.

But is it enough to make an argument for Johnny Kling?
   242. Dag Nabbit is a cornucopia of errors Posted: May 28, 2004 at 07:07 PM (#649260)
I was looking over the credentials of Konetchy in relation to Beckley. and I was wondering if someone who supports Beckley could give me anargument of why he desrves to be on the ballot instead of Konetchy. The only real advantage I see using Win Shares is the career number, which was accumulated over more seasons. Konetchy has a slightly better peak and rate per 162 games.

Really, I'm unimpressed by both of these 1st base candidates, but I'm getting the feeling Beckley has much more support than Konethcy and I can't really figure out why.


Well you admit that Beckley's got better career numbers but you can't understand why someone would vote for him? Some of place a heavier emphasis on career value.
   243. ronw Posted: May 28, 2004 at 07:18 PM (#649285)
Re: Beckley and Konetchy

I think the reason Konetchy is behind Beckley among those voting for Beckley is simple: About 5 years. Those voting for Jake are not taking peak into account very much, so Konetchy's higher peak is not that impressive to them. Five years, however, is very impressive to a career voter.

I tend to agree with karl on this one. It seems that sabermetrics help identify overlooked players, but can also be used to downplay consistent longetivity. Beckley played for 26 years, and in only a few of those years was he below All-Star level. His sabermetric problem is that he was rarely much above All-Star level. He was an All-Star when there were three leagues, two leagues, one league, and two again. He was an All-Star both before and after the 1893 pitching distance change. He was an All-Star (admittedly a cut below) when Brouthers, Connor, and even Anson were still dominating the league, and he was an All-Star when Chance, Harry Davis, and Fred Tenney were the premier 1B. Beckley really didn't have Pete Rose hanging-on value, he was among the top 1B in the league in the vast majority of his seasons, at least through 1905. Some might say he played at the level of his competition. When the 1B were strong, he had his best seasons, when they were weak, he has his worst, but he was always just below the top 1B in the game. For 20 years, that's impressive to me.

Here's a little positional breakdown on Jake's career, by WS:

1888 NL - 14 (4th, behind Connor 32, Anson 29 and Brouthers 27)
1889 NL - 19 (4th, behind Brouthers 28, Connor 26, Anson 21)
1890 PL - 21 (2nd, behind Connor 25)
1891 NL - 16 (T3rd with Foutz, behind Connor 23, Anson 21)
1892 NL - 19 (T5th with Anson, behind Brouthers 34, Connor 24, Virtue 22, Ewing 20)
1893 NL - 17 (1st)
1894 NL - 17 (T2nd with Doyle, behind Brouthers 21)
1895 NL - 18 (2nd, behind LaChance 19)
1896 NL - 10 (T4th with Cartwright, LaChance and Tucker, behind Doyle 17, Connor 14, Anson 12)
1897 NL - 15 (T3rd with Tenney and Werden, behind Lajoie 21, Doyle 18)
1898 NL - 14 (6th, behind Joyce 25, Wagner 22, McGann 21, Everitt 18, Tenney 17)
1899 NL - 20 (T2nd with McGann, behind Tenney 25)
1900 NL - 21 (1st)
1901 NL - 18 (T1st with Kelley)
1902 NL - 18 (2nd, behind Tenney 25)
1903 NL - 17 (4th, behind Chance 31, Tenney 21, Doyle 19)
1904 NL - 23 (2nd, behind Chance 29)
1905 NL - 16 (T5th with Gessler, behind Chance 25, McGann 24, Barry 19, Tenney 17)
1906 NL - 5 (7th, behind Chance 35, Jordan 25, Nealon 18, Tenney 17, McGann 16, and Bransfield 14)
1907 NL - 0
   244. ronw Posted: May 28, 2004 at 07:37 PM (#649330)
Another overlooked fact is that Beckley almost always played for horrible teams, a fact which lowers his WS. However, keeping with his theme of playing with the competition, he was rarely the best player on his team, but was almost always one of the best.

Here's his team rank each year of his career:

1888 Pitt NL - 14 (2nd, behind Kuehne 18)
1889 Pitt NL - 19 (2nd, behind Caroll 23)
1890 Pitt PL - 21 (1st)
1891 Pitt NL - 16 (2nd, behind Miller 17)
1892 Pitt NL - 19 (4th, behind Smith 31, Miller 24, Shugart 22)
1893 Pitt NL - 17 (4th, behind Smith 25, Lyons 21, Van Haltren 20)
1894 Pitt NL - 17 (3rd, behind Stenzel 24, Smith 19)
1895 Pitt NL - 18 (2nd, behind Stenzel 28)
1896 Pitt/NYG NL - 10 (overall behind Tiernan, Van Haltren, Davis and Gleason on NYG and behind Smith, Stenzel, Donovan, Lyons and Ely on Pitt)
1897 NYG/Cin NL - 15 (overall behind Davis, Van Haltren, Tiernan, Joyce and Gleason on NYG, and behind Irwin and Hoy on Cin)
1898 Cin NL - 14 (6th, behind Smith 27, McBride 21, Miller 21, Corcoran 18, McPhee 15)
1899 Cin NL - 20 (2nd, behind Selbach 23)
1900 Cin NL - 21 (2nd, behind Barrett 23)
1901 Cin NL - 18 (2nd, behind Crawford 24)
1902 Cin NL - 18 (2nd, behind Crawford 23)
1903 Cin NL - 17 (4th, behind Seymour 24, Donlin 24, Steinfeldt 21)
1904 StL NL - 23 (1st)
1905 StL NL - 16 (T2nd with Grady, behind Smoot 23)
1906 StL NL - 5 (6th or 7th or 8th, behind Bennett, Grady, Arndt, Burch, and some traded StL players)
1907 StL NL - 0

I don't think my posts will push a peak voter in Beckley's direction, but they at least show that he was a solid player for a long time.
   245. Sean Gilman Posted: May 28, 2004 at 08:39 PM (#649437)
Another overlooked fact is that Beckley almost always played for horrible teams, a fact which lowers his WS.

No, it doesn't. Win Shares doesn't work that way.
   246. John (You Can Call Me Grandma) Murphy Posted: May 28, 2004 at 09:35 PM (#649489)
I'm glad we're having this Beckley/Konetchy discussion because I almost overlooked the latter. I'm going to have Konetchy in my top ten. He played in a more competitive time than Beckley and was even a little bit better first baseman and base runner to boot. I have him as the best major league first baseman for 1910, 1911 and 1916 (very close in 1909 and 1911) and the best NL first baseman for 1909, 1911 and 1919.
   247. ronw Posted: May 28, 2004 at 10:14 PM (#649530)
Sean is right. Win Shares does not penalize players for playing on bad teams. In broad generalizations, the good teams have more available win shares, but also more good players to take them. The bad teams have less win shares, but less good players to take them.

Player A on the 1927 Yankees (136 OPS+) and Player B on the 2003 Tigers (142 OPS+) have roughly the same relative offensive statistics.
Although they play different positions, they are surprisingly roughly equivalent defensively. Despite the extreme records of these teams, under Bill James' system Player A receives 21 WS, and Player B receives 19 WS. Player A has to compete with the likes of Ruth and Gehrig, while Player B is clearly the best player on his team. I won't go through the math, but Player A is Bob Meusel, and Player B is Dmitri Young. I could probably do a more similar player comparison, but I had these two teams handy.

James' WS book makes similar detailed comparisons of pitchers on good and bad teams, and reaches Sean Gilman's conclusion.
   248. ronw Posted: May 28, 2004 at 10:24 PM (#649539)
Here's the fun of Ed Konetchy when compared to other league 1B (who is in the mid 20's on my ballot). It's also interesting that Ed replaced Jake Beckley at 1B for the Cardinals in 1907.

