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, August 21, 2006

1984 Ballot Discussion

1984 (September 4)—elect 2
WS W3 Rookie Name-Pos (Died)

269 75.3 1959 Ron Fairly-1B/RF
261 68.5 1962 Jim Fregosi-SS
190 78.0 1963 Wilbur Wood-P
189 55.4 1963 Bob Bailey-3B
171 56.3 1966 Davey Johnson-2B
127 44.6 1964 Jerry Grote-C
132 38.1 1966 Tito Fuentes-2B
122 40.8 1965 Jim Lonborg-P
116 41.4 1965 Nelson Briles-P (2005)
122 32.6 1965 Mickey Stanley-CF
110 31.6 1969 Wayne Garrett-3B
102 32.8 1970 Don Gullett-P
104 24.8 1965 Sandy Alomar-2B

Players Passing Away in 1983
HoMers
Age Elected

81 1961 Earl Averill-CF

Candidates
Age Eligible

88 1925 George Halas-RF/NFL HOF
85 1942 Charlie Grimm-1B
82 1943 Fred Schulte-CF
76 1946 Jackie Hayes-2B
76 1954 Hilton Smith-P
74 1959 Dutch Leonard-P
70 1953 Chet Laabs-LF/RF
67 1955 Stan Spence-CF
60 1967 Del Rice-C
58 1966 Willie Jones-3B
58 1969 Vic Wertz-RF/1B
39 1982 Carl Morton-P

Thanks, Dan!

John (You Can Call Me Grandma) Murphy Posted: August 21, 2006 at 02:20 AM | 216 comment(s) Login to Bookmark
  Related News:

Reader Comments and Retorts

Go to end of page

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

Page 2 of 3 pages  < 1 2 3 > 
   101. DL from MN Posted: August 24, 2006 at 10:17 PM (#2155673)
I gave Easter Tony Perez' bat and Dave Winfield's glove, subtracted 2 seasons due to Easter's injuries and came up with a player equivalent to Gil Hodges - around 60th on the ballot. If I give Easter Winfield's bat and Perez' glove he's at the top of the ballot. Split the difference and he's 9th, just behind Quincy Trouppe.

Trying again using Easter's numbers after age 33 but Winfield or Cepeda before age 32:

High 3rd (Bob Johnson)
Low 44th (Ben Taylor)
Avg 10th (Jim Wynn)

He may have been an average hitting 1B after he left the majors but he was described as "immobile" and "limping".

I'm not ready to slot him ahead of Cepeda just yet but he's got a good argument. After working with a composite of Easter, Hodges, Cepeda, Winfield and Perez I think I've got something reasonable. This year he will place 18th on my ballot.
   102. Howie Menckel Posted: August 25, 2006 at 12:44 AM (#2155838)
Here's a fun offshoot:

HOMers in the 1950s and 1960s so far

NL HOMers 1950-1960s, * is part-time (10+ G, fewer than half in field or 1 IP per G/35 G)
1950 (10) - Slaughter, Reese, Musial, Spahn, JRobinson, Snider, Campanella, Ashburn, Roberts, Irvin
1951 (11) - Slaughter, Reese, Musial, Spahn, JRobinson, Snider, Campanella, Ashburn, Roberts, Irvin, Mays
1952 (13) - Slaughter, Reese, Musial, Spahn, JRobinson, Snider, Campanella, Ashburn, Roberts, Irvin*, Mays*, Mathews, Wilhelm
1953 (13) - Slaughter, Reese, Musial, Spahn, JRobinson, Snider, Campanella, Ashburn, Roberts, Irvin, Mathews, Wilhelm, Banks*
1954 (14) - Reese, Musial, Spahn, JRobinson, Snider, Campanella, Ashburn, Roberts, Irvin, Mays, Mathews, Wilhelm, Banks, Aaron
1955 (16) - Reese, Musial, Spahn, JRobinson, Snider, Campanella, Ashburn, Roberts, Irvin*, Mays, Mathews, Wilhelm, Banks, Aaron, Koufax*, Clemente
1956 (18) - Reese, Musial, Spahn, JRobinson, Snider, Campanella, Ashburn, Roberts, Irvin, Mays, Mathews, Wilhelm, Banks, Aaron, Koufax*, Clemente, Drysdale*, FRobinson
1957 (16) - Reese, Musial, Spahn, Snider, Campanella, Ashburn, Roberts, Mays, Mathews, Wilhelm, Banks, Aaron, Koufax*, Clemente, Drysdale*, FRobinson
1958 (14) - Reese*, Musial, Spahn, Snider, Ashburn, Roberts, Mays, Mathews, Banks, Aaron, Koufax, Clemente, Drysdale*, FRobinson
1959 (15) - Musial, Spahn, Snider, Ashburn, Roberts, Mays, Mathews, Banks, Aaron, Koufax, Clemente, Drysdale, FRobinson, Gibson*, BWilliams*

1960 (17) - Musial, Spahn, Snider*, Ashburn, Roberts, Mays, Mathews, Banks, Aaron, Koufax, Clemente, Drysdale, FRobinson, Gibson*, BWilliams*, Santo, Marichal*
1961 (17) - Musial, Spahn, Snider*, Ashburn, Roberts*, Mays, Mathews, Banks, Aaron, Koufax, Clemente, Drysdale, FRobinson, Gibson, BWilliams, Santo, Marichal
1962 (17) - Musial, Spahn, Snider*, Ashburn, Roberts, Mays, Mathews, Banks, Aaron, Koufax, Clemente, Drysdale, FRobinson, Gibson, BWilliams, Santo, Marichal
1963 (16) - Musial, Spahn, Snider, Roberts, Mays, Mathews, Banks, Aaron, Koufax, Clemente, Drysdale, FRobinson, Gibson, BWiliams, Santo, Marichal, Allen*
1964 (17) - Spahn, Snider*, Roberts, Mays, Mathews, Banks, Aaron, Koufax, Clemente, Bunning, Drysdale, FRobinson, Gibson, BWilliams, Santo, Marichal, Allen
1965 (16) - Spahn, Roberts, Mays, Mathews, Banks, Aaron, Koufax, Clemente, Bunning, Drysdale, FRobinson, Gibson, BWilliams, Santo, Marichal, Allen
1966 (14) - Roberts*, Mays, Mathews, Banks, Aaron, Koufax, Clemente, Bunning, Drysdale, Gibson, BWilliams, Santo, Marichal, Allen
1967 (11.8) - Mays, Mathews*, Banks, Aaron, Clemente, Bunning, Drysdale, Gibson, BWilliams, Santo, Marichal, Allen
1968 (11) - Mays, Banks, Aaron, Clemente, Bunning, Drysdale, Gibson, BWilliams, Santo, Marichal, Allen
1969 (11) - Mays, Banks, Aaron, Clemente, Bunning, Drysdale*, Gibson, BWilliams, Santo, Marichal, Allen


AL HOMers 1950-1960s, * is part-time (10+ G, fewer than half in field or 1 IP per G/35 G)
1950 (14) - Appling*, DiMaggio, Feller, Mize, Gordon, TWilliams, Boudreau, Doerr, Newhouser, Wynn, Lemon, Doby, Berra, Ford*
1951 (13) - Paige*, DiMaggio, Feller, Mize, TWilliams, Boudreau, Doerr, Newhouser*, Wynn, Lemon, Doby, Berra, Mantle
1952 (9) - Paige, Feller, Mize*, Newhouser, Wynn, Lemon, Doby, Berra, Mantle
1953 (11) - Paige, Feller, Mize*, TWilliams*, Wynn, Lemon, Doby, Berra, Ford, Mantle, Kaline*
1954 (11) - Feller*, TWilliams, Newhouser*, Wynn, Lemon, Doby, Berra, Ford, Mantle, Kaline, Slaughter*
1955 (12) - Feller*, TWilliams, Wynn, Lemon, Doby, Berra, Ford, Mantle, Slaughter, Bunning*, Killebrew*
1956 (11) - Feller*, TWilliams, Wynn, Lemon, Doby, Berra, Ford, Mantle, Kaline, Slaughter, Bunning*, Killebrew*
1957 (10) - TWilliams, Wynn, Lemon*, Doby, Berra, Ford*, Mantle, Kaline, Slaughter*, Bunning
1958 (12) - TWilliams, Wynn, Lemon*, Doby, Berra, Ford, Mantle, Kaline, Slaughter*, Wilhelm, Bunning, Killebrew*
1959 (10.8) - TWilliams, Wynn, Doby*, Berra, Ford, Mantle, Kaline, Slaughter**, Wilhelm, Bunning, Killebrew

1960 (9) - TWilliams, Wynn, Berra, Ford, Mantle, Kaline, Wilhelm, Bunning, Killebrew
1961 (8) - Wynn*, Berra, Ford, Mantle, Kaline, Wilhelm, Bunning, Killebrew
1962 (7) - Wynn, Berra, Ford, Mantle, Kaline, Wilhelm, Bunning, Killebrew
1963 (8) - Wynn*, Berra, Ford, Mantle*, Kaline, Wilhelm, Bunning, Killebrew
1964 (5) - Ford, Mantle, Kaline, Wilhelm, Killebrew
1965 (5) - Ford, Mantle, Kaline, Wilhelm, Killebrew
1966 (6) - Ford*, Mantle, Kaline, Wilhelm, Killebrew, FRobinson
1967 (5.3) - Mantle, Kaline, Mathews*, Wilhelm, Killebrew, FRobinson
1968 (6) - Mantle, Kaline, Mathews*, Wilhelm, Killebrew, FRobinson
1969 (4) - Kaline, Wilhelm*, Killebrew, FRobinson
   103. Juan V Posted: August 25, 2006 at 01:11 AM (#2155874)
Regarding Moore, I have him roughly comparable with Bancroft (on positioning in my rankings, not that they are similar players), which would be 20-ish in my ballot and fourth amongst the middle infield guys (the top 3 are in my ballot, as stated above).


Scratch this. I found a little mistake on my spreadsheet regarding Moore, and he´s higher than that. Not quite in the ballot, as I still think Fregosi has a slightly better case, but that lifts him to the 16-20 range.
   104. Howie Menckel Posted: August 25, 2006 at 03:06 PM (#2156448)
HOMers Through The Years....

These figures include all players appearing in at least 10 games in a season (or estimated equivalent).

Notice a steady uptick from 1856 to 1889.
Then a "bubble" from 1890-93, then back to 1880s levels.
A minor increase follows from 1908-21.
The explosion follows from 1922 until 1942, when WW II makes its mark.
Totals sink briefly to 1880s levels again, followed by a brief post-war burst - and then a sag again.

It's VERY clear that the 'cause' of the explosion is the willingness to induct 1920s-30s Negro Leagues players while not reducing the number of AL/NL electees in those years. We've more or less been electing the same number of major leaguers from the 1890s to the 1960s (mid- to high-20s), aside from many HOMers missing in WW II (with a slight, steady increase over time, given that in the early years of that span it took the addition of a handful of Negro Leaguers annually to get the tally up to par).

Basically, this plays to HOM founder Joe Dimino's 'a pennant is a pennant is a pennant' theory. Note that there tended to be about 13 to 16 Negro League HOMers playing annually in the 1920s and 1930s, and that those numbers match the entire amount of 'extra' players from that era. I'm pretty agnostic on that issue, but it's clear what has happened.

The numbers:

19th CENTURY
1856-59 - 1
1860-65 - 2 (3 in 1864)
1866-67 - 4
1868-71- 6/8/9/10
1872-78 - 11 or 12
1879-80 - 16/17
1881-84 - 20 to 22
1885-89 - 23 to 25
1890-93 - 29/32/30/27
1894-99 - 23 (24 in 1895)
1900---- - 22
Waddell, Childs, Beckley, Duffy, C Jones, Van Haltren, and Browning are current top 25 vote-getters who would add to the mix, mostly in the 1890s.