1907 NL - 10 (6th, behind Chance 23, Jordan 22, Tenney 22, Ganzel 14, McGann 11)
1908 NL - 18 (4th, behind Bransfield 20, Chance 20, Tenney 19)
1909 NL - 24 (1st)
1910 NL - 21 (1st)
1911 NL - 26 (1st)
1912 NL - 22 (1st)
1913 NL - 19 (2nd, behind Saier 26)
1914 NL - 13 (6th, behind Saier 24, Miller 21, Daubert 19, Schmidt 19, Merkle 17)
1915 FL - 27 (1st)
1916 NL - 26 (1st)
1917 NL - 17 (T 3rd with Chase and Holke, behind Merkle 19, Luderus 18)
1918 NL - 9 (T 6th with Chase and Mollwitz, behind Merkle 22, Magee 18, Daubert 15, Luderus 15, Paulette 11)
1919 NL - 18 (T1st with Luderus)
1920 NL - 18 (3rd, behind Daubert 24, Fournier 19)
1921 NL - 13 (5th, behind Fournier 24, Kelly 24, Grimes 21, Grimm 14)
   249. ronw Posted: May 28, 2004 at 10:32 PM (#649545)
And team by team:


Here's the fun of Ed Konetchy when compared to other league 1B (who is in the mid 20's on my ballot). It's also interesting that Ed replaced Jake Beckley at 1B for the Cardinals in 1907.

1907 StL NL - 10 (4th, behind Murray 16, Byrne 15, Holly 11)
1908 StL NL - 18 (2nd, behind Murray 27)
1909 StL NL - 24 (1st)
1910 StL NL - 21 (1st)
1911 StL NL - 26 (1st)
1912 StL NL - 22 (1st)
1913 StL NL - 19 (1st)
1914 Pitt NL - 14 (4th, behind Wagner 19, Carey 17, Viox 16)
1915 Pitt FL - 27 (1st)
1916 BosB NL - 26 (2nd, behind Maranville 27)
1917 BosB NL - 17 (3rd, behind Maranville 22, Smith 22)
1918 BosB NL - 9 (T3rd with Herzog, behind Smith 21, Wickland 15)
1919 Broo NL - 18 (4th, behind Myers 23, Wheat 21, Olson 19)
1920 Broo NL - 18 (T 4th with Kilduff, behind Wheat 28, Myers 27, Johnston 20)
1921 Broo/Phil NL - 13 (behind Johnston 24, Wheat 23, Griffith 17 on Broo; behind Williams 17 on Phil)

Konetchy may have played 5 less years, but he was likely the top player on that horrible Cardinal team from 1909-1913, and also the best 1B in the National League.
   250. ronw Posted: May 28, 2004 at 10:33 PM (#649547)
Delete the second paragraph from the second post from your minds.
   251. KJOK Posted: May 28, 2004 at 11:25 PM (#649638)
I think Mr. Chance belongs in this conversation, as HE was the BEST deadball era 1st baseman:

Offensive Win%
Chance - .720
Konetchy - .603
Beckley - .596

Runs Above Average 1st baseman:
Chance - 309
Beckley - 245
Konetchy - 180

FIELDING
Konetchy - EXCELLENT
Chance - VERY GOOD
Beckley - AVERAGE

PLAYED ON PENNANT WINNER
Chance - 4
Konetchy - 1
Beckley - 0
   252. John (You Can Call Me Grandma) Murphy Posted: May 28, 2004 at 11:42 PM (#649668)
I think Mr. Chance belongs in this conversation, as HE was the BEST deadball era 1st baseman:

I have him on my ballot so I agree, but Chance just didn't play as long as Konetchy.
   253. John (You Can Call Me Grandma) Murphy Posted: May 28, 2004 at 11:48 PM (#649680)
I have him on my ballot so I agree

That is, I agree that he belongs in the conversation, but not that he's the best deadball first baseman.
   254. KJOK Posted: May 29, 2004 at 12:04 AM (#649719)
Using OWP, playing time, and defense (Win Shares/BP) for position players, applied to .500 baseline. Using Runs Saved Above Average and Support Neutral Fibonacci Wins for Pitchers.

1. JOHN McGRAW, 3B. .727 OWP. 459 RCAP. 4,909 PAs. Def: AVERAGE. Was CAREER ALL-TIME OBP% leader until Ruth qualifies in 1923, EVEN adjusting for League, and is STILL #3 behind Williams and Ruth. He didn’t have a long career, but he’s being discounted for low playing time way too much as he provided more value in those few appearances than all of his contemporary 3rd baseman.

2. JOE JACKSON, LF/RF. .780 OWP. 449 RCAP. 5,690 PA’s. Def: AVERAGE. Best hitter of his generation except for Cobb. Had “MVP-type” years 1911-1913.

3. PETE BROWNING, CF/LF. .745 OWP. 478 RCAP. 5,315 PAs. Def: POOR. Baseball’s premier hitter in the 1880’s. Much better hitter than Thompson.

4. HUGHIE JENNINGS, SS. .607 OWP. 263 RCAP. 5,650 PAs. Def: EXCELLENT. Best SS of the 1890’s. Great offensively and defensively.

5. ROGER BRESNAHAN, C. .651 OWP. 282 RCAP, 5,373 PA’s. Def: AVERAGE. Best Catcher between Ewing and Cochrane/Dickey.

6. RUBE WADDELL, P. 254 RSAA, 222 Neut_Fibonacci_Wins, and 134 ERA+ in 2,961 innings.

7. JOE McGINNITY, P. 238 RSAA, 208 Neut_Fibonacci_Wins, and 121 ERA+ in 3,441 innings.

8. FRANK CHANCE, 1B. .720 OWP. 308 RCAP. 5,099 PAs. Def: VERY GOOD. Excellent hitter and good fielder back when 1st base was more important defensively. Leader of one of the greatest teams in history, and the next inductee from that team should be Chance, not Sheckard.

9. CUPID CHILDS, 2B. .609 OWP. 354 RCAP. 6,762 PAs. Def: AVERAGE. Best 2nd baseman of the 1890’s. Best remaining 2nd baseman.

10. BOBBY WALLACE, SS. .522 OWP (.546 thru 1910). 195 RCAP (211 thru 1910). 9,612 PAs. Def: VERY GOOD. Hung around ala Pete Rose after 1910. Also outstanding defensive 3B in the 2 years he played there. Not quite the player Jennings was.

11. TOMMY LEACH, CF/3B. .552 OWP, 121 RCAP, 9,051 PA’s. Def: EXCELLENT – 3B, VERY GOOD – CF. Just slightly below Collins defensively, and he played longer. Basically did everything well, but doesn’t have the one outstanding area to get noticed.

12. LARRY DOYE, 2B .632 OWP, 273 RCAP, 7,382 PA’s. Def: FAIR. Best hitting 2B between Lajoie and Hornsby. Won MVP in 1912, finished 3rd in 1911. Finished in Top 10 in OPS+ 8 times.

13. BOB CARUTHERS, P/RF. 179 RSAA. 177 Neut. Fibonacci Wins, 123 ERA+. .668 OWP. 243 RCAP. 2,906 PAs. Only shortness of career keeps Caruthers from being an “inner circle” superstar.

14. SAM THOMPSON, RF. .684 OWP. 387 RCAP. 6,510 PAs. Def: AVERAGE. Not the hitter Browning was, but still an offensive force.


15. DICKEY PEARCE, SS. He WAS basically, along with Harry Wright, the old guy in the league 1871-1877, and his fielding was still league average, but didn’t hit nearly as well as Harry (who played CF). May have been Ozzie Smith, but hard to tell for certain. However, I’m finally convinced there is enough evidence to place him in the top 15.


LEFT OFF THE BALLOT:
JIMMY SHECKARD, LF. .626 OWP. 135 RCAP. 9,117 PAs. Def: EXCELLENT. Another player who was good but not great offensively, played a long time. Don’t see why so much love for him. He was a great defensive LFer, but in era with fewer fly balls. Similar to Hugh Duffy.
TONY MULLANE, P. 241 RSAA, 240 Neut_Fibonacci_Wins, and 118 ERA+ in 4,531(!) innings. AA discount puts him off ballot until I finally get around to my AA vs. NL study.
LIP PIKE, CF. Perhaps best hitting CF of the 1870’s, but short career puts him off ballot.
JAKE BECKLEY, 1B. .596 OWP. 245 RCAP. 10,492 PAs. Def: AVERAGE. Another good for a long time player who is just below elite status.
GEORGE VAN HALTREN, CF. .620 OWP. 167 RCAP. 8,992 PAs. Def: FAIR. A notch below the elite OF’ers both offensively and defensively.
PETE HILL, CF/LF. . Right now looking like Dale Murphy/Andre Dawson, so is just off the ballot.
ED KONETCHY, 1B. .603 OWP. 180 RCAP. 8,664 PA’s. Def: EXCELLENT. Excellent defensive 1Bman in era when that mattered, so he edges just ahead of Jimmy Ryan.
JIMMY RYAN, CF/RF. .609 OWP. 205 RCAP. 9,114 PAs. Def: AVERAGE. Not quite up to top OF hitters, and only average defense won’t move him up.
   255. ronw Posted: May 29, 2004 at 12:43 AM (#649848)
Fine. Frank Chance compared to others at the same position in the same league, WS:

1898 - 4 (behind a lot of other C, including his own teammate, Tim Donahue)
1899 - 7 (still behind most starters, including his own teammate, Tim Donaohue)
1900 - 7 (T4th with McGuire, behind McFarland 14, Zimmer 10, Peitz 9)
1901 - 9 (backup RF, behind most starting RF, including teammate Jock Menefee)
1902 - 13 (5th at 1B, behind Tenney 25, Beckley 18, Bransfield 15, McGann 14)
1903 - 31 (1st)
1904 - 29 (1st)
1905 - 25 (1st)
1906 - 35 (1st)
1907 - 23 (1st)
1908 - 20 (T1st with Bransfield)
1909 - 14 (T3rd with Jordan and Bransfield, behind Konetchy 24, Hoblitzell 23)
1910 - 14 (5th, behind Konetchy 27, Merkle 20, Hoblitzell 18, Daubert 17)
1911 - 5 (behind all starters, including teammate Vic Saier 6)
1912 - 0
1913 - 1

That's 6 great years and 2 solid years, by my count, compared to 16 solid years for Beckley and 11 solid years for for Konetchy.
   256. ronw Posted: May 29, 2004 at 12:51 AM (#649869)
And more from the Peerless Leader, compared to team:

1898 - 4 (behind most of his team, including his own positionmate, Tim Donahue 10)
1899 - 7 (still behind most of his team, including his own positionmate, Tim Donahue 8)
1900 - 7 (T7th with Ganzel, behind all other starters Mertes 18, Bradley 16, Green 14, Childs 12, McCarthy 12, Ryan 12)
1901 - 9 (4th, behind Hartsel 27, Green 20, Menefee 11)
1902 - 13 (6th, behind Slagle 23, Kling 17, Tinker 17, Menefee 14, DJones 14)
1903 - 31 (1st)
1904 - 29 (1st)
1905 - 25 (T1st with Slagle)
1906 - 35 (1st)
1907 - 23 (T1st with Sheckard)
1908 - 20 (4th, behind Tinker 32, Evers 28, Kling 22)
1909 - 14 (7th, behind Evers 27, Hofman 27, Steinfeldt 25, Tinker 24, Sheckard 23, Schulte 18)
1910 - 14 (T7th with Kling, behind Hofman 31, Schulte 26, Sheckard 23, Evers 22, Tinker 22, Steinfeldt 15)
1911 - 5 (behind all starters, including positionmate Vic Saier 6)
1912 - 0
1913 - 1

Had only 6 years as the primary team weapon. A very good peak, almost Jenningsian, plus one year.
   257. sunnyday2 Posted: May 29, 2004 at 01:12 AM (#649944)
Speaking as perhaps the main FO1860s, I hope DavidFoss is still around for the weekend. David, can we get Al Reach? And then I think we've probably got the entire generation (worthies, that is) who straddled 1871.

I've devised a simple rating scheme for this generation:

• The amateur era through 1867-68-69, depending on the player's level: players are rated on (R/G +/- other factors) X seasons. e.g. Dickey Pearce scored about 3.25 R/G for 10 years, played important defensive positions (that generally earn about 67% of their WS value on offense, 33% on defense), and played with the best team of the era, meaning his R/G probably came in more than an avg. number of PA. So Pearce = 3.25 + 1.625 - 10% = 4.4 x 10 seasons ('57-'66) = 44 "WS"

• The early pro era 1867-70 for some players, '69-'70 for others. In this case Pearce played with one of the best teams, which I push into the "pro" category in '67, so he's got 4 years at that level. Same formula X years X 2, in the soft belief that the competition was twice as good. Pearce earns 4.95 (10% discount already in there) X 4 X 2 = 39.6 "WS"

• The NA/early NL era. Pearce played 6 more years with OPS+ 80 and +4.3 FR/year. 9.8 (above a hypothetical defensive replacement) - 2.0 x 6 x 1.5 (6 years, 3X the difficulty/but while the above indeces revolve around 5 this one revolves around 10) = 71.4 "WS" (I am obviously not going to defend this number, it is simply an effort to find a reasonable value that revolves around "5".)

Pearce's "value" is 44 + 39.6 + 71.4 = 155

Next example:

Harry Wright played 8 years in the amateur era, much of it at P-SS-C with 3 R/G on lousy teams (fewer PA), so he gets 3.0 + 2.0 + 10% = 5.5 "WS" x 8 = 44. it is a total coincidence that he ties Pearce. However, he plays 2 fewer years (i.e. 2 years in '69-'70) at an early pro (elite) level, so those 44 "WS" are in 2 "extra" years. As an early pro, he scores just 5 "WS" x 2 years x 2 (for difficulty) or 20. In the NA he plays 4 years not terribly effectively and scores 1.7 X 4 X 1.5 = 10.2

I'll spare you the rest of the details, but just a couple highlights.

G. Wright has just 2 years at an amateur level, 3 as an early pro (Wash Nationals '66 + '69-'70) and of course 10 years in the NA/NL. Was a C initially, then SS.

For Pike it's 3 + 3 (PhiAth in '66 + '69-'70) + 8 at OF-3B-2B-CF
Meyerle 2 + 2 + 7 as a P and then a lousy 3B
McVey 0 + 2 + 9

Totals

Pearce 44 + 39.6 + 71.4 = 155 "WS"
H. Wright 44 + 20 + 10.2 = 74.2
G. Wright 9 + 48 + 181.5 = 238.5
Pike 9.9 + 36 + 109.4 = 155.3
Meyerle 15 (remember he was a P) + 16 + 66.2 = 97.2
McVey 0 + 16.4 + 116.1 = 132.5


The numbers, of course, are more in the way of indices, not "values." But there's something here for everybody. The basic formula is rate X years for you career voters, there's a steep timeline 1-2-3. Certainly the NA/NL numbers pass the sniff test:

G. Wright 181.5
McVey 116.1
Pike 109.4--McVey and Pike close as they should be, Pike ahead on value/year 9.1-8.6, but McVey with one more year
Meyerle 97.2
Pearce 71.4
H. Wright 10.2--not much above average probably

For the pre-NA the scores are:

Pearce 83.6 (14 years)
H. Wright 64 (10 years)
G. Wright 57 (5 years)
Pike 45.9 (6 years)
Meyerle 31 (4 years)
McVey 16.4 (2 years)

Remember Meyerle was a P for 2 years.

So, we knew G. Wright was the best of this crew. Pearce and Pike are pretty close, is that a surprise? I mean Pearce played 20 years, Pike 14, but Dickey's first 6 (10 really, but 6 more than Lip) were against weaker competition. I personally will continue to rank Pearce ahead of Pike but, hey, they're close.