20th CENTURY
1901-07 - 22 to 25
1908-15 - 26 or 27 (28 in 1913)
1916------ 33
1917-21 - 25 to 27 (23 in 1918)
1922-25 - 31/33/38/42
1926-32 - 44 to 46 (42 in 1929)
1933-41 - 38 to 42
1942-45 - 36/26/19/20
1946-48 - 30 to 31
1949-53 - 24 (22 in 1952)
1954-56 - 28 or 29
1957-60 - 26
1961-63 - 24 or 25
1964-69 - 22/21/20/17/17/15
Mendez, Moore, and Redding are current top-25 vote-getters who would add to the 1901-19 mix, along with brief adds in the early 1900s from some of the 1800s guys.
   105. DanG Posted: August 25, 2006 at 03:13 PM (#2156466)
Howie:
Could you repeat the chart in #104 totaling every five year period? That is 1856-60, 1857-61, 1858-62, etc. I think it would make the location of the "bump" much clearer.
   106. DavidFoss Posted: August 25, 2006 at 03:26 PM (#2156483)
Thanks Howie!

McCovey (1959) is the only new candidate left from the 1950s (Kaat's 1959 cup of coffee is not enough to register). Brooksie (1957-59 plus cups of coffee) appears destined for induction.

1926-32 and 1933-41 are a quite high, but otherwise we're not doing too bad on era balance. I'm OK with 1894-07 being a bit lower. The AL expansion took some time to take hold and the spike from 1890-93 looks to be an anomaly of players from different eras overlapping. Short eras appear to be the 1910s and 1950s, but neither of those are too bad.
   107. yest Posted: August 25, 2006 at 03:27 PM (#2156484)
Howie I noticed a simaler problem in my pHoM about a "decade ago in my pHoM so I decided to see what my system would be like from 1947-80 if baseball was still under segragation to see if and how much I was overating 20's and 30's players so by taking out all black player's from the records (I only did offensive stats and not even close to as thourough a job as it deserved) so only white players would get black ink, gray ink, MVPs ext. and to see how my system would react to it but the thing that surprised me was there wasn't a single person who would have made my pHoM that oterwise wouldn't and the only noticable differance was borderliners became clear cut and clearly no brainers
   108. Joey Numbaz (Scruff) Posted: August 25, 2006 at 03:36 PM (#2156491)
"1926-32 and 1933-41 are a quite high, but otherwise we're not doing too bad on era balance. I'm OK with 1894-07 being a bit lower. The AL expansion took some time to take hold and the spike from 1890-93 looks to be an anomaly of players from different eras overlapping. Short eras appear to be the 1910s and 1950s, but neither of those are too bad."

1926-41 should be higher - Negro Leagues in full swing, and baseball hasn't had an expansion in 25-40 years. There aren't any other sports pulling significantly from the talent pool yet either. This was the "Golden Era" - The competition level didn't pass that of the late 30s until the mid-1980s IMO.

The dip in the early 50s is understandable, those are the 3 or 4 HoMers we've never heard of that were killed in WWII who would have been in their primes. Same for the smaller dip in 1957-60 (probably lost one or two in Korea).

The overall dip in the 1950s is most certainly caused by a combination of the wars and other sports like the NBA and NFL pulling talent. Those factors outweigh integration, IMO. I'm pretty confident we'll jump up in the 1960s.
   109. Joey Numbaz (Scruff) Posted: August 25, 2006 at 03:38 PM (#2156494)
"The competition level didn't pass that of the late 30s until the mid-1980s IMO."

BTW, when I say level, I mean average, per-team level. Of course there were more great players in the 70s than the 30s, but there weren't 50% more, and there were 50% more teams in the 1970s.
   110. DL from MN Posted: August 25, 2006 at 04:12 PM (#2156539)
It's getting crowded with sluggers with Cepeda, Kiner, Keller, Cravath, Frank Howard and now Luke Easter all between 16 and 26 on my ballot.
   111. Chris Fluit Posted: August 25, 2006 at 04:22 PM (#2156548)
104. Howie Menckely: It's VERY clear that the 'cause' of the explosion is the willingness to induct 1920s-30s Negro Leagues players while not reducing the number of AL/NL electees in those years. We've more or less been electing the same number of major leaguers from the 1890s to the 1960s (mid- to high-20s), aside from many HOMers missing in WW II (with a slight, steady increase over time, given that in the early years of that span it took the addition of a handful of Negro Leaguers annually to get the tally up to par).

Howie, your above chart does not prove this point that you're making. Maybe if you showed us the full chart (as you did for the 1950s and 1960s), or the 5-year chart that Dan G requests, it would become clear to us. But right now, you're making an unfounded assertion.

This is the Hall of Merit breakdown by decade (as based on the chart that Brent shows every once in a while)

1860s: 1
1870s: 9
1880s: 17
1890s: 14 + 1 NeL (Grant) = 15
1900s: 13 + 3 NeL (Foster, Hill, Johnson) = 16
1910s: 11 + 4 NeL = 15
1920s: 10 + 6 NeL = 16
1930s: 21 + 8 NeL = 29
1940s: 13 + 3 NeL = 16
1950s: 10/11 + 3/4 NeL (Campanella, Doby, Robinson and maybe Banks) = 14

By Brent's chart, it is clear that the Negro League players are not responsible for any huge jump. Negro League players were inducted at a slowly increasing rate of 1-3-4-6-8 per decade from the 1890s to the 1930s. It is the number of Major Leaguers inducted from the 1930s that creates that huge spike. Admittedly, Brent's chart is more simplistic than the one you presented. But without your full chart, I'm not willing to accept your assertions or conclusions. So I guess I'm echoing Dan G's request

Furthermore, the nature of the Negro Leagues insures that Negro League players would be over-represented in a year-by-year chart rather than one that focuses on decade of birth (as in sunnyday's chart) or on peak (as in Brent's chart). Because the replacement level was lower in the Negro Leagues than it was in the Majors, and because Negro League players were paid less than their white counterparts, Negro League players would often hang around for longer careers. Oscar Charleston played for 35 years. Cool Papa Bell for 28. Pete Hill 27. Satchel Paige 26. Pop Lloyd 25. Jud Wilson 23. Martin DiHigo 22. That means that a Negro League player is going to chart more often on a year-to-year analysis and yeah, you're going to get years of 14 to 16 Negro League players at a time. But that doesn't mean that we've elected a comparable number of Major League and Negro League players. Fewer players account for more entries in a year-by-year analysis. That gives you more entries, but not more players. The decade of birth and the decade of peak charts show that we haven't elected as many Negro League players as you seem to be claiming.

Admittedly, even those other charts do show a spike. Brent's chart shows 8 players with their careers centered in the 1930s, 2 more than any other decade. Sunnyday's chart shows 10 players born from 1900-1909, 5 more than any other decade. Those of us who support Negro League players and are pushing for their induction acknowledge there's a spike. But that's why we're promoting the players who played before the spike, or after it. We would want to see the spike evened out, by electing pre-Negro League players born from 1880-1899 and 1910-1929 (or those playing in the '10s and '20s, and those playing in the '40s and '50s). We want to be as fair to those eras as we were to players from the '30s. Telling us that we've already elected enough Negro League players from the '30s is not the same thing as telling us that we've already elected enough Negro League and pre-Negro League players.
   112. Chris Fluit Posted: August 25, 2006 at 04:25 PM (#2156551)
ps. Sorry about the "Menckely" in there. I honestly don't know how that "y" got in there. Must have slipped in when I was typing the colon. There was no insult intended.
   113. sunnyday2 Posted: August 25, 2006 at 04:34 PM (#2156556)
>The overall dip in the 1950s is most certainly caused by a combination of the wars and other sports like the NBA and NFL pulling talent. Those factors outweigh integration, IMO. I'm pretty confident we'll jump up in the 1960s.

>It's VERY clear that the 'cause' of the explosion is the willingness to induct 1920s-30s Negro Leagues players while not reducing the number of AL/NL electees in those years.

>>Howie, your above chart does not prove this point that you're making.

This is an important point, as it provides direction as to what (if any) remedial action might be considered.

Howie has posted charts before, or maybe it was somebody else, showing inductees by year from the AL, NL and NeLs. Howie is right. We have elected a fairly consistent number of white players. Black players have fluctuated all over the place and it seems to me that the talent pool of black players has had nothing to do with it. It is all opportunity. Now, granted, some black players lost some opportunities to WWII, maybe. And there is, I believe, a dip in the number of white players we elected during the war, too--obviously opportunities were lost to WWII by both black and white players.

But that does not explain the fact that the differential in black HoMers from the golden age vs. the war/integration era is much much larger than the differential for white players.

So I come back to this: The NeLs declined and disappeared 6-7-8-10 years before the MLs became fully integrated. Initially the Larry Dobys and Roy Campanellas and Jackie Robinsons and then the Mays and Banks types got the opportunity to play in the MLs. The Clarksons and Artie Wilsons and Willard Browns did not, really. This is primarily why there is a dip in black HoMers 1945-55, IMO

If this is something that can be "fixed" (which is not necessarily the case, but if it is) Luke Easter, to me, is emerging as part of the fix.
   114. sunnyday2 Posted: August 25, 2006 at 04:40 PM (#2156563)
Oh, OK, Chris found the other chart:

I would look at the '50s and ask not how many NeLers we inducted, but how many blacks. The assumption is that the black talent pool was more or less constant in the '30s, '40s and '50s, though, yes, the war had a small impact on that.

The numbers of black HoMers is not, then, a function of the talent pool but of opportunity. For the '50s the relevant comparison is not blacks who had played in the NeLs but the entire talent pool, wherever they played. If we elected fewer blacks from the '40s than the '50s, this reflects their differentially more limited opportunity.

(I think even more to the point, we might find that we elected fewer blacks from '45 to '55 but we don't have that data, at least not right here, right now on this thread. But if we did elect fewer black players from '45-'55 than from '35-'45 or from '55-'65, that is a function of opportunity, not of the talent pool.)

This at least is my sense of what the numbers and the history mean.
   115. Mark Donelson Posted: August 25, 2006 at 04:52 PM (#2156578)
In a period when the NeLs were dying out but most African-American ballplayers weren't getting to the majors either, is it possible that some of the potential HOMers we're missing from this pool for that period just...did something else that made a better living for them instead? It doesn't seem as if, for this brief period, baseball was a terribly secure career path. So perhaps the missing African-American HOMers born in the '20s just never played long enough at high enough levels to even gain our notice?

I admit this is rather like karlmagnus's Leever argument. Of course, I'm not arguing that we elect anyone in particular based on it--if this is true, it's good to know (or theorize, anyway--hard to prove either way)--but we can't really fix it.

I have to admit Luke Easter is becoming more appealing to me too. However, like the good Doctor C., I can't quite make myself pull the trigger either, though--so much uncertainty there.
   116. Dr. Chaleeko Posted: August 25, 2006 at 05:21 PM (#2156603)
It doesn't seem as if, for this brief period, baseball was a terribly secure career path. So perhaps the missing African-American HOMers born in the '20s just never played long enough at high enough levels to even gain our notice?

In short, I'd bet against it. We're talking an America that's still heavily segregated and in which black workers are probably making some fraction on the dollar of what whitey is making. I'd bet that economic and financial opportunity for black men in any kind of organized baseball where they acted like something of a free agent was much better than what they could have done in the labor force...which incidentally some of them may have had no training for if they'd played in some incarnation of organized black/caribbean/white ball since they were 17 or 19.
   117. sunnyday2 Posted: August 25, 2006 at 05:40 PM (#2156616)
I do think it is plausible that football began to siphon off blacks at this time, and maybe basketball and probably boxing too. But I doubt that the pool shrunk by more than 10 percent, if that, whereas we are electing <50 percent as many.

Again, I've never said I know what the fix is, other than to use some imagination in looking at the careers of the next tier of candidates like Easter, Wilson and Clarkson.

I would throw another name in the hopper, though he was the older generation--representing the over-represented golden age: Bobby Estalella. His situation is exactly analogous to the integration era blacks, however. Like some of them, Estalella actually got the opportunity to play in the majors. But like them it wasn't a fair chance, it was a truncated chance in which he was evaluated against a different/higher standard than white players.

The problem is we look at the record of these players and it looks like they got a shot, so we intuitively say, well, here, here's the record, here's the kind of player he was. Those are the hardest ones to evaluate. The ones who never got any shot whatsoever, we evaluate accordingly. With the integration era guys and Bobby Estalella, we don't really know if the gaps were based on their true talent and ability, or on the color of their skin, so we just don't know how to adjust.
   118. Dr. Chaleeko Posted: August 25, 2006 at 06:56 PM (#2156698)
Just read this article in the wikipedia: http://en.wikipedia.org/wiki/Black_players_in_American_professional_football.

Seems like the NFL may not have siphoned much talent away from baseball. The article, which has a pretty sketchy chronology, seems to suggest that after 1934 the NFL had zero black players for about 12 years. The article isn't real specific about how many blacks played in the wake of the Rams' Strode and Washington, but it seems apparent that the answer is "not many." Indirectly it sites stats that show the NFL was 30% black in 1970.

That said, it's also apparent from the article that black men were playing football in college and semi-pro ranks with regularity, so it's possible that the supposition that football, but not pro football, drained the ranks is workable. Still, with football a fall/winter sport and baseball a spring sport, and football players being less bulky and more lithe back then, it's possible that many footballers played baseball simultaneously and could have leveraged the opportunity into a pro baseball career.

I'm kind of thinking that football didn't become a really serious drain until the AFL started signing a lot of black men in the 1960s.
   119. Mark Donelson Posted: August 25, 2006 at 07:49 PM (#2156737)
I'm going on vacation and will be w/o Internet access for at least part of it--John, can you post this (my prelim) as my final ballot on the ballot thread on Monday? Thanks.

I’m an extreme peak voter; career numbers matter very little to me, except as a tiebreaker. I rely heavily on WS for hitters, with OPS+ and some WARP thrown in as well. For starting pitchers, it’s PRAA, with some WS and ERA+ adjustments for good measure. For relievers, I’ve adopted a mix of career total PRAA and year-by-year peak PRAA, with an emphasis on the latter, which seems to produce the most sensible results I can come up with.

pHOM: Slaughter and Torre.

1983 ballot:

1. José Méndez (pHOM 1960). Comparable with some of the best ML pitchers of his era, and those are some pretty shiny names.

2. Rube Waddell (pHOM 1919). Love his PRAA, love his strikeouts, and the unearned runs don’t bug me that much (especially with the revelation that they’re not even as bad as they at first seemed).

3. Dobie Moore (pHOM 1932). Fantastic peak, even if it’s not quite what we thought before the new MLEs. Like all the early NeLers, he’s hard to evaluate, but I’m pretty confident this guy was the real deal.

4. Hugh Duffy (pHOM 1930). I’ve read all the pro and con arguments, and I have to say I keep ending up in the same place on Duffy: He belongs, at least for anyone with as much of a peak emphasis as I have.

5. Cupid Childs (pHOM 1938). I’m convinced he was the class of his position at his time, dominant for long enough to get my vote.

6. Ralph Kiner (pHOM 1964). He still looks pretty good to a peak voter.

7. Dizzy Dean (pHOM 1967). Sure, it’s a really short peak (which is why he’s not even higher), but he was inarguably dominant during it.

8. Charlie Keller (pHOM 1973). A peak I just couldn’t argue around anymore. With even fairly conservative war credit, he’s VERY close to Kiner.

9. Bill Freehan (pHOM 1983). The best catcher we’ve seen in a while. Nearly identical to Elston Howard, but the slightly better five-year peak edges out Howard’s possibly very slightly better defense.

10. Eddie Cicotte (pHOM 1972). Clear enough dominance for long enough, in my book.

11. Ed Williamson (pHOM 1931). A lost cause, but still the best of the backlog 3Bs, for my taste.

12. Vic Willis (pHOM 1961). Not the most dominant pitcher of his era, perhaps, but he presents an awfully appealing peak.

[12a. Enos Slaughter (pHOM 1984). Looking again at his numbers alongside those of Clemente and Kaline made me realize I’d done Country wrong.]

13. Elston Howard (pHOM 1976). After my reconsideration, I found I’d been underrating Howard a bit. The various extenuating circumstances of his career can’t hide the great (if short) peak.

14. Al Rosen (pHOM 1968). Another very short peak, but five great years at this position are enough for me.

15. Charley Jones (pHOM 1976). After a long approach, he finally lands on my ballot. It’s hard to see through the AA haze, but he looks dominant enough to me.

16-20: Torre (1984), Trouppe (1967), Browning (1979), B. Robinson, Walters (1968)
21-25: Cravath, Pierce, Fox, Wynn, Gomez
26-30: Bresnahan (1973), Berger, [Faber], F. Howard, McGraw, [B. Williams], H. Smith
31-35: Roush, Pesky, Redding (1975), Trout, Boyer
36-40: Joss, [Reese], H. Wilson, [Lyons], Leach, [Wynn], Oms, [W. Ford], McCormick
41-45: Doyle, Minoso, Chance, J. Ryan, Elliott
46-50: Cepeda, [Lemon], G. Burns, Colavito, Rizzuto, Dunlap

Required Explanations and Newbies:

•Brooks Robinson. An awful lot like Boyer, who I’m not overly high on, so the question is: How much of a boost do I give him because WS underrates defense? Given his reputation, then and now, as well as the defensive numbers, I’ve given him a lot, but it still isn’t quite enough to get him on ballot. He’ll make my pHOM shortly, though. At #19.

•Torre. As with Norm Cash, I resist the temptation to overvalue the one outlying peak season, but he gets position boosts that help him out, too. It’s a bit complicated figuring out exactly how much to give him, of course…in the end, I can’t see him above Freehan or Howard (or Rosen) on peak with positional adjustments, so he ends up just off-ballot. (It was tempting to boost him a slot just to get one of the likely electees this year on my ballot. But I was strong!) He’s a pHOM guy for me this year, so that’s something… At #16.

•Sewell. There’s just not remotely enough peak here for me. In my re-evaluation, Stephens jumps over him again, and now both are just outside my top 50.

•Minoso. Kind of the rich man’s Van Haltren: very good for a long time, but he doesn’t really have the kind of peak I’m looking for. At #44.

•Fairly. Not even close—just no peak at all. Not even in my consideration set.

•Fregosi. Better than I knew, but he looks like Sewell-lite to me, and I don’t like Sewell all that much. Not in or terribly close to my top 50.

•Wood. He DOES have a peak, but not quite enough of one. Oddly enough, his PRAA peak looks a lot like that of a very different pitcher in everything but last name, Joe Wood. Both are hanging around well outside my top 50, but within the consideration set.
   120. Howie Menckel Posted: August 25, 2006 at 07:54 PM (#2156743)
Well, the 1925-42 numbers of major-leaguers elected ARE a bit higher than 1901-24, so it's probably fair to say that the original comment was overstated.
No offense taken at anyone wanting more evidence, of course. But it appears we will indeed elect similar numbers of MLB players over a very long stretch of the 20th century.

MLB HOMers
1901-24 - 18 to 22 (25 in 1916; 17 in 1918 and 1920; 23 in 1924)
1925-42 - 26 to 30 (31 in 1926)
1943-45 - 17/11/10
1946-52 - 24 to 25 (21 in 1946; 22 in 1952)
1953-61 - 26 to 29
then slowly sliding, as more and more future HOMers are not yet eligible.

NeLg HOMers
1901-09 - 3 to 4 (2 in 1901)
1910-19 - 6 to 7 (8 in 1916)
1920-22 - 9/11/11
1923-41 - 13 to 16 (12 in 1933; 11 in 1938)
1942-46 - 8 to 11
1947-48 - 6/5
   121. Chris Fluit Posted: August 25, 2006 at 07:59 PM (#2156751)
Thanks for the breakdown, Howie.
   122. Howie Menckel Posted: August 25, 2006 at 08:50 PM (#2156792)
Highlighting a couple of the years in which we elected the largest number of HOMers:


1926 (45)
NL (12) - Wheat, Alexander, Carey, Groh*, Rixey, Hornsby, Frisch, Vance, Hartnett, Terry, Waner, Ott*
AL (19) - Cobb, ECollins, WJohnson, Speaker, Faber, Ruth, Sisler, Heilmann, Covaleski, Goslin, Gehrig, Simmons, Lyons, Cochrane, Grove, Foxx*, Ruffing, Gehringer, Cronin*
NeL (14) - Lloyd, SJWilliams, Torriente, Charleston, Rogan, Beckwith, Mackey, JWilson, CPBell, Stearnes, BFoster, Suttles, Wells, Dihigo

1928 (46)
NL (11.9) - Alexander, Carey, Rixey, Hornsby, Sisler**, Frisch, Vance, Hartnett, Terry, Waner, Ott, Hubbell*
AL (18.1) - Cobb, ECollins*, Speaker*, Faber, Ruth, Sisler**, Heilmann, Covaleski*, Goslin, Gehrig, Simmons, Lyons, Cochrane, Grove, Foxx, Ruffing, Gehringer, Cronin*, Dickey*
NeL (15) - Lloyd, SJWilliams, Torriente*, Charleston, Rogan, Beckwith, Mackey, JWilson, CPBell, Stearnes, BFoster, Suttles, Wells, Dihigo, Paige

as opposed to
1956 (29)
NL (18) - Reese, Musial, Spahn, JRobinson, Snider, Campanella, Ashburn, Roberts, Irvin, Mays, Mathews, Wilhelm, Banks, Aaron, Koufax*, Clemente, Drysdale*, FRobinson
AL (11) - Feller*, TWilliams, Wynn, Lemon, Doby, Berra, Ford, Mantle, Kaline, Slaughter, Bunning*, Killebrew*

* is part-time (10+ G, fewer than half in field, or fewer than 1 IP per G or 35 G)
   123. Howie Menckel Posted: August 25, 2006 at 09:21 PM (#2156803)
possible adds to the 1956 list of 29 HOMers, with 1983 ballot rank in parentheses: BRobinson* (3rd), Minoso (10th), Pierce (11th), NFox (14th), KBoyer (20th).

possible adds to 1926 list of 45 HOMers or the 1928 list of 46 HOMers: Sewell (7th) (and maybe Moore in 1926??)
   124. sunnyday2 Posted: August 25, 2006 at 10:07 PM (#2156824)
The real point is how many blacks are there:

1926--14, all in the NeL obviously, none in the AL or NL
1928--15

1956--JRobinson, Campy, Irvin, Mays, Banks, FRobby, Doby = 7 from a black talent pool that had to have been as big or bigger than the one in the '20s. With only Minoso as a real remaining candidate.

Or did it? Some say no, the talent pool was smaller for some reason.

I believe that it was that they (the other guys not named JRobinson, Campy, Irvin, Mays, Banks, FRobby, Doby) lacked the opportunity to play, not in 1956 per se, but they lacked the opportunity to have a "normal" career whether in the MLs (as in the 1960sff, or the NeLs as in the '20s).
   125. yest Posted: August 25, 2006 at 11:20 PM (#2156855)
as we get more into the expansion era how do we deal wit the fact that teams were playing teams that weren't in their division less then teams in their division if either division was extremly weak or strong?
   126. Joey Numbaz (Scruff) Posted: August 26, 2006 at 02:06 AM (#2157140)
Yest - that's a very minor consideration . . . especially in the AL, which essentially had a balanced schedule once they expanded in 1977 (13 in division, 12 out).

In the NL, it's more pronounced since they played and 18/12 schedule, but fortunately, in the NL the divisions were pretty balanced most of the way.

If you didn't even consider it at all you'd be missing very, very little.

I've done some studies of my own, messing around with extreme schedules, and it very rarely means more than one or two games in the standings, no matter which division you play against more. In the most extreme cases it could swing like 4 games in the standings, but that's generally only if the division is very small (4 teams) and either very good or very bad.
   127. Joey Numbaz (Scruff) Posted: August 26, 2006 at 02:13 AM (#2157153)
I do think that Win Shares need to be adjusted for AL hitters, who are at a disadvantage due the offense being split 9 ways instead of 8. I think AL offensive WS post 1972 need a 10% boost, with nothing to offset it pitching/fielding wise - they simply added more offense, and WS doesn't account for it.

For the rate stats, I know BB-ref doesn't include pitchers in the league totals when computing OPS+, but I do think they include the DH. If that's the case, AL OPS+ are also going to be understated post-1972.

Marc - I disagree that pitchers that hit can't be less valuable than pitchers that don't. It's different than the DH. Pitcher offensive replacement level is 0 - no manager selects his starting pitcher based on his offense.

But for fielding, I do agree that even bad fielders have more value defensively than DHs. Take Manny vs. Ortiz. Manny's ability to play the field, however badly he does it, allows Ortiz to be in the lineup. If Big Papi could play in the field, it would enable them to get another bat in the lineup.

Since I use average as my replacement level for fielding . . . I'd rate all DHs as bad as I rate the worst fielders that play regularly.