I also think this underrates McVey. Pike beats him by 23 "WS," but was ahead by 30 as of the start of the 1871 season. Maybe Lip shoulda rated higher, especially for me since I value pre-1870.

So, again, I realize these numbers do not translate to anything (runs, wins, etc.) but I am compelled by them nevertheless to move Harry Wright down. Dickey Pearce will stay where he is on my ballot--#1 or 2. And Pike, who has been hovering around #10 on my ballot for many years, will probably move up based on these numbers.
   258. sunnyday2 Posted: May 29, 2004 at 01:21 AM (#649983)
Re. the 1B glut...I keep going back and forth as to whether Beckley or Chance is the top guy in the pool. And as a peak/prime voter, that tells how crazy about Chance I am...NOT.

As for Beckley, he may have ranked highly against he peers but when your peers are Fred Tenney and Candy LaChance and Jake Stenzel I am prepared to be underwhelmed.

I looked at Konetchy and since he clearly falls behind Beckley, even for me, and Beckley has yet to crack my top 20, ho as in ho, hum as in hum. Ho hum. ;-)
   259. Kelly in SD Posted: May 29, 2004 at 02:34 AM (#650230)
Some Mickey Welch facts and stats (not including 1891 cos I was too tired. He only went 7-10.):

Against current members of the HoM:

Pud Galvin: 26-11
John Clarkson: 12-5
Hoss Radbourn: 17-10
John M. Ward: 4-7
Amos Rusie: 2-0
Kid Nichols: 0-1

Against other notables:
McCormick: 18-20
Larry Corcoran: 6-13
Goldsmith: 3-7
Jim Whitney: 14-13
Charles Buffington: 15-9

Against teams over .500: 129-130 or .498
Against teams below .500: 179-73 or .710

Year by Year against teams by finish (his team's finish is indicated by the team name)

80: (0-7) (5-6) (3-6) (Try) (5-5) (5-6) (9-0) (7-0)
81: (0-2) (2-2) (9-2) (2-3) (Try) (0-3) (5-3) (3-3)
82: (2-3) (0-2) (4-2) (3-3) (2-5) (1-1) (Try) (3-0)
83: (4-4) (4-4) (3-3) (4-2) (3-5) (NYG) (4-3) (6-2)
84: (3-3) (6-5) (5-2) (3-6) (NYG) (7-1) (7-4) (8-1)
85: (7-2) (NYG) (7-1) (6-1) (7-2) (6-2) (5-1) (6-2)
86: (5-6) (2-5) (NYG) (2-3) (5-3) (6-2) (8-1) (5-3)
87: (0-1) (2-3) (5-1) (NYG) (4-3) (5-2) (1-4) (6-1)
88: (NYG) (3-6) (5-4) (6-3) (1-2) (4-2) (3-1) (4-1)
89: (NYG) (1-3) (4-1) (5-1) (2-3) (5-1) (6-2) (5-0)
90: (1-2) (2-3) (3-1) (3-3) (3-2) (NYG) (3-2) (6-1)

Also, it looks like he was injured in 1887 when he went 22-15 while Keefe went 35-19. He only made 8 starts over a two month period between mid-May and mid-July, instead on his usual every-other game routine with Keefe. Up to mid-July, he was 7-8, but afterward, he did not miss a start until he and Keefe got the last two weeks off and went 16-7.
In 1889, he kept the Giants in the race single-handedly. At the end of July, he was 19-6; the rest 25-23. Then the team went on a 39-14 run to finish the season.
This was the second season in a row the Giants got hot. In 1888, the went 41-13 in the middle of the season when Keefe went 21-3 and Welch was 16-6.

I hope these numbers help the discussion.
   260. DavidFoss Posted: May 29, 2004 at 03:00 AM (#650324)
pre-NA number for AL REACH:

============================

AL REACH

1861 -- Brooklyn Eckford (8-4)
-- Pos -- 3B-1B-SS
-- Competition -- NYC+/Newark
-- Hitting -- 8th on team with 2.22 R/G (Campbell 4.17, Manolt 3.37, Wood 3.2)
1862 -- Brooklyn Eckford (14-2) (8 games with boxes)
-- Pos -- 1B-OF-3B-SS
-- Competition -- NYC+/Philly/Newark
-- Hitting -- 5th on team with 4.17 R/G (Mills 4.88, Manolt 4.63, Wood 4.5)
1863 -- Brooklyn Eckford (10-0)
-- Pos -- 1B
-- Competition -- NYC+/Philly
-- Hitting -- 5th on team with 2.0 R/G (Wood 3.5, Sprague 3.11, Manolt/Beach 3.0)
1864 -- Brooklyn Eckford (1-4)
-- Pos -- ??
-- Competition: NYC+/Newark
-- Hitting -- 4th on team with 2.75 R/G (Wood 4.0, Manolt 2.8, Brown 2.8)
1865 -- Philadephia Athletic (15-3)
-- Pos -- 2B
-- Competition -- NYC+/Philly/Newark/Altoona
-- Hitting -- 5th on team with 3.8 R/G (McBride 4.0, MSmith 4.0, Berkenstock 3.93)
1866 -- Philadephia Athletic (23-2)
-- Pos -- 2B
-- Competition -- Philly/NJ/NYC+/Wash/WilkesBarre
-- Hitting -- 5th on team with 5.83 R/G (McBride 6.4, Hayhurst 6.31 (13G), Pike 6.25)
1867 -- Philadephia Athletic (44-3)
-- Pos -- 2B
-- Competition -- Philly/Wash/NYC+/Boston
-- Hitting -- 1st on team with 6.0 R/G (McBride 5.89, Sensenderfer 5.60, Kleinfelder/Fisler 5.33)
1868 -- Philadephia Athletic (47-3)
-- Pos -- 2B
-- Competition -- Philly/Pit/Cin/Lou/STL/Chi/Rock/Roch/Syr/NYC+/Wash/Alb/Cle/NJ
-- Hitting (no H-data)
---- Runs -- 1st on team with 5.14 R/G (Radcliffe 5.0, Fisler 4.91, Cuthbert 4.8)
---- TB -- 2nd on team with 6.10 TB/G (Fisler 6.47, Cuthbert 5.6, Sensenderfer 5.42)
1869 -- Philadelphia Athletic (45-8) (Pro:15-7)
-- Pos -- 2B
-- Competition -- Philly/NJ/Cincy/NYC+/Bos/Bal/Wash/PA/Alb
-- Hitting (all games)
---- 1st on team with 5.39 R/G (Cuthbert 5.13, McBride/Sensenderfer 5.00)
---- 2nd on team with 5.26 R/G (McBride 5.35, Sensenderfer 4.56, McMullin 4.53)
---- 1st on team with 9.11 TB/G (McBride 8.18, Cuthbert 7.70, Fisler 7.67)
1870 -- Philadephia Athletic (65-11-1) (Pro: 26-11-1)
-- Pos -- 2B
-- Competition -- Phi/Bal/NYC+/CT/Cin/NJ/Chi/Wash/FtW/Cle/Buf/Bos
-- Hitting (pro games only)
---- Hits -- 5th on team with 2.02 H/G (Fisler 2.36, Malone 2.32, McBride/Sensenderfer 2.30)
---- TB -- 6th on team with 3.05 TB/G (Fisler 3.86, McBride 3.66, Malone 3.56)
   261. DavidFoss Posted: May 29, 2004 at 03:10 AM (#650354)
There you go Marc/sunnyday2... you caught me just before I went out of down. Traffic should be really light now.

Reach certainly knew how to play for winners.... and he had a monster hitting spree from 67-69. Was 2B an easy position back then? I tend to recall that it was, but I'm not sure.

GWright's already inducted into the HOM so it doesn't matter, but don't discount his Union of Morrisania numbers. They were a major club when he was there. I popped Morrisania into mapquest figuring it'd be way upstate, but its in the heart of the Bronx.

I don't have time to look at your analysis at the moment, but it sounds like you need to buy this book, Marc!