For WS, it's not an issue, since all fielders get credit, and DH's don't.
   128. Brent Posted: August 26, 2006 at 04:21 AM (#2157391)
I do think that Win Shares need to be adjusted for AL hitters, who are at a disadvantage due the offense being split 9 ways instead of 8. I think AL offensive WS post 1972 need a 10% boost, with nothing to offset it pitching/fielding wise

No. That's inconsistent with the logic of win shares. By construction they have to equal 3 times wins. James explained this all very clearly in the Win Shares book.
   129. Brent Posted: August 26, 2006 at 05:03 AM (#2157419)
To explain a little more, every meaningful measurement has to be in well-defined units. For win shares, James defined the unit of measurement as 1/3 of a win -- for example, if a team wins 81 games, they will have 81*3 = 243 win shares. If you add a 10% boost to hitters without offsetting it among pitchers/fielders, the total win shares would be about 255, but the number of wins would still be 81. So the unit of measurement for the AL would be .318 wins, while the unit of measurement for the NL would be .333 wins. They would no longer be in the same units, so it would no longer be valid to compare them.
   130. sunnyday2 Posted: August 26, 2006 at 05:19 AM (#2157430)
OTOH is it valid to compare them now? Look up the current 2006 WS totals. Beltran and Pujols lead the ML with about 27 and the rest of the top 5 are all in the NL. Then comes the best in the AL, Manny at about 23. Clearly AL hitters are at a disadvantage for the reason Joe mentions.
   131. Joey Numbaz (Scruff) Posted: August 26, 2006 at 05:42 AM (#2157439)
"No. That's inconsistent with the logic of win shares. By construction they have to equal 3 times wins. James explained this all very clearly in the Win Shares book."


He was very clear. And very wrong. :-)

I know what Win Shares are, I've built a spreadsheet that calculates them by hand. I probably know the system in detail in terms of how it works and all of the little things within it as well as anyone who isn't Bill James or Jim Henzler.

What Bill James refused to acknowledge was that the AL artificially added offense with the DH.

The only way that I can think of, to adjust for this artificial addition of offense (via the DH) is to add offensive Win Shares to offset this.

It makes perfect logical sense when you think about it. The AL uses a 9 man offense, the NL an 8 man offense. AL players are at an automatic 12.5% disadvantage. You have to account for this. I was just throwing 10% because it's a nice eyeball way to do it conservatively, but if you want to get technical, all AL offensive WS post-1972 should be multiplied by a factor of 1.125 to keep pace with the NL.

Now you might say the NL was better than the AL in the 1970s and early 80s (and you'd be right). But that's a separate adjustment.
   132. Brent Posted: August 26, 2006 at 06:11 AM (#2157459)
No, Joe. I think Bill James understood his own system better than you do. When I read his explanation it makes perfect sense. AL hitters are in higher run environment (5.02 R/G vs. 4.74 for NL), so their hits are worth less. I agree AL hitters are at a disadvantage because of the rules, but their disadvantage results in less value. We judge players in the context of the rules they play in. In the 1960s, the rules allowed high mounds and large strike zone; in the 1870s the rules allowed fair-foul bunts; in the 1910s the rules allowed spitballs. These rules put some players at a disadvantage, just as the DH puts AL batters at a disadvantage. But their disadvantage is a real loss in value, and you can't just pretend it isn't there by adding X% and pretending that your unit of measurement hasn't changed.

Beltran is ahead of Manny because of his fielding and playing in a lower run-scoring environment.
   133. Joey Numbaz (Scruff) Posted: August 26, 2006 at 11:07 AM (#2157483)
Of course Bill James understands his system better than me, I never said he doesn't.

But he's flat out wrong on this part of it, whether he realizes it or not.

"AL hitters are in higher run environment (5.02 R/G vs. 4.74 for NL), so their hits are worth less."


Huh? WS already normalizes to league average - what does the higher run enviroment have to do with it?

I guess we'll have to agree to disagree. But the way James handles it basically penalizes a player for playing in the AL, which makes zero sense, whether they technically have less value or not.

"These rules put some players at a disadvantage, just as the DH puts AL batters at a disadvantage. But their disadvantage is a real loss in value, and you can't just pretend it isn't there by adding X% and pretending that your unit of measurement hasn't changed"


You can do that. And you should. If you are artificially adding offense to the league in real life, you also need to do it in your uber-system. You can't just close your eyes and hold your nose and ignore it because it's tough to deal with, or prevents everything from adding up correctly.

Adding to AL win shares post-DH is not any different than lowering the league average 5% (which is standard) when figuring things like OWP for AL hitters.

It has to be done, or you are basically penalizing a player for signing an American League contract, which is flat out wrong, no matter how you slice it.
   134. TomH Posted: August 26, 2006 at 01:25 PM (#2157508)
Two different things happening here:

If you are trying to measure WINS (or VALUE), James' method is correct; there are only so many wins, and they need to be divided equally, not "fudged".

But if you are measuring ABILITY, NL hitters have an unfair advantage over AL (using the DH) hitters.
Think about it - if the AL suddenly went to a 11-man batting order, Manny Ramirez would be worth less; he would only come to the plate 540 times a year instead of 660. The "wins" would be more scattered / dispersed to other playes. But is this fair to Manny? No. He didn't get any "worse" as a hitter; the rules merely changed. If he had played in the NL, he would have been worth more wins. What if the AL adopted a 180 game schedule. Voila, more Wins! Their hitters are 'better'! :)

Conclusion: for the HoM, we are measuring value, but also attempting to put all players on an equal playing field as much as possible (ability). In order to do that, some adjustment needs to be made for post-1972 AL hitters. But I haven't done the math to figure out how much that adjustment should be.
   135. sunnyday2 Posted: August 26, 2006 at 01:40 PM (#2157516)
In a perfect world it would be nice if we didn't need any fudge, but James himself introduced the bullshirt dump. When we fudge (or at least some of the fudge), all we're doing is trying to quantify the bullshirt, which is no less valid and somewhat more rigorous than what James does.

OTOH I never though James' bullshirt dump was particularly valid, at least when players seem to jump all over the damn place as a result. A little movement maybe, but as we know, his lists bear very little resemblance to his own numerical results. WTF!?

Well, I try to keep it to 3 WS per win. In the 19C I move WS from pitchers to fielders. If I give MLE credit to a player (Cravath, Grove, Averill, the NeLers, etc.) I am in theory flushing a bunch of replacement level (and on up toward the average) out of the league. I don't have to have the players affected on my spreadsheet in order to do it. There is no reason why it doesn't work but it doesn't affect anybody I care about. (OK, some players stay in the league longer in real life than they would with the larger pool, you got me there.)

During WWII, I give "extra credit" to players who are in the military, and I discount everybody who is actually playing. That and similarly flushing out the replacement through average level players.

What to do with the DH is the trickiest of all, I think. Well, that and adjusting pitchers for changing work loads. So, my question:

Can one fudge with the basic separation of WS among arms, bats and gloves and come up with something valid? Is offense more important, more valuable (worthy of more of the WS pie) in the AL than the NL?
   136. Dr. Chaleeko Posted: August 26, 2006 at 02:00 PM (#2157521)
I don't know as much as Bill or Joe or Brent, but...

The NL does not use an 8 man offense, it uses a nine man offense. In fact, the NL uses 8 men, pitchers, and many, many pinch hitters. And while pinch hitters come to bat at a disadvantage, they are much better than pitchers.

What percent of PAs are going to pitchers since interleague play began?

AL: 2760 pPAs vs 881053 lgPAs, or .3% of all PAs
NL: 57977 pPAs vs 975753 Total PAs, or 5.9% of all PAs

So for the interleague period at least, the difference in opportunity is not 10% at all, but rather 5.6%. To model a non-interleague period, the difference is 6%, presuming similar pinch-hitting trends (we'd have to check on that). How about in terms of productivity?

AL: pitchers created 123 runs, the league as a whole created 116206, so pitchers created .1% of all runs.
NL: pitchers created 2195 runs, the league as a whole created 124275, so pitchers created 1.8% of all runs.

This one's even narrower, a difference of 1.7%, and it wouldn't be much higher without the DH (though pitcher batting has decreased in producitivty since the DH's advent, as part of the historical trend toward crappier pitcher batting).

More broadly,
From 1996-2005, non-pitcher batters in the AL created 116206 runs in 605961 outs or 5.18 rc/27. From 1996-2005, non-pitcher batters in the NL created 122070 runs in 630905 outs or 5.22 rc/27.

but

From 1996-2005, total team offense in the AL created 116331 runs in 608382 outs or 5.16 rc/27.
From 1996-2005, total team offense in the NL created 124275 runa in 669550 outs or 5.01 rc/27.

A three percent difference between the leagues.

The eight men plus pinch hitters of the NL are as productive or more so then the 9 men in the NL are on a per game basis. Adding in the pitchers knocks 4% off their offensive efficiency.

One more thing. How many more PAs does a lineup generate with the DH than without?

In our study period...

AL: 881053 PAs in 22662 (22680 scheduled games), or 38.9 PAs per game. That's 6301.8 per annum.
NL: 975753 PAs in 25270 (25272 scheduled games), or 38.6 PAs per game. That's 6253.2 per annum.

Again, despite the presence of the easy outs in the nine hole, there's hardly any difference. In fact, it comes down to five extra turns through the lineup of an AL team plus another turn for the top four hitters in the lineup. Or 5.0-5.5 runs if you want to look at it like that. (It would be higher in the non-interleague era.)

This data (from the SBE) suggests to me that there is some difference, but that for the interleague period at least:
-it's not as big as the 1-in-9 vs. 1-in-8 argument would say it was (i.e. not 10%)
-the DH makes no difference in overall productivity of non-hitting players (unless Coors Field is having an extreme effect here, I don't think it's all that likely though)
-the DH prevents only a small amount of offensive hemorhagging due to today's extensive use of pinch hitting and relief pitching.
-the difference in opportunity is very small, amounting to a game's worth of PAs per year
-it's possible that the pre-interleague period has modest differences in all these things and should perhaps be treated slightly differently than the interleague period.

So returning to the question of WS, OPS+, and other numbers. I think we need to partition WS and OPS+. OPS+ describes skill and performance. To compare guys in two leagues, I think it could be appropriate to filter for the DH in some way. I'm not sure that's true, but it's possibly true. But WS describes a player's value to his team. His value is not pegged to his performance per se, but to how much that performance contributes relative to the contributions of the other guys on his team. OPS+ doesn't care about the team, Win Shares cares mostly about the team.

An NL player's value to his team will be a little higher since there's a slight difference in the runs created by the teams and his contribution will be higher than that of the pitcher's slot. In addition because pinch hitters and pitchers as individuals aren't batting terribly often per person (but are holding down a lineup spot in the aggregate), they are probably not contributing many marginal runs. Therefore, the NL guys are probably netting a higher percentage of their teams' marginal runs (warning: unsupported assertion), while the AL guys are dividing the pie more equally. That jibes with what Joe is saying in most ways.

But unlike OPS+, WS is measuring value to team. If a team fields nine Manny Ramirez clones, none of them will have more BWS than another for the obvious reason that their performance is identical and identically valuable to the team. No matter how much Shaughnessey-ignored-or-bandwagoned greatness or Shaughnessey-lampooned crubminess the clones show. I know that it's about marginal runs, not percentage of the pie, but at a certain point it is about percentage of the over-the-margin pie. In the AL, it's not easy for Manny to contribute Pujolsian value to his team because more hitters are contributing in that region. But IS absolutely appropriate to measure his contribution relative to the total contribution of his teammates. WS is after how many wins the guy contributes to his team. You can't just tack on BWS because it not only throws off the hitting/pitching balance, it also distorts the balance among the batters on the team...does the DH get the 10% DH bonus proposed for Manny? Why not? If the Sox trade Manny for Hafner tomorrow, and Ortiz moves to first (and KY to LF) to accomodate Hafner at DH, does Ortiz only get a 1.6% bonus? How about when a guy plays 51% of his games at DH but 49% at 1B? Does he get a 4.9% DH bonus?

OK, but what effect might the DH have on players elsewhere on the continuum between good and good-riddance? Well, it seems like it should (in all likelihood) be easier for a stinky hitter to contribute little if any batting value in the AL. This should stand to reason: if the level of marginal production increases due to the presence of another good hitter, but Stinky Middleman hits like Ozzie Guillen, he won't keep up with the rise in marginal value. Put another way, if the rise in marginal value is about the same thing as bumping hitters seven and eight down one slot, then our Guillensque hitter is now the ninth hitter and the line drawn probably below the eight hitter, which he used to be, is now above him. He's producing, in an AL environment, at a lower level, relative to his teammates, than his NL counterpart, Suck E. Middledude who hits similarly but has the pitcher-slot buffer to keep him over above the marginalia. This may have the effect (or may not?) of shifting some small number of Win Shares away from the lower-order hitters on teams toward the better hitters. Dunno.

Or I could be wrong, because I do recognize that my understanding of the system is less nuanced than it should be. But I think Brent has got it right by making the distinction of a value-driven system versus a performance-driven system. Two different goals, two different measurements, two different needs.
   137. Brent Posted: August 26, 2006 at 04:48 PM (#2157561)
Huh? WS already normalizes to league average - what does the higher run enviroment have to do with it?

Yes, it's the normalizing to league average is what causes hitting performance in the AL to be less valuable than the same performance in the NL. The reason the DH was instituted was because the AL wanted to boost offense.

I'll agree that there is a distinction between value/wins and ability/merit. James designed his system to measure the former; he recognized (correctly) that it would be logically inconsistent to credit AL players with more than their actual number of wins in a system that is designed to assign credit for wins.

For HoM voting purposes, most of us (including me) deviate from pure value for a variety of reasons. If some voters want to adjust AL batting win shares for purposes of parity with the NL, I won't object, but let's recognize that that's a choice about what you think is fair, not an analytical flaw of the win shares system. I don't think there's a compelling case to make that adjustment.

In fact, there's an analogous case for which I don't think any voters made a similar adjustment: the grandfathering of spitball pitchers when the spitball ban was instituted in 1920. This provision put some pitchers at an advantage, other pitchers who had to face them or who were being compared to them at a disadvantage. I'm not aware that any voters boosted the results of the pitchers who weren't grandfathered for their disadvantage in having to be compared to pitchers facing a different set of rules.

Also, I think Dr. C is correct that the effect of the DH is much smaller than 10%.
   138. OCF Posted: August 26, 2006 at 04:56 PM (#2157565)
There are at least two other ways to get at this, for people who prefer to use offensive stats a little closer to raw measures than WS. One is to always rely on league averages that have the pitchers removed.

But suppose that you're using things that don't have the pitchers removed. By exactly what factor would you estimate that the DH rule increases league-wide scoring? I we had a factor like that to use, we could simply compound it on top of park factors.
   139. Chris Cobb Posted: August 26, 2006 at 06:38 PM (#2157606)
Some comments on the DH issue.

FWIW, WARP2 in the post-DH era makes exactly the sort of changes that Joe is arguing should be made to win shares when comparing NL and AL batters.

If you want to see the effects of the DH on the relative offensive prowess of hitters, check out the swings in EQA and BRAR for players who switch leagues. It is considerable. Omar Vizquel offers a particularly vivid example, as I recall. Jim Thome's good Philly year provides an example at the other end of the batting spectrum.

Dr. Chaleeko's analysis, if I understand it correctly, is largely irrelevant to the question of how to adjust batting win shares for AL players. The overall question is: what is the difference between the marginal offensive contribution of the DH slot in AL batting order, as a percentage of overall offense and the marginal offensive contribution of the pitcher/pinch-hitter slot in the NL batting order?

The total number of offensive win shares available to batters in both leagues will be the same, but if the DH gets an average hitter share, but the P/PH slot gets only say, 1/3 of an average hitter share (half PA being taken by pitchrs whose overall marginal run production is less than zero, half PA taken by PH whose overall production is below average), then average AL hitters will lose the difference between those two shares. An average batting slot share, assuming a 9-batter lineup, is 12.96 win shares/162 games. 1/2 of that is 4.32, so 8 non-DH hitters in a DH league vs. a P/PH league lose 8.64 ws to the DH per 162 games, so each slot individually loses 1.08 ws/162 games.

This seems to me to be the correct approach to estimating the impact of the DH. This analysis could be fine-tuned by establishing the marginal offensive production of the DH (possibly somewhat above average, overall) and the P/PH overall. 1/3 of an average hitting slot might be too low, or it might be too high. I just picked it as a reasonable estimate.

It looks to me like Joe's 10% estimate is a bit too high, but that Dr. Chaleeko's estimate that the DH has very little impact at all is far too low.

However, a model that looks more closely at what is actually happening with marginal runs than I have is needed. If pitchers are actually so far below other hitters that their negative marginal run production offsets the positive marginal run production of pinch-hitters who share their batting slot, then the rough 1 per 9 bws vs. 1 per 8 bws would turn out to be correct.
   140. karlmagnus Posted: August 26, 2006 at 06:42 PM (#2157608)
If pitchers get 5.6% more of the PAs in the NL, then AL offensive WS should be scaled up by 99.7/94.1 to equalize the two offenses, shouldn't they? You could take it off the defense, since the theroem that offense and defense must have equal WS is plucked out of the air and has no firm basis (It's this kind of thing that makes me mostly ignore WS, which over a career seem to me as much as 25-30% out in a number of cases.)

You have an additional question for a DH, which is his greater ability to play hurt, which has arisen specifically in the Manny case. Manny has a sore knee and a hamstring problem, which aggravate each other, and can therefore play LF only sparingly (I bet he leaves the game early tonight, for example.) He will thereby lose PA and WS, and may even lose more PA and WS by hitting less regularily, not getting the chance to see the pitcher 3 times if he plays only part of a game, etc. On the other hand, we're told hitting itself is no problem, so except that he might miss a few hits by being immobile on the bases, he could play DH freely.

Meanwhile the Great Lump plays DH with no problem, sails above 50HR, hits a few game winners because he hits ahead of Manny, and is touted for the MVP that Manny has never won and deserves to get. This is bloody unfair, which is typical of life, but we ought at least to recognize it by e.g. penalizing Edgar Martinez a bit when he comes due (though I recognize there's an argument for MiL credit for Edgar as well.)
   141. Brent Posted: August 26, 2006 at 07:26 PM (#2157627)
You have an additional question for a DH, which is his greater ability to play hurt, which has arisen specifically in the Manny case.

Yes. In addition, you've got the effect of the DH on career length, which has allowed players careers to be extended. The DH changed the nature of the game, adding real value to some players and subtracting it from others. James talked about all of this in his book, giving the example of Jim Rice, whose value was reduced by the DH during the time he played outfield and was raised by the DH during the time he was able to DH. When you folks think about this as just an issue of normalizing batting rate statistics across the two leagues, you're thinking much too narrowly. I don't believe it's possible to accurately adjust for it simply by tinkering with a few formulas.
   142. Chris Cobb Posted: August 26, 2006 at 08:54 PM (#2157715)
When you folks think about this as just an issue of normalizing batting rate statistics across the two leagues, you're thinking much too narrowly. I don't believe it's possible to accurately adjust for it simply by tinkering with a few formulas.

I would agree, but figuring out an appropriate normalization for batting rate stats is still useful. Then issues of playing time, defensive value, and career length need to be thought through as well . . .
   143. yest Posted: August 27, 2006 at 02:39 AM (#2158204)
Yest - that's a very minor consideration . . . especially in the AL, which essentially had a balanced schedule once they expanded in 1977 (13 in division, 12 out).

but it can sevely effect pitchers.
   144. sunnyday2 Posted: August 27, 2006 at 03:13 AM (#2158263)
All of these are reasons why a 300 WS or 350 WS, or 500 HR, or 3000 hit career should no longer be taken as necessarily denoting a "great," HoM-caliber player. Especially Brent's point about the DH extending careers.
   145. karlmagnus Posted: August 27, 2006 at 02:26 PM (#2158414)
Sunnyday2, which is why people should look more closely at the 2930 hits of Jake Beckley, equivelent to about 3350 on a 162 game schedule. Back in the 1890s men were men, a hit was a hit, and your career could be terminated by dissing John McGraw while rounding third base.
   146. Jeff M Posted: August 27, 2006 at 02:29 PM (#2158416)
I'm with Brent all the way on this one. I'm not going to pluck a guy off an AL team and pretend he played for an NL team. By evaluating what he contributed, I'm not "penalizing" him for playing in the AL. The circumstances are what they are. The player contributed what he contributed. If I give an AL hitter more WS, where do I take them from? Defense? What's the justification for that? Or is the argument that we tear down the whole system and ignore the 3 WS per win framework? If that's the case, WS falls apart at its foundation and I don't see the point of using it at all.

I see a growing trend (particularly in the last calendar year) to imagine careers that don't exist, perhaps motivated by the necessity of doing so for Negro Leaguers and war vets, or perhaps because it is interesting and fun. I'm not immune. Of course, that approach is fraught with danger because every adjustment we make that we think normalizes a player's contribution would ripple in all sorts of ways that we can never understand. The Butterfly Effect.

We had little choice for Negro Leaguers; some things are unavoidable. I think, however, that we have to be careful about when we choose to imagine what could have been for a player, particularly if we are not clear about how to make the adjustments. It's a sliding scale. The more reliable the information, the less compelling a case we need to apply those adjustments. The more compelling the case, the less sure we need to be about the adjustments.

For example, in most cases we have good park factor information. The abundance and quality of that information allows us to make adjustments to seemingly put everyone on a level playing field, regardless of how compelling we find the individual case. On the other hand, we don't have a lot of solid information about how to translate Negro League stats, but we have a compelling need, so we accept the risk and adjust as best we can.

With respect to upward adjustments to WS for AL hitters in the DH era, I don't think we have a reliable way of making the adjustment, for the reasons that have been discussed already (primarily by Brent). And I certainly don't see the AL hitters' situation as so compelling that I'll accept the risk of making the adjustments incorrectly.
   147. Howie Menckel Posted: August 27, 2006 at 03:06 PM (#2158435)
"If that's the case, WS falls apart at its foundation and I don't see the point of using it at all."

Well, that's another way of looking at it :)

All kidding aside, I think if the only two choices were relying to a great degree on WS in picking a top 15, or in not relying on them at all, I'd recommend the latter.

Using them as an additional piece of the puzzle, ok.
But once James, or someone else, further refines WS, I expect to see some 'buyer's remorse' here from those who voted for the downgradees (smiling gently, again).
   148. DavidFoss Posted: August 27, 2006 at 04:06 PM (#2158470)
but it can sevely effect pitchers.

In a 12-team league, that's 90 games in the division and 72 games outside the division. That's a 10% difference that only comes into play when there's a divisional imbalance. What's the most extreme years for this? 1973-74 NL? Looking at two roughly .500 teams those years, what is Montreal's edge over Houston? How much is the effect of more Reds/Dodgers games offset by more Padres games?

Starting in 1995, things get much trickier of course.
   149. Dr. Chaleeko Posted: August 27, 2006 at 05:59 PM (#2158517)
I think that to start with, it would be helpful to recalculate OPS+ to examine the effect of the DH and to arrive at a DH-neutral OPS+ measurement...if such a thing could be done without someone having to partition the DH and non-DH at bats of every guy in the league for several seasons. This way, we could more readily compare the offensive prowess of players between leagues.

That process would probably give us a bunch of good ideas for seeing how we could tinker with this for WS or WARP as needed.

Chris, to clarify, I don't think that the DH thing has to be ignored, but I don't see where it's as big as 10% and I'm not certain 5-6% is worth the potential trouble, except when we're making direct comparisons against guys who are really close in value. Particularly since even most career DHs played a significant portion of their careers afield. I should have said it that way before.
   150. Chris Cobb Posted: August 27, 2006 at 06:28 PM (#2158520)
Chris, to clarify, I don't think that the DH thing has to be ignored, but I don't see where it's as big as 10% and I'm not certain 5-6% is worth the potential trouble, except when we're making direct comparisons against guys who are really close in value.

Well, it's fortunate that we don't have to do much of that :-) !

Seriously, if an AL player is earning 1 batting win share less per 162 games (that's about a 6% reduction) than he would if he performed exactly the same way in the NL, and he has a 2000 game career, that's 12 career win shares. Given the tight bunching of players at the borderline, 12 win shares (or 4 WARP) is a significant amount. 5 win shares over a five-year peak? Also a significant amount.