Have a great holiday weekend everyone!
   262. KJOK Posted: May 29, 2004 at 06:06 AM (#650465)
Fine. Frank Chance compared to others at the same position in the same league, WS:

Win Shares, with it's 20% baseline, is always going to undervalue players with great but shorter careers like Chance vs. players with lesser but longer careers like Beckley.

And even most people here who are believers in WARP would probably (possibly?) agree a 20% baseline is too low to measure value...
   263. John (You Can Call Me Grandma) Murphy Posted: May 29, 2004 at 06:13 AM (#650466)
Win Shares, with it's 20% baseline, is always going to undervalue players with great but shorter careers like Chance vs. players with lesser but longer careers like Beckley.

That's why I factor in WS per 162 games.
   264. David Concepcion de la Desviacion Estandar (Dan R) Posted: May 29, 2004 at 08:09 AM (#650474)
What is the 20% baseline? I'm not familiar...
   265. Daryn Posted: May 29, 2004 at 01:39 PM (#650491)
Dan, it is a very low replacement level, which I like BTW.
   266. ronw Posted: May 29, 2004 at 03:52 PM (#650508)
I would generally think that career numbers would be undervalued with WS, but not the year-by-year comparison I posted above. I haven't done it with WARP (because I don't know how it is calculated) but I would think the year-by-year positioning wouldn't be too far off.
   267. KJOK Posted: May 29, 2004 at 07:02 PM (#650628)
I realize this is a bit of selective endpointing, but it does show what type of player Chance was in his prime:

1902-1908 RCAA
1 Honus Wagner 562
2 Nap Lajoie 327
3 Frank Chance 298
4 Elmer Flick 276
5 Sam Crawford 258
6 Fred Clarke 223
7 Mike Donlin 222
T8 Topsy Hartsel 202
T8 George Stone 202
10 Roger Bresnahan 193
   268. KJOK Posted: May 29, 2004 at 07:48 PM (#650656)
3rd Basemen - All-time RCAP OWP PA
1 Eddie Mathews 633 .704 10101
2 Mike Schmidt 576 .688 10062
3 Wade Boggs 575 .665 10740
4 George Brett 508 .663 11624
5 John McGraw 459 .727 4939
6 Chipper Jones 391 .692 6067
7 Stan Hack 370 .631 8506
8 Bill Joyce 367 .697 4154
9 Home Run Baker 364 .659 6659
10 Denny Lyons 326 .658 5021
11 Toby Harrah 283 .579 8766
12 Bob Elliott 241 .610 8190
13 Sal Bando 220 .597 8288
T14 Deacon White 216 .600 5668
T14 Heine Groh 216 .598 7035
T16 Pie Traynor 209 .529 8293
T16 Harlond Clift 209 .566 6894
18 Al Rosen 205 .666 4374
19 Ron Santo 200 .588 9396

Sure McGraw had a short career, but he could have 3 NEGATIVE 30 RCAP years and he'd still be ahead of Home Run Baker in roughly the same # of PA's.
   269. KJOK Posted: May 29, 2004 at 08:04 PM (#650668)
Just a reminder, the defensive spectrum pre-1920 was roughly:

C....SS...3B.2B...1B..CF.LF.RF
   270. Esteban Rivera Posted: May 30, 2004 at 12:22 AM (#651030)
Hi everyone. I'm going to be off the island for the next week or so and I do not know when I'll have computer access again. Therefore, I'd like to post a provisional ballot for this election in the discussion thread, if it's alright with you guys.
If I don't post again by Friday, would you please count the following as my ballot for 1927? Thanks and have a great Memorial Day weekend.

1. Joe Jackson - With the one year boycott done, Jackson debuts at number one. Granted, I understand he was not one of the dominant superstars of his time( i.e. Cobb, Cpeaker, Collins, etc.), but none of them are on this ballot. Jackson's tenure edges the eligible field.

2. Sam Thompson - A heck of an offensive machine. Reputed to have the best arm of his time. Doesn't the 1890's Philadelphia outfield kind of resemble the mid 1990's Cleveland outfield?

3. Joe McGinnity - Compiled an awesome record in only a decade and began past the usual starting age for a ballplayer in the majors. The best pitcher or runner up for half his career.

4. Lip Pike - One of the best players in early baseball. Definitely deserves more attention.

5. Pete Browning - Was a heck of a hitter and did it under tremendous duress. I buy the "greatness can't take full advantage off lower competition" idea. Proved he could hold his own in the player's league.

6. Rube Waddell - Was a special picher. I buy the run support analysis and also believe in the higher value of being a phenomenal K artist in his time and place. His career record isn't that impressive but you have to remember that there were some stretches where he was jettisoned because his managers did not know how to deal with his unique personality.

7. Bill Monroe - Keep gaining confidence in him. Seems to be one of the best second basemen of his time.

8. Hughie Jennings - A historical monster for five years.

9. Hugh Duffy - His credentials are that he was for a time one of the best players and he produced during the 90's. Then he just fell off. However, I feel his peak gives him the slight edge over Ryan and Van Haltren.

10. Mickey Welch - Flip flops with Griffith this week as I begin to feel that I may have overreacted earlier in his downgrading.

11. Clark Griffith - The more that I look at him the more I realize I have been underestimating his accomplishments.

12. Jake Beckley - The career man makes it back on my ballot. There is a definite dearth of long career's at first.

13. Bob Caruthers - Head for the hills! It's Parisian Bob! The combo man finally reappears on my ballot. Let's see if its to stay.

14. Roger Bresnahan - Hanging by a thread. He has his favorable points such as his offense and being versatile. However, playing time and defensive issues make me a bit wary of going higher with him. Will most likely wait until he can be properly placed alongside Santop and Petway.

15. Pete Hill - . I see Hill as a worthy candidate. Seems to be a very good defensive player and his hitting for the first half of his career seems to be great. However, the second half of his career seems to be an adventure in park effects. Until the "mess" is sorted out, Hill starts here.

Jimmy Scheckard is next in line but the addition of Hill (who I place above Sheckard) and Jackson still keep him off ballot.. After him are Wallace (long consistent career but not the best at his postion so he does not yet get on my ballot), Williamson, and Van Haltren( another one with consistency but not the best at postion). Dickey Pearce,well, I still need to decide how much confidence I can muster in competition levels of the 50's and early 60's. With my current level of confidence, he's 20th.
   271. David Concepcion de la Desviacion Estandar (Dan R) Posted: May 31, 2004 at 04:22 PM (#652202)
When is the actual 1927 ballot going up?
   272. Daryn Posted: May 31, 2004 at 05:07 PM (#652221)
It is only Canadians on the site today -- we'll have to wait for after the fireworks to get our ballots up.
   273. Devin has a deep burning passion for fuzzy socks Posted: May 31, 2004 at 06:05 PM (#652266)
I swear, I hadn't looked at the discussion thread for awhile when I did this. This is similar to what I did with Catcher OPS+ a while back, except in this case it's first basemen and Win Shares. I averaged the Win Shares for starting 1Bmen for each season (generally the first player listed in the Win Shares book for each team), then found how much each player differed from the average and added them up (and adjusted for games played). The time period I looked at was 1888-1932, so I'd have all of Beckley and Sisler's careers. Here's some lists, with a minimum of 3 years required (75 players qualified).

Total "WSAA"
1. Gehrig 138.18
2. Brouthers 82.58
3. Chance 76.28
4. Connor 72.79
5. Beckley 67.82
6. Konetchy 66.71
7. Terry 64.3
8. Foxx 60.12
9. Sisler 58.17
10. H. Davis 56.9
(14. Anson 44.22, 17. Merkle 23.4, 29. Chase 8.71, 32. Pipp 5.84, 33. G. Kelly 4.75)

Average
1. Gehrig 17.27
2. Foxx 15.03
3. Brouthers 10.32
4. Terry 9.19
5. Chance 8.48
6. Foutz 8.45
7. Connor 8.09
8. Bottomley 5.01
9. T. Jordan 4.52
10. Konetchy 4.45
(11. Anson 4.42, 12 Davis 4.38, 13 Sisler 4.16, 15. Beckley 3.57)

And this is arbitrary, but I defined a "Positive year" as a year more than 2.5 WS above average, and looked just at those.