It seems to me that unless the difference is no more than 2%, it's too large a factor to be safely ignored.

Further study along the lines you've started might show that the difference is 2% or less, but that hasn't been shown yet.
   151. Dr. Chaleeko Posted: August 27, 2006 at 08:38 PM (#2158612)
A point for Mr. Cobb! I was thinking about big unwieldy studies but you're right at the level we look at....
   152. Yoenis Cespedes, Baseball Savant Posted: August 28, 2006 at 12:19 AM (#2158867)
Here's a little lazy Sunday reading for you HOMers. It's a little something that I've been working on over the weekend. Eventually, I'd love to post a complete story to the Hardball Times or BTF. I'd love to hear what you think!

Following here is a little exercise in "what-if": What if Dick Allen played in the 1990s? This isn't just a statistical translation but also a look at how he might have reacted to the modern business of baseball and how the same things might have happened in different ways.

1991: The 21-year-old Allen finishes his minor-league career with a bang, leading the International League in home runs and RBI with Scranton and winning the league MVP. He makes local headlines in August after he goes into the stands to confront a local heckler who Allen said made a series of racial remarks concerning him and his family. The confrontation came after Allen struck out with the tying and winning runs in scoring position to end the first game of a doubleheader against Columbus. After order was restored, Allen played the second game and hit three home runs with seven RBI. The incident doesn’t sour Philly brass on the career .306 hitter, who earns a September 1 call-up to the big club. He hits .292/.280/.458 in a 10-game cup of coffee.

The Phillies finish a mediocre 78-84 on the year due mostly to a punchless offense. First baseman John Kruk is the only regular to manage at least a 105 OPS+. Aging outfielder Dale Murphy is the only other starter over 100. The team misses Lenny Dykstra, who is out most of the year after sustaining injuries in a car accident.

1992: His bat in midseason form, Allen tears through Spring Training pitching. However, his glove cannot shake off the winter frost, making six errors at third base. A week before the season starts, manager Jim Fregosi announces that Dave Hollins will start at third base and Allen will move to left field. Allen voices displeasure to the media about being “moved off my natural position,” but ultimately accepts the move.

The Phillies improve – by one game – to 79-83. But the story is Allen, the unanimous National League Rookie of the Year, first-time All-Star and strong fifth-place finisher in the MVP voting. Allen plays all 162 games and finishes with a .314/.388/.546 season. His name is all over the National League leaderboards: leading the league in runs scored (123), total bases (343) and extra-base hits (82), finishing second in doubles (44), third in hits (197) and triples (12), fourth in the NL in slugging percentage, OPS and OPS+ (162), sixth in home runs (26), batting average and on-base percentage, and tied for tenth in both RBI (90) and walks (76).

With Kruk, Allen, catcher Darren Daulton and Hollins all posting breakout seasons, the Phillies have the best three-through-six hitters in baseball. Sabermetricians note that the if the Phils performed to their Pythagorean expectation, they would have finished 87-75, good for second place in the NL East.

1993: It’s a dream season for Allen and Philadelphia, winning the East in a cakewalk as the Phillies finish 10 games ahead of Montreal with a 103-59 record, the best in franchise history and only the third time a Phillies team has won at least 100 games. Dykstra finally stays healthy and hits a career-best .305/.420/.482 (143 OPS+) while Allen shows no sign of a sophomore slump. He hits .319/.394/.525 (145 OPS+) with 24 home runs and 96 RBI, finishing among the league leaders in all of the rate stats and leading the league in hits with 202. He even throws in 24 steals in 26 attempts. Allen makes his second All-Star team and renegotiates his Phillies contract, signing a five-year, $23 million deal that will keep him with the team through the 1998 season.

In the World Series, the Phillies face the defending champion Toronto Blue Jays. Allen hits a three-run homer during the seventh inning of Game Six as part of an eight-run Phillies rally to give them a 9-5 lead. Phillies closer Mitch Williams surrenders a three-run homer to Toronto’s Joe Carter in the bottom of the ninth with one out to make it 9-8, but manages to strand the game-tying and Series-losing runs in scoring position to force a Game Seven. Allen goes 2-for-4 with an RBI in Game Seven as the Phils win 5-2 to claim their second World Series championship.

Controversy briefly erupts in the off-season when the MVP vote is announced. Allen finishes third in the voting behind San Francisco’s Barry Bonds and Dykstra. Both Philadelphia writers give their first-place votes to Dykstra, writing that his career year was the main reason that the Phillies won the World Series. Allen implies that the writers have racial motives in picking Dykstra, saying, “This city always likes so-called ‘working-class heroes.’ And it’s funny how this ‘working-class hero’ always looks the same. I think you know what that means.”

Local columnist Bill Conlin leads the charge of sportswriters and talk radio hosts blasting Allen, writing, “His remarks are especially irresponsible, considering his position of authority. Not only is he one of the best players in baseball, but he’s the Phillies’ new players’ representative for labor talks, effective Friday.”

1994: The World Champion Phillies look anything but as injuries sink any chance of a title defense before June rolls around. Allen himself starts the season on the 15-day disabled list after dislocating his shoulder in spring training. Only Allen and outfielder Jim Eisenreich manage to play even 100 games. Talk of a possible labor strike makes the mood in Veteran’s Stadium tense – it’s almost as if the team didn’t win anything. A lost season turns into a public-relations disaster for the Phillies as Allen becomes one of the leading voices agitating for a strike. The boobirds come out in full-force in July after Allen is quoted by several reporters in a postgame interview as saying, “This is about having the freedom to determine our economic destiny, to provide for our kids. I think that’s something a lot of folks can understand. The owners are treating us worse than I would treat my horses.”

On the field, Allen is picked for his third All-Star team and finishes ninth in the MVP voting with a sensational season. He hits .332/.423/.687 (181 OPS+) with 32 home runs and 90 RBI in just 100 games. The slugging percentage sets a new single-season franchise record, breaking the mark set in 1930 by Hall-of-Famer Chuck Klein .6868 to .6867.

But after the final games are played on August 11, Allen holds a press conference with local and national media before leaving Philadelphia. His final statement is a parting shot that rocks the team and the city: “When this thing gets settled, the first thing Bill Giles needs to do is call up Lee Thomas and tell him to get to work on trading me.”
   153. sunnyday2 Posted: August 28, 2006 at 03:13 AM (#2159038)
Here's a cheap shot, but true:

Joe Sewell, 3 SB in 19 attempts in 1927

74 SB and 72 CS for career
   154. TomH Posted: August 28, 2006 at 11:35 AM (#2159144)
lots of busted hit and runs?
This is already figured into OWP, RCAP, and WS for those who use those measures.
   155. DL from MN Posted: August 28, 2006 at 01:56 PM (#2159203)
Now I'm certain that nobody reads my prelim ballots.
   156. Joey Numbaz (Scruff) Posted: August 28, 2006 at 05:29 PM (#2159428)
I came to the 12.5% conclusion thinking that the DH is typically at least an average offensive player, and as such is going to take up at least 1/9th of the marginal WS on a typical team.

So the average NL player gets 12.5% of the offensive WS (P get very few WS in modern times, almost a negligible amount) while the average AL player only gets 11.1%. 1.4/11.1 = 12.6% of an AL penalty - assuming the DH is an average offensive player. I agree that it would be a good idea to figure the total marginal runs contributed by DHs, but I'd imagine the answer would end up being similar, if not higher.

Now you could make the argument that many DHs would be in the lineup anyway, playing somewhere else, so the edge isn't that high. But if that's the case, then AL players are also losing fielding WS, because the overall fielding level of the league is higher too. If those DHs would have been in the lineup, they would have been fielding (generally poorly) and lowering the overall level of fielding in the league as well.

*********

Regarding the Jim Rice comment, it still rings as hollow for me as it did when James said it. I interpret it to say - Rice's peak value is understated because of the DH, but I don't want to do anything about it, because it will be messy.

I don't understand the issue with not having it add up. In this one instance, it makes complete logical sense. The AL hasn't 'added up' when compared to the NL since 1972. The AL went and artificially added offense, when compared to all previous years in MLB. The only way to make it truly add up is to do the same with Win Shares and artificially add offensive WS, to account for the AL artificially adding offense. They don't have to come from anywhere, the come from thin air, just like the offense added by the implementation of the DH.
   157. Dr. Chaleeko Posted: August 28, 2006 at 06:26 PM (#2159501)
I don't understand the issue with not having it add up. In this one instance, it makes complete logical sense. The AL hasn't 'added up' when compared to the NL since 1972. The AL went and artificially added offense, when compared to all previous years in MLB. The only way to make it truly add up is to do the same with Win Shares and artificially add offensive WS, to account for the AL artificially adding offense. They don't have to come from anywhere, the come from thin air, just like the offense added by the implementation of the DH.

I see it the other way. There's nothing artificial, it's just the rules. You could say that all offense after 1892 is artificial since they moved the mound back. And by that same logic, you'd be correct! The rules are just the rules. The offensive context is exactly what it is. Manny's or A-Rod's value IS diminished by the DH because it should be. There's nothing inconsistent there. It's all a question of ability versus value. I don't think it's at all problematic to say that OPS+ should in some way be adjusted to allow hitters from each side to compete on equal footing because we're describing ability. But to say that an AL guy should get extra Win Shares for being in a DH league is to ignore the simple, obvious truth that individual run production is diluted slightly in a DH league. It has to be.

If you want to tinker with WS, you shouldn't be doing it on the end by adding more BWS. You should be doing it within the inner workings where the offensive context is adjusted. Then you don't have to worry about any of the thorny mess at the end with unbalanced offensive/defensive stuff. But the trick will be whether or not you can reasonably get away with making a single leaguewide DH adjustment, or whether you have to rejigger each team's BWS individually because their DHs contributed different levels of marginal offense compared to the rest of the league. Then you've got another thorny question. What do you do with the actual DHs? Does this mean that they have to produce a higher number of marginal runs to get BWS? Why is that appropriate? This is a question you'll face no matter how you adjust BWS. If Manny gets more BWS for being in a DH league, does Ortiz? And do you partition his playing time into non-DH and DH-only PAs to credit him with BWS?

It's a big, big mess.

I'm not trying to be a pain, but this feels like a deeper issue than simply tacking on WS.
   158. DavidFoss Posted: August 28, 2006 at 07:50 PM (#2159571)
I'm not trying to be a pain, but this feels like a deeper issue than simply tacking on WS.

It is. WS has never been perfect. Seasonal numbers for individual pitcher numbers are incredibly era dependent, yet we've been trying to adjust so that players are compared to their contemporaries in a fair manner. WS doesn't necessarily handle positional adjustments ideally either. Myself, I've only been using it to compare players at the same position -- others may have some other strategy for boosting C's and SS/2B types.

I see no difference between that and this AL/NL issue. If an AL player gets 600 PA and has an OPS+ of 150 and gets 35 WS, it doesn't seem fair that their NL clone gets 37 WS (or even a bit more after THT's adjustments to remove the zero-ing of pitcher bWS). (I will agree that 10% is too high due to PH's).

Many of us already are making tweaks to WS, so this AL/NL adjustment simply seems in line with those. So, I suppose I advocate leaving raw WS alone (because its nice that it 'adds up'), but perhaps reporting the percentage of time spent as an AL non-DH so that we can all make our own adjustments.
   159. Chris Cobb Posted: August 28, 2006 at 09:17 PM (#2159642)
Then you've got another thorny question. What do you do with the actual DHs? Does this mean that they have to produce a higher number of marginal runs to get BWS? Why is that appropriate? This is a question you'll face no matter how you adjust BWS. If Manny gets more BWS for being in a DH league, does Ortiz? And do you partition his playing time into non-DH and DH-only PAs to credit him with BWS?

My plan is to create "NL-equivalent" batting win share adjustment for all AL batters I evaluate from the DH era but to exclude from those adjustments their batting as a DH, insofar as I can effectively isolate it. The designated hitter benefits from the DH rule by gaining playing time, so I plan to exclude them from the adjustment.

I plan to make the adjustment by giving the player some amount of win shares per game, as for example .8 or 1.0 win share/162 games, which I will add in on a seasonal basis. When I do the calculation, I'll exclude DH games from the player's total for purposes of the adjustment. This method will require more bookkeeping, but not an unduly burdensome amount.