Total in Positive Years (#of PYs)
1. Gehrig 139.81 (7)
2. Sisler 83.99 (7)
3. Brouthers 83.02 (7)
4. Konetchy 78.28 (10)
5. Chance 74.33 (6)
6. Beckley 74.23 (11)
7. Connor 69.87 (7)
8. Fournier 63.46 (6)
9. Terry 60.47 (6)
10. Davis 60.47 (9)

"Positive Year" Average
1. Gehrig 19.97
2. Foxx 15.03
3. Chance 12.39
4. Sisler 12
5. Brouthers 11.86
6. Foutz 11.74
7. J. Reilly 10.72
8. Fournier 10.58
9. Terry 10.49
10. E. Morgan 10.31
(11. Connor 9.98, 21. Konetchy 7.83, 22. Anson 7.77, 27. Beckley 6.75, 28. Davis 6.72)
   274. Devin has a deep burning passion for fuzzy socks Posted: May 31, 2004 at 06:48 PM (#652322)
Some notes on all that (and I could use some advice on some of this)

1. The thing that bugs me the most is there are some seasons where a team's 1Bman has only 1 or 2 Win Shares, and I can't help but think that unnaturally skews the average for that year, because that's below replacement level. I'm just not sure what the default minimum should be.

2. Obviously, my definition of "positive years" is completely arbitrary and means whatever you want it to mean.

3. I adjusted for games played after finding the difference, even though I know it's the wrong way to go about it. IOW, in 1888 Cap Anson had 29 Win Shares. The league average was 15.81, so Cap has 13.19, then I multiply that by the "GP Factor" to get 15.39 in a 154-game equivalent. I know that's the wrong way to do it (I should adjust everyone's Win Shares before I get the average), but it's also the easier way, and I don't think the distortions make a significant difference.

4. Just to double check, the Games Played, as far as I can tell, went like this:
1888-1891, 1893-1897: 132
1892, 1898-99: 150
1900-1903: 140
If any of that's wrong, please tell me.

5. For the early years, I included all 2 or 3 leagues and didn't adjust for league strength (again, for simplicity), which, if it does anything, will penalize Beckley by having the average higher than it should be, but I doubt it would be a dramatic difference. I left out the Federal League except for Chase and Konetchy. I didn't adjust the numbers for 1915 because I wasn't sure of the factor. (Chase's 1914 FL numbers I arbitrarily reduced)

6. I added Harry Davis' 1896 and 1898, because he was playing significant time but just splitting it between teams, and didn't rank as anyone's #1. This actually hurt his ranking because he didn't have great years. I don't think there are any similar questions around any of the other top candidates.

7. Obviously, the numbers for Gehrig, Foxx, Anson, Connor and Brouthers only cover parts of their careers. Bill Terry's gets most but not all.

8. The trendline for the adjusted average starts at 14.3 and rises to 16.3, which doesn't seem unreasonable given the rise in offense in the 1920s. The actual average, however, drops off sharply in 1893 and stays in the 11-13 range until 1901 (except a spike in 1897). Then it stays in the 14-15 range with occasional spikes until the 1920s, when it moves into th 16-17 range. That does tend to indicate Beckley was the best of a weak period.

9. What's it all mean? Obviously none of the candidate between the ABC guys and Gehrig are great candidates, but we knew that. Peak voters might have a decent argument with Chance. Beckley and Konetchy, OTOH, are more or less career guys, but this doesn't look like a great accumulation. They can't match what Connor did in his age 31-40 seasons, or Brouthers in his age 30-38. You don't have to be those guys to make the HoM, but it'd be nice to be 50 or 60% of them.
   275. Devin has a deep burning passion for fuzzy socks Posted: May 31, 2004 at 06:52 PM (#652331)
Oops, meant to put this in just for fun. The worst 1Bmen by this measure:

71. Charlie Grimm -34.26
72. George Stovall -39.75
73. Doc Johnston -43.04
74. Walter Holke -43.88
75. Phil Todt -54.26 (exhibit #1 in Why the Red Sox Sucked in the '20s)
   276. John (You Can Call Me Grandma) Murphy Posted: May 31, 2004 at 07:21 PM (#652400)
Obviously none of the candidate between the ABC guys and Gehrig are great candidates, but we knew that.

I agree. Konetchy will be the top first baseman on my ballot this year at #9. That doesn't bode too well for him (or the rest of the available candidates).
   277. Jeff M Posted: June 01, 2004 at 01:22 AM (#652759)
C....SS...3B.2B...1B..CF.LF.RF

What's the authority for this? More specifically:

1. What authority that 1b was tougher than CF? I understand that 1b was tougher than it is now, and I understand that it may have been tougher than LF or RF at some point (very early), but better than CF? I've never heard that assertion.

2. If 1b was tougher than CF (or any other OF position), what evidence that it continued to be that way until 1920? I would think 1b sunk to its normal role a lot earlier than 1920.

3. If 1b was tougher than all the OF positions until 1920, why do we keep talking about defense of Sheckard, Thompson, Magee, Hill, Browning, etc. as a significant factor in their rankings? I think we focus too much on OF defense already, but if they were lowest on the spectrum all the way to 1920, then even more so.

4. I also haven't seen the assertion that RF was lower on the spectrum than LF. I can see that argument very early when nobody hit lefthanded (and maybe fewer opposite field hitters), but I'd be very surprised if that carried to 1920. Among the other factors, RF is generally tougher because you have to have a better arm to play out there.

5. I thought 2b passed 3b on the spectrum in the 1930s.

I ask these questions because (A) the assertions surprised me and I want to know more about it and (B) we still have a number of fairly new voters who assume most of what we post is correct.
   278. John (You Can Call Me Grandma) Murphy Posted: June 01, 2004 at 01:51 AM (#652772)
C....SS...3B.2B...1B..CF.LF.RF

What's the authority for this?


That certainly piqued my interest when I saw it, too.
   279. KJOK Posted: June 01, 2004 at 02:39 AM (#652794)
C....SS...3B.2B...1B..CF.LF.RF
We actually had a defensive spectrum thread at one time where we discussed this, but I think it was lost in the transition to the new site.

1. What authority that 1b was tougher than CF? I understand that 1b was tougher than it is now, and I understand that it may have been tougher than LF or RF at some point (very early), but better than CF? I've never heard that assertion.

IIRC, there are two parts to this argument. One is the strictly defensive argument that there were fewer balls hit to the OF in the early days of baseball, many more bunt plays and throws to the bases for 1st basemen, and with more errors on the infield, more opportunities for a good 1st baseman to prevent errors from other infielder throws.

The 2nd part is to infer the defensive spectrum by looking at the offense at the postion:
1st base 1876-1919: .333 OBP, .375 SLG
CF 1876-1919: .346 OBP, .371 SLG

2. If 1b was tougher than CF (or any other OF position), what evidence that it continued to be that way until 1920? I would think 1b sunk to its normal role a lot earlier than 1920.

1st base 1910-1919: .332 OBP, .368 SLG
CF 1910-1919: .350 OBP, .377 SLG

Looks like it was still holding that first base was still a "tougher" defensive position.

3. If 1b was tougher than all the OF positions until 1920, why do we keep talking about defense of Sheckard, Thompson, Magee, Hill, Browning, etc. as a significant factor in their rankings? I think we focus too much on OF defense already, but if they were lowest on the spectrum all the way to 1920, then even more so. Exactly.

4. I also haven't seen the assertion that RF was lower on the spectrum than LF. I can see that argument very early when nobody hit lefthanded (and maybe fewer opposite field hitters), but I'd be very surprised if that carried to 1920. Among the other factors, RF is generally tougher because you have to have a better arm to play out there.

Not only were there not as many LH hitters, but the ballparks were primarily rectangularly shaped, with very large LF areas (400 ft plus down the lines in some cases) and very small RF areas (250-300 ft to the lines) so LF'ers were actually the ones that had to have strong arms and good range. Teams often played players with
"CF type" skills such as Duffy and Sheckard in LF, while put their less talented outfielders in RF.

5. I thought 2b passed 3b on the spectrum in the 1930s.

Yes, which is exactly consistent with the spectrum above which shows 3B still ahead of 2B on the defensive spectrum.
   280. KJOK Posted: June 01, 2004 at 02:47 AM (#652799)
Trying to anticipate the next question, it looks like it was between 1922 and 1928 that CF moved to the left of 1B on the defensive spectrum.

YEARPosOPSPosOPSDiff
19191B0.729CF0.742-0.01
19201B0.736CF0.801-0.07
19211B0.787CF0.811-0.02
19221B0.811CF0.8080.00
19231B0.806CF0.7890.02
19241B0.806CF0.780.03
19251B0.847CF0.8390.01
19261B0.795CF0.7980.00
19271B0.808CF0.7960.01
19281B0.832CF0.7610.07
19291B0.873CF0.8210.05
19301B0.872CF0.7940.08
19311B0.826CF0.7750.05
19321B0.826CF0.770.06
19331B0.769CF0.7620.01
19341B0.867CF0.7870.08
19351B0.812CF0.7730.04
19361B0.859CF0.7860.07
19371B0.834CF0.770.06
19381B0.86CF0.7750.09
19391B0.846CF0.7730.07
19401B0.813CF0.7820.03
   281. KJOK Posted: June 01, 2004 at 02:53 AM (#652802)
Oops, all my formatting was wiped out:

YEAR Pos OPS Pos OPS Diff
1919 1B 0.729 CF 0.742 -0.01
1920 1B 0.736 CF 0.801 -0.07
1921 1B 0.787 CF 0.811 -0.02
1922 1B 0.811 CF 0.808 0.00
1923 1B 0.806 CF 0.789 0.02
1924 1B 0.806 CF 0.780 0.03
1925 1B 0.847 CF 0.839 0.01
1926 1B 0.795 CF 0.798 0.00
1927 1B 0.808 CF 0.796 0.01
1928 1B 0.832 CF 0.761 0.07
1929 1B 0.873 CF 0.821 0.05
1930 1B 0.872 CF 0.794 0.08
1931 1B 0.826 CF 0.775 0.05
1932 1B 0.826 CF 0.770 0.06
1933 1B 0.769 CF 0.762 0.01
1934 1B 0.867 CF 0.787 0.08
1935 1B 0.812 CF 0.773 0.04
1936 1B 0.859 CF 0.786 0.07
1937 1B 0.834 CF 0.770 0.06
1938 1B 0.860 CF 0.775 0.09
1939 1B 0.846 CF 0.773 0.07
1940 1B 0.813 CF 0.782 0.03
   282. PhillyBooster Posted: June 01, 2004 at 12:59 PM (#652891)
It is possible, of course, that offense is a trailing indicator -- that centerfield had become more important years earlier but that it took a decade or so for managers to get out of the conventional wisdom rut.

The Phillies' continued playing of Travis Lee as a great defensive first baseman in 2002 shows that the problem remains today.
   283. karlmagnus Posted: June 01, 2004 at 03:06 PM (#653000)
Given KJOK's very helpful figures, doesn't it make sense to believe that CF became more important than 1B over the period 1911-20, as first the level of offense increased, and then the HR numbers went up. It would seem inconceivable that 1B wasn't more importyant than CF during the dead ball era, or indeed in the 90s, when the offense was based on the running game and strategy.

Doesn't this mean people have been WAY underrating Beckley, who should be regarded as closer to a 2B or 3B than an OF, given his time period. He had almost 1000 hits more than his contemporary Jimmy Collins, at an OPS+ of 125 compared to 113.

For Konetchy, the case is much less clear; his prime was in the 10s, after the 1911 offesnive explosion.
   284. jimd Posted: June 01, 2004 at 03:39 PM (#653043)
Average OPS by position (% + or - from MLB avg)
(Reprinted from some old thread.)

Decad 1B LF RF CF 3B 2B Ca SS Pit
1870's +1 +4 -1 +4 +2 +2 +0 +1 -13
1880's 13 +6 +1 +5 +1 -1 -7 -2 -17
1890's +6 +9 +7 +7 +0 -2 -6 -2 -22
1900's +6 10 +9 +8 +0 +2 -9 -1 -29
1910's +6 +7 +9 10 +1 +1 -7 -4 -31
1920's +9 10 10 +8 -3 +1 -4 -7 -32
1930's 13 +8 10 +5 -1 -3 -3 -4 -36
1940's +8 11 +9 +7 +2 -3 -4 -4 -37
1950's +9 10 +7 +7 +4 -3 -1 -5 -40
1960's 11 +9 11 +7 +4 -5 -3 -6 -46
1970's 10 +8 +8 +5 +3 -5 -2 -11-45
1980's +8 +6 +6 +2 +3 -4 -4 -8 -48
1990's +9 +4 +6 +1 +1 -3 -4 -7 -50

Mean.. +9 +8 +7 +6 +1 -2 -4 -5 -36
   285. John (You Can Call Me Grandma) Murphy Posted: June 01, 2004 at 03:39 PM (#653044)
For Konetchy, the case is much less clear; his prime was in the 10s, after the 1911 offesnive explosion.

KJOK stated that first base was tougher than CF during the teens also, so I'm confused by your statement. The twenties is when things started to change between the two positions in terms of offensive production.

I do agree with you that first base during the Deadball Era has been shortchanged by some of the voters here
   286. OCF Posted: June 01, 2004 at 04:29 PM (#653131)
One thing to be cautious of about the numbers jimd quoted in #284 is how sensitive they might be to outlier individuals. Yes, CF is listed as "+10" for the 1910's. During that decade, Ty Cobb averaged an OPS+ of 192 and Tris Speaker averaged 163. They accounted for 1/8 of the regular CF'ers, and at least 1/10 of the playing time. If you were to replace them with CF's averaging 130 OPS+, the average for the position would drop by about 5 points - enough to put it behind the 1B'men.

The CF position came up with Cobb and Speaker for the teens - and the 1B position wouldn't come up with Gehrig until the middle of the 20's. I don't see that these outliers are by themselves evidence for an inverted defensive spectrum.
   287. Max Parkinson Posted: June 01, 2004 at 04:31 PM (#653135)
One explanation for the better hitting out of CFers in the '10s is simply personnel.

Let's look at the top 16 CFers for that decade, in terms of games played - a simple proxy for the starters of the 16 ML teams (the Fed only lasting 2 years and all...) The list includes:

Speaker
Cobb
Roush
Kauff
Milan
Carey

These guys are pretty good hitters. That doesn't mean that they weren't good defenders. It's a blip, a coincidence of strength at a position. It happens, especially in "small sample sizes" like a 16-team league. The fact that the group of CFers (dominated by these guys, and other good ones like Paskert and Strunk) out hit the group of first basemen is not sufficient evidence to say that 1B was a much tougher position to play.
   288. Max Parkinson Posted: June 01, 2004 at 04:33 PM (#653137)
I mean, what OCF said...
   289. DavidFoss Posted: June 01, 2004 at 04:34 PM (#653138)
KJOK stated that first base was tougher than CF during the teens also, so I'm confused by your statement. The twenties is when things started to change between the two positions in terms of offensive production.

There was a reprieve from the Dead Ball era in the early teens. 1911 and 1912 cannot be really be considered dead ball years. The dead ball era returned later in the decade, though. Scoring levels were quite low again from 1914-1919.

How much of this defensive spectrum switch is due to superstar players? Are Cobb & Speaker single-handedly raising the bar for CF's in the teens? When CF offense levels rise above both LF/RF, that tells me that that might be happening.