The difficult part will be determining the size of the adjustment, particularly if it turns out that it should not be "one size fits all." The appropriateness of a simple adjustment that is the same, say, for Omar Vizquel and for David Ortiz: that, for me, is the thorny question right now.

For WARP, I plan to use WARP2 batting numbers for post-1972 AL players.
   160. Chris Cobb Posted: August 28, 2006 at 09:17 PM (#2159643)
Then you've got another thorny question. What do you do with the actual DHs? Does this mean that they have to produce a higher number of marginal runs to get BWS? Why is that appropriate? This is a question you'll face no matter how you adjust BWS. If Manny gets more BWS for being in a DH league, does Ortiz? And do you partition his playing time into non-DH and DH-only PAs to credit him with BWS?

My plan is to create "NL-equivalent" batting win share adjustment for all AL batters I evaluate from the DH era but to exclude from those adjustments their batting as a DH, insofar as I can effectively isolate it. The designated hitter benefits from the DH rule by gaining playing time, so I plan to exclude them from the adjustment.

I plan to make the adjustment by giving the player some amount of win shares per game, as for example .8 or 1.0 win share/162 games, which I will add in on a seasonal basis. When I do the calculation, I'll exclude DH games from the player's total for purposes of the adjustment. This method will require more bookkeeping, but not an unduly burdensome amount.

The difficult part will be determining the size of the adjustment, particularly if it turns out that it should not be "one size fits all." The appropriateness of a simple adjustment that is the same, say, for Omar Vizquel and for David Ortiz: that, for me, is the thorny question right now.

For WARP, I plan to use WARP2 batting numbers for post-1972 AL players.
   161. John (You Can Call Me Grandma) Murphy Posted: August 28, 2006 at 09:23 PM (#2159648)
Many of us already are making tweaks to WS, so this AL/NL adjustment simply seems in line with those. So, I suppose I advocate leaving raw WS alone (because its nice that it 'adds up'), but perhaps reporting the percentage of time spent as an AL non-DH so that we can all make our own adjustments.

I like that idea, David.

The thing I like about WS is the "tweakiness" of it. I don't think anybody would claim that WS is perfect (I know I wouldn't), but the system can be manipulated to correct flaws, unlike other systems.
   162. sunnyday2 Posted: August 28, 2006 at 10:15 PM (#2159673)
I like that every player everywhere any time had the opportunity to help his team get 3 WS per game. If some players can participate in 3 and others 3.3 and others some other number, that is not "fair to all eras," in my book.
   163. Sean Gilman Posted: August 29, 2006 at 08:18 AM (#2159993)
My plan is to create "NL-equivalent" batting win share adjustment for all AL batters I evaluate from the DH era but to exclude from those adjustments their batting as a DH, insofar as I can effectively isolate it. The designated hitter benefits from the DH rule by gaining playing time, so I plan to exclude them from the adjustment.

Do I follow correctly:

Say Frank and David each create 30 raw Batting Win Shares. Frank played 50 games at DH and 100 games at 1B while David played 100 games at DH and 50 games at 1B. Your adjustment would then give Frank twice as many "bonus" wins shares as David, say Frank adjusts to 34 WS and David to 32 WS. Effectively you're saying Frank's batting is 2 WS more valuable than David's because of the position Frank played when he wasn't batting.

Additionally, Edgar created 30 raw Batting Win Shares while playing 150 games at DH. He gets no adjustment and is therefore 4 batting Win Shares less valuable than Frank because of the position he played.

This seems, on face, to be unfair to players who happened to have played in the DH era (the playing time rationale for not giving a bonus to actual DHs being particularly egregious: as if there's a HOM-calibur DH who wouldn't have played the field in a non-DH world) and counter to a basic principle of Win Shares: that batting is batting and fielding is fielding and irrelevant to a player's batting value is the position he plays when he's fielding.
   164. Chris Cobb Posted: August 29, 2006 at 12:15 PM (#2160015)
The example is correct, except that the size of the adjustment is far larger than any actual adjustment would be.

A realistic outcome of the adjustment would be

Frank 30.6 bws
David 30.3 bws
Edgar 30.0 bws

to add the relevant comparsion, Manny, who earned 29 batting win shares while playing 140 games in the outfield, gets a bonus that raises him to 29.9 batting win shares.

And we should remember, that Albert played 150 games at first base in the NL, and he earned 31 bws for a performance that would have been statistically equal to Frank's, if he had done it in the AL. The point of the adjustment is to diminish the gap in _measured merit_ between the two players of equivalent individual performance whose _win-value_ is slightly different because they are playing under different rules. If Manny played 140 games in the field in the AL, then its clear that he would have been able to play 140 games in the field in the NL, so to say that a hitter of equal performance in the NL is slightly more worthy of induction into the HoM merely because he played in the NL is absurd. The bws adjustment is necessary for that reason.

On the other hand, I'm not prepared to say that a player who played 150 games at DH would have played 150 games in the field in the NL. I'm certainly willing to allot that player merit for all the value he created as DH, but I'm not willing to assume that his batting merit would have been equivalent in the NL, because I don't believe his playing time could have been equivalent.

This seems, on face, to be unfair to players who happened to have played in the DH era (the playing time rationale for not giving a bonus to actual DHs being particularly egregious: as if there's a HOM-calibur DH who wouldn't have played the field in a non-DH world) and counter to a basic principle of Win Shares: that batting is batting and fielding is fielding and irrelevant to a player's batting value is the position he plays when he's fielding.

Is there a HOM-calibre DH whose career was not obviously lengthened, both in total seasons and in playing time within seasons, by the DH? I am in no way convinced that there is any long-time AL DH whose batting career would have developed similarly in the National League, so I don't see it as in any way egregious to assess the DH's merits purely within the context of the DH-based game that enabled the player to have the career he did.

The National-League equivalent I am proposing converts an AL player's win-share batting value into the batting value his performance would have achieved in the NL. There is no NL-equivalent for a DH, so I will not, in that case, create an NL-equivalent value. The theoretical separation of batting value from fielding value in win shares is irrelevant to the question of this adjustment, because the adjustment depends upon _playing time_, and, in the National League, the capacity of a player to play in the field _is highly relevant_ to his actual batting value, because in the NL players sit on the bench even when they are capable of hitting if they are incapable of playing the field. When a player is DHing in the AL, we have no way to know for certain how capable he is of playing the field, so I think it inadvisable to translate his batting performance into NL conditions.
   165. DanG Posted: August 29, 2006 at 01:12 PM (#2160053)
but to exclude from those adjustments their batting as a DH, insofar as I can effectively isolate it.

Chris: You may know this already, but the annual TSN Baseball Guide has detailed stats for every player as a DH. I own the guides from the late 60's forward, so if you or anyone would like any of this data, I can post it.
   166. Dr. Chaleeko Posted: August 29, 2006 at 01:52 PM (#2160105)
DavidFoss,

I think you are exactly right to compare the DH vs SP innings effects. The more I thought about the comparison, the more logical it seemed to me. I guess I just needed to foam at the mouth a while about it until I could get a grip on it. Meanwhile Chris Cobb says...

The difficult part will be determining the size of the adjustment, particularly if it turns out that it should not be "one size fits all." The appropriateness of a simple adjustment that is the same, say, for Omar Vizquel and for David Ortiz: that, for me, is the thorny question right now.

Right, and my concern here is that you'll find that, especially in WS, you have to readjust team-by-team for the DH because each teams DHs produce a different amount of marginal runs. I hope you don't have to go to that length! Fortunatley I think that retrosheet's splits ought to at least give you a running head start on partitioning player's values between DH and Fielder. Er, not Cecil Fielder.


OK, two more issues to ask about, one kind of easy, one that might be really annoying.
1) EASIER: Do NL players who DH in Interleague games get some kind of backmasked version of the DH adjustment? Since we adjust to the NL, they must for the six or seven games they play at DH (which is probalby as many as Manny will play at DH a year in the Big Papi Strikes Again! Era).
1a) Or do NL players who do not DH in interleague games get the upward adjustment just for those games?
1b) My head is spinning.

2) ANNOYINGER: What about AL pitchers? How is their ERA+ or WS or WARP affected by the DH, and should they get some kind of adjustment for playing in the DH league?
2a) With similar cavaets for interleague play as above?
2b) And also wouldn't we have to partition AL/Int starts for AL starters since they wouldn't get the benefit of an adjustment for NL park games?

Oyyyyy.
   167. rawagman Posted: August 29, 2006 at 02:56 PM (#2160175)
These issues speak to the core of my reluctance to use uber-stats.
As far as I can tell, the uber-stats were created to act as a BS-dump for all of the traditional stats.
Yet, the uber-stats have managed to create a whole new breed of BS that needs dumping.
Kinda defeats the purpose, doesn't it?
   168. Chris Cobb Posted: August 29, 2006 at 03:27 PM (#2160217)
These issues speak to the core of my reluctance to use uber-stats.
As far as I can tell, the uber-stats were created to act as a BS-dump for all of the traditional stats.
Yet, the uber-stats have managed to create a whole new breed of BS that needs dumping.
Kinda defeats the purpose, doesn't it?


Not in the least.

The comprehensive metrics still give us the most accurate overall assessments we have available of the value of a player to his team in a given season. This is what win shares tells us, and it does a good, though not perfect job, of representing this value.

The DH problem is one of deciding how _merit_ is related to value, and how to assess the relative merit of players who are competing under different conditions. No statistic that measures prodution or value can provide an answer to that question without adjustments being applied to it. But those adjustments can be made. As we are taking _quantitative_ approaches to figuring out what adjustments are appropriate, the analogy between these adjustments and the "BS dump," as Bill James called it, does not hold, because the BS dump is precisely the place where unquantifiable factors get applied.

BP's WARP3, in fact, offers us a comprehensive metric that already includes the sort of adjustments we are talking about making to win shares to adjust for the effects of the DH.

For any stat we might use, the question for us is: how much are we going to sweat the details in accounting for all the factors that may influence the statistical record such that the statistic gives us a misleading view of the relative merits of players. The comprehensive metrics sweat more details in the process of calculation than do any other stats that we have, so they give us a better starting point for making additional necessary adjustments.

Making careful use of the comprehensive metrics leads to much better result, I would argue, than putting a lot of weight on, say, black ink.
   169. rawagman Posted: August 29, 2006 at 03:45 PM (#2160245)
touche
   170. DavidFoss Posted: August 29, 2006 at 03:50 PM (#2160249)
1a) Or do NL players who do not DH in interleague games get the upward adjustment just for those games?

This reminds me why I hate interleague play! To be consistent with pre-1973, then they would. It would be a small percentage of a small bonus, so it might be negligible. Same with AL players who play in non-DH games. They get no bonus for that. Again if you just report how many games are played as a non-DH in a DH-games -- and then use that for your bonus. (5-10% depending on how you incorporate PH's).

Anyhow, no need to worry about this secondary adjustment until the 2000's elections.

What about AL pitchers? How is their ERA+ or WS or WARP affected by the DH, and should they get some kind of adjustment for playing in the DH league?

I'm pretty sure this is a big no. From the pitcher's standpoint, they can't tell if the extra offense is coming from a 9th hitter or just a random change in context due to era, parks, etc.

(I know there is some pitchers batting correction in pWS which I have periodically forgotten, but I thought it was to adjust a pitchers context relative to his teammates. No pitchers batting means that adjustment washes out, no?)
   171. Mark Shirk (jsch) Posted: August 29, 2006 at 05:41 PM (#2160414)
I am with dave on the pitchers, playing in a DH league should already be accounted for in the league offense level adjustment.

I also agree taht interleague play is the spawn of the devil. I can't even get excited about Yankees/Mets, Yankees/Phillies (favorite vs. hometown) or Yankees/Padres (favorite vs. brother's favorite). But I guess that isn't really germane to this thread huh?
   172. sunnyday2 Posted: August 29, 2006 at 05:43 PM (#2160418)
>>These issues speak to the core of my reluctance to use uber-stats.
As far as I can tell, the uber-stats were created to act as a BS-dump for all of the traditional stats.
Yet, the uber-stats have managed to create a whole new breed of BS that needs dumping.
Kinda defeats the purpose, doesn't it?

>Not in the least.

Both right.

Bill James, inventor of WS, has a huge bullshirt dump as an integral part of his system.