Do the spectrum values change when you drop the top X players at each position?
   290. DavidFoss Posted: June 01, 2004 at 04:35 PM (#653142)
Oops... yeah... what OCF said... his post wasn't there when I clicked send.
   291. DavidFoss Posted: June 01, 2004 at 04:44 PM (#653152)
Sure McGraw had a short career, but he could have 3 NEGATIVE 30 RCAP years and he'd still be ahead of Home Run Baker in roughly the same # of PA's.

McGraw's leagues: .291/.358/.389
Bakers' leauges: .266/.334/.349

Do your RCAP numbers need to be era adjusted at all? I know its a difference stat, so its re-zero-ed by era. But does the scale need to be adjusted as well?

I ask this because both players have matching 135 OPS+'s. I knew of OPS's bias towards SLG, but recent posts have explained to me that OPS+ doesn't have the same bias.
   292. jimd Posted: June 01, 2004 at 05:11 PM (#653185)
One thing to be cautious of about the numbers jimd quoted in #284 is how sensitive they might be to outlier individuals.
In general, I'm not disagreeing with OCF's caution.

If you were to replace them with CF's averaging 130 OPS+, the average for the position would drop by about 5 points -

But there's a problem with the magnitude of his calculations. My table uses OPS not OPS+. OPS+ is calculated as ((OBP/LgOBP) + (SLG/LgSLG) - 1) * 100. An OPS+ of 192 represents somebody who is about 45% above average. So the estimate of 5 points of change is more like 2 or 3, depending on which way the fraction rounds.

Also, the effect is longer than just the 1910's. CF'ers outhit 1bmen in the 1890's, the 1900's, the 1910's, before reversing in the 1920's (which includes a half-decade of Gehrig). Cobb and Speaker didn't impact the two decades prior; they just accentuated a pre-existing condition. The argument that they moved CF ahead of LF and RF is one that I'd buy into.

Centerfield as a "glove" position rivaling 3b is a product of the last 20-30 years; before that it was just the best glove of the 3 outfielders and CF was expected to carry a real bat.
   293. John (You Can Call Me Grandma) Murphy Posted: June 01, 2004 at 05:38 PM (#653230)
CF'ers outhit 1bmen in the 1890's, the 1900's, the 1910's

I'm not going to buy the next argument that will be presented that first basemen just forgot to hit for thirty something years. Was it you Jim that came up with the differences in fielding responsibility for the position throughout the years?
   294. KJOK Posted: June 01, 2004 at 08:54 PM (#653602)
McGraw's leagues: .291/.358/.389
Bakers' leauges: .266/.334/.349

Do your RCAP numbers need to be era adjusted at all?


They are park and era adjusted for RUNS. The only further adjustments would be to convert RUNS to WINS, as it takes slightly fewer RUNS to equal a WIN in Baker's leagues than in McGraw's.
   295. Jim Sp Posted: June 01, 2004 at 09:07 PM (#653622)
An OPS+ of 192 represents somebody who is about 45% above average.

OPS+ of 192 represents a player whose OBP and SLG are each 45% above average, which is a player who creates twice as many runs as the average player.
   296. The definitely immoral Eric Enders Posted: June 01, 2004 at 09:40 PM (#653654)
I've been lurking in these HOM discussions for a while now, and I just wanted to take a moment to thank all you guys for your wonderfully informative posts. I wish I had enough time to become a full-fledged HOM participant. Keep up the good work.
   297. John (You Can Call Me Grandma) Murphy Posted: June 01, 2004 at 09:52 PM (#653665)
Keep up the good work.

Thanks, Eric!

Since you're an authority on the Negro Leagues, feel free to correct us if we're wrong about something posted here. Most of us (including myself) are novices for that era of baseball.
   298. jimd Posted: June 01, 2004 at 11:17 PM (#653754)
OPS+ of 192 represents a player whose OBP and SLG are each 45% above average, which is a player who creates twice as many runs as the average player.

It represents many things. A player whose SLG is 92% above average and OBP is average, a player which is the reverse, and many players which are many (additive) mixtures of the two, including 46% of each. The number of RC is approximately double, ranging from 192% to 213% depending on the balance.

In any case it has little to do with (OPS/LgOPS), which is what this table was calculated with. A +13% from this table is roughly equivalent to a 126 OPS+, a -5% to a 90 OPS+. So the 5 point change in OPS+ caused by Cobb/Speaker would correspond to a 2 or 3 point change in the metric of the table.

Original post and discussion: posts 187-199
   299. OCF Posted: June 01, 2004 at 11:41 PM (#653784)
Ah, jimd's link was to the very thread and very subject upon which I first posted in the Hall of Merit. Scroll down a few entries to find me disavowing any intent to become a voter.

I'll concede jimd's central point - even though the outliers matter, even without them the deadball centerfielders did outhit the first basemen. I'm not so sure it follows that first basemen deserve a bonus for it, however. In this era, base stealing was an obligatory part of offense and triples were an important part of what we would mean by power. As is few other times, players were selected for speed because of what speed meant on offense. The best hitters were fast. Where do you use fast guys on defense? The speed matters more in centerfield than it does at first base. Yes, all those bunts did increase the defensive burdens on first basemen, but the old (and probably slower) player-managers often played first. It's at least possible that the centerfielders were just, on average, better players than the first basemen.
   300. EricC Posted: June 02, 2004 at 12:52 AM (#653953)
[in reponse to Dan Rosenheck #101]

I will try to evaluate 30 players per election..... I take the player's five best years...... and WS adjusted for season length..... I'd love to hear thoughts/criticisms from the statisticians among us on this method so as to improve it.

Welcome, Dan. I use standard deviations in my rating system too, so I'm aware of some pitfalls.

I see a potential problem if you are only evaluating 30 players per election- this is nowhere near enough to get good statistics and leaves open the possibility that you are overlooking candidates that would otherwise score as ballot-worthy in your system. Is it possible for you to automate your procedure so that you can look at all eligible players? If not, you should expand your consideration set as large as is convenient. Don't forget to include the players that have already been elected to the HoM, because they help to set the bar.

A second problem is the treatment of pre-1893 pitchers in comprehensive rating systems such as WS, which is compounded if you are adjusting WS for season length by prorating it. Consider, for example, Jim Devlin. In 1876, he earned 53 WS in 69 team games. In 1877, he earned 60 WS in 61 team games. If you prorate to 162 team games per season, you get 124 + 159 = 283 WS in two years, more than, say, Randy Johnson earned in 16 years through 2003. It would be absurd to argue that Jim Devlin was a greater pitcher than Randy Johnson. How you treat pre-1893 pitchers is up to you to figure out, but you might want to try your system on the following players that had at least some significant pre-1893 pitching value: (HoMers starred) Tommy Bond, Charlie Buffinton, Bob Caruthers, John Clarkson*, Larry Corcoran, Candy Cummings, Jim Devlin, Charlie Ferguson, Dave Foutz, Pud Galvin*, Guy Hecker, Bill Hutchison, Tim Keefe*, Matt Kilroy, Silver King, Terry Larkin, Bobby Mathews, Jim McCormick, Ed Morris, Tony Mullane, Kid Nichols*, Charlie Radbourn*, Toad Ramsey, Amos Rusie*, Elmer E. Smith, Al Spalding*, Jack Stivetts, Monte Ward*, Mickey Welch, Gus Weyhing, Will White, Jim Whitney, and Cy Young*.

Finally, I'd like to point out that taking the player's 5 best years is an arbitrary choice, and will cause you to overrate those players who happen to fall in the "sweet spot" of having exactly 5 good years.
Page 3 of 4 pages  < 1 2 3 4 > 

You must be Registered and Logged In to post comments.

 

 

<< Back to main

BBTF Partner

Support BBTF

donate

Thanks to
BDC
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Hot Topics

Syndicate

Page rendered in 1.0840 seconds
43 querie(s) executed