The way I see it, just figure everybody's WS the same way. If NL hitters have a 2.5% or 12.5% or 100% advantage, fine, work it out in the bullshirt dump. If pitchers today can't pitch 300 innings (hell, they can barely pitch 200 innings), work it out in the bullshirt dump.

But figuring out new and different adjustments for every nuance, no, it just creates new and different inequities, IMO.
   173. Dr. Chaleeko Posted: August 29, 2006 at 06:16 PM (#2160474)
Thanks for the ideas on pitchers viz the DH. My one remaining question would be whether AL starting pitchers and NL starting pitchers since the advent of the DH retain similar durability profiles. Does the presence of an extra hitter fatigue pitchers more quickly and cause them to lose their edge more quickly than in the NL where the starter gets a nice breather in the second/third and fourth/fifth innings? Or does the presence of more pinch hitting in the NL make it a wash? Or maybe does the free substitution of pitchers in the AL actually lead to shorter outings than in the NL? And is there enough difference to merit caring one way or the other?
   174. jimd Posted: August 29, 2006 at 06:47 PM (#2160519)
The initial impact of the DH on AL starters was MORE innings.

Back then, it was common for a starter who was behind to be lifted late for a PH, not ineffectiveness. The DH removed that necessity.

***

There is one impact of the DH on pitchers that should be directly measurable in the current game.

AL pitchers face the DH for the whole game. NL pitchers face a mix of the opposing starter and a collection of PHs. However, this is not evenly distributed. NL starters tend to face the opposing starter, while NL relievers tend to face the PHs. This should slightly benefit NL starters relative to AL starters (as measured by ERA+ or other league normalized stats), while it would slightly hurt NL relievers relative to AL relievers. However, I don't know the magnitude of the effect.
   175. Sean Gilman Posted: August 29, 2006 at 06:54 PM (#2160530)
Is there a HOM-calibre DH whose career was not obviously lengthened, both in total seasons and in playing time within seasons, by the DH? I am in no way convinced that there is any long-time AL DH whose batting career would have developed similarly in the National League, so I don't see it as in any way egregious to assess the DH's merits purely within the context of the DH-based game that enabled the player to have the career he did.

If that's your standard for not including DHs in the adjustment, then I don't see how you can include anyone. Is there a HOM-calibre hitter who played in the NL over the last 30 who did not have his career lengthened (at least in season) by the DH?

When a player is DHing in the AL, we have no way to know for certain how capable he is of playing the field, so I think it inadvisable to translate his batting performance into NL conditions.

It's the assuming the negative here, when we have no way of knowing one way or the other, that rankles me. It's not Frank and David and Edgar's fault they lived in a DH world, I don't think they should be punished for it.
   176. Sean Gilman Posted: August 29, 2006 at 06:56 PM (#2160535)
<strike>. . . . who played in the NL over the last 30 who . . . .</strike>

. . . . who played in the AL over the last 30 years who . . . .
   177. andrew siegel Posted: August 29, 2006 at 07:22 PM (#2160556)
When did they ban player moves between leagues? The longevity of players in both leagues should be equally effected by the DH Rule.
   178. Joey Numbaz (Scruff) Posted: August 29, 2006 at 07:33 PM (#2160564)
"(I know there is some pitchers batting correction in pWS which I have periodically forgotten, but I thought it was to adjust a pitchers context relative to his teammates. No pitchers batting means that adjustment washes out, no?)"


The pitcher's batting correction is for pitchers that hit below the level of 'marginal' offense. So any pitcher, after adjusting for park that creates runs at a rate less than 1/2 of the league average has his pitching WS 'docked'. It's not an adjustment relative to teammates, it's relative to the league.

There's a minor side issue that stems from this - the batting WS of the entire team essentially absorb the hit for all players that are below the marginal offense level.

Example: Team creates 200 marginal runs. Individual marginal runs totals . . . we'll assume the team only has 5 hitters in this league for simplicity.

Player 1 120
Player 2 50
Player 3 30
Player 4 20
Player 5 -20

The team gets 200 marginal runs towards their WS. Player A should get 60% of 200. However he actually only gets 120/220 or 54.5% of his team's marginal runs. That's because player 5's -20 is zeroed out so there aren't negative WS for any player. But the offense is still only credited with 200 marginal runs (which is compared to the defense (pitching+fielding) marginal runs and wins to determine actual Win Shares).

This is a not insignificant flaw in the system. I don't think it's a deal breaker, because every team has some submarginal players, so it's not really biased against anyone except players that play on truly awful teams - or have terrible hitting pitchers (relative to the league average that year). But it's definitely something to be aware of.
   179. Joey Numbaz (Scruff) Posted: August 29, 2006 at 07:37 PM (#2160567)
"If that's your standard for not including DHs in the adjustment, then I don't see how you can include anyone."


I agree - I would include the DHs in the offense adjustment, but they get their 'demerit' on the fielding side, where they have no value at all. But I do think DHs should be included in any AL offense adjustment.
   180. Joey Numbaz (Scruff) Posted: August 29, 2006 at 07:40 PM (#2160569)
"The initial impact of the DH on AL starters was MORE innings."


Just to nip it in the bud in case anyone is wondering, my pitching Pennants Added do account for this, since seasonal innings are translated based on the league leaders for that season. If the DH were causing AL pitchers to throw more innings, that would show up when comparing the totals of the league leaders in each league, so I am catching this.
   181. TomH Posted: August 30, 2006 at 01:41 AM (#2160840)
lotta people gettin down (not in a good way) on Joe Sewell lately.

Joe Sewell had a career OWP of .549. A team of Sewells, playing every position with avg pitching and defense, would have a record of 85-69 each year. Of course, that's pretty good when your shortstop can hit that well. Sewell's OWP is higher than some bum named Ripken (no, not Billy...).

Name one shortstop that we have not elected to the HoM who played as many games as he did, and had a career OWP within 50 points of Sewell's (through 1978 stats, our current year).

Bzzz. None.

Every other shortstop on the ballot has an OWP of under .500, OR played fewer games than Sewell. Or, in most cass, BOTH.

This may be true of other people on our ballot; but most of them would be less-than-stellar in the field. This guy was a grade A- shorstop! How much extra should he get for his fielding.? If he saved an average of only 8 runs a year, that would take his "effective" OWP up to over .600.

And the best part of all of this is: it doesn't change the numbers if you add back in a few guys who were excluded from the majors at the time. I think it's time to stop concentrating on a few shortstops who did't play in Sewell's day, and look at what DID happen; a man played a key position, hit well, fielded well, had a good career length (and yes, got moved off of shortstop for the last 5 years, but some of that was a team with a choice of riches, much like A-Rod and the Yankees).

He'll be at the of my ballot next "year", assuming 2 certain guys get elected.

1876-1978 Shortstops with at least 1200 games played
OWP OWP G
1 Honus Wagner . .749 2792
2 Arky Vaughan . .692 1817
3 George Davis... .614 2368
4 Hughie Jennings .606 1285
5 Lou Boudreau . .603 1646
6 Joe Cronin ...... .583 2124
7 Luke Appling .... .577 2422
8 Jack Glasscock . .576 1736
9 Vern Stephens .. .572 1720
10 Jim Fregosi ..... .565 1902
11 Bill Dahlen ... .559 2443
12 Johnny Pesky . .555 1270
13 Ed McKean ..... .553 1654
14 Joe Sewell .... .549 1903
15 Harvey Kuenn ... .547 1833
16 Cecil Travis ..... .546 1328
17 Kid Elberfeld ... .540 1292
18 Bobby Wallace ... .522 2383
19 Woodie Held .... .520 1390
20 Rico Petrocelli . .519 1553
21 Denis Menke .... .512 1598
22 Eddie Joost .... .507 1574
23 Pee Wee Reese .504 2166
24 Travis Jackson . .503 1656
25 Monte Ward ..... .499 1825
26 Dave Bancroft . .498 1913
27 Art Fletcher ... .496 1533
28 Al Dark ......... .494 1828
29 Phil Rizzuto ... .494 1661
30 Freddy Parent . .493 1327
31 Joe Tinker ..... .487 1803
32 Woody English . .483 1261
33 Bert Campaneris .480 2043
34 Johnny Logan ... .477 1503
35 Maury Wills ..... .473 1942
36 Herman Long ... .471 1874
37 Dick Bartell ... .468 2016
   182. sunnyday2 Posted: August 30, 2006 at 02:19 AM (#2160896)
Two words: Dobie Moore.

This list also fairly screams Vern Stephens.
   183. KJOK Posted: August 30, 2006 at 04:42 AM (#2161090)
Joe Sewell had a career OWP of .549. A team of Sewells, playing every position with avg pitching and defense, would have a record of 85-69 each year.

I'd argue this under-states his case, as I would use the OWP of the League SS's (or NL and AL SS's) as the "Loss" side instead of .500 since a team with Sewell playing every position would be 'wasting' his defensive skills, so his "RECORD" then becomes, instead of 85-69:

vs. AL 100-54
vs. AL/NL 98-56
   184. TomH Posted: August 30, 2006 at 11:32 AM (#2161211)
1. Dobie Moore very likely DID have a better peak than Sewell. If you're a peak or preak/prime guy, go ahead and vote for him. Just don't knock Joe.
2. Take away about 80 career runs from Stephens relative to Sewell for defense, and Vern loses on both career length AND effectiveness.
   185. sunnyday2 Posted: August 30, 2006 at 12:22 PM (#2161221)
TomH, it's your list, not mine, and Stephens is #9 and Joe #14, and Stephens played 120 more games at SS, and as a SS impressed at least some people more than Johnny Pesky who is #12. I'm not knockin' Joe, he's been around #20 on my ballot forever, but I just don't see how he leaps forward among your list of SSs, 4 of whom are eligible and ahead of Joe on, like I said, your own list. (Stephens isn't on my ballot either, BTW, which is sorta my point.)
   186. Dr. Chaleeko Posted: August 30, 2006 at 01:13 PM (#2161253)
10 Jim Fregosi ..... .565 1902
13 Ed McKean ..... .553 1654
14 Joe Sewell .... .549 1903
15 Harvey Kuenn ... .547 1833
16 Cecil Travis ..... .546 1328



TomH,

How do McKean, Kuenn, and Travis (with war credit) grab you?
   187. TomH Posted: August 30, 2006 at 01:38 PM (#2161277)
"ahead of Joe" - well, yes, I did sort it by OWP, by I also listed games played, since John ("3 for 3") Paciorek comes out higher in career OWP than Babe Ruth, I kinda thought that was important.
If I convert OWP and G to equiv RCAA by the formula games/2*(sqrt(OWP/(1-OWP))-1), and then add is FRAA from the BP cards, here is a snippet of the previous table:
player .............. OWP G... RCAA FRAA RCAA+FRAA
Vern Stephens .. .572 1720 134 ..25 ........159
Jim Fregosi ...... .565 1902 133 -43 ...........90
Johnny Pesky .. .555 1270 ..74 .48 ..........122
Ed McKean ...... .553 1654 ..93 -166 ....... -73
Joe Sewell ...... .549 1903 ..98 121 .........219
Harvey Kuenn . .547 1833 ..91 -83 ............8
Cecil Travis ..... .546 1328 ..64 .29 ..........93

Yes, give Pesky and Travis some war credit, and they are in Stephens' ballpark. Not Sewell's. I'll point out the RCAA, wiht "above average" being the salient measure, hurts the guys with longer careers, since most of us agree that "average" has value.

Any questions, class?? :)
   188. Mark Shirk (jsch) Posted: August 30, 2006 at 01:49 PM (#2161285)
I like Rizzuto more than Sewell as well. While Rizzuto definitely wasn't the hitter that Sewell was, he was a better fielder and is missing 2-3 prime seasons because of the war. This would raise his OWP over .500, I would presume, and he would have as many or more games total, and many more at SS, than Sewell. Rizzuto also has the higher peak. They are close and within 8-10 spots on my ballot but I prefer Scooter.

And why not Fregosi? He has the higher OWP by 16 games with more games at SS and one fewer game overall? Sure Sewell was the better fielder, but if anything, Tom's list shows me that historically he wasn't anything special. Cases could be made for Stephens, Pesky, Fregosi, and Rizzuto and Travis with war credit.

I also wish we would not rely so much on career rate stats in our arguments. Career counting stats fine, career counting/rate combination stats (uberstats, VORP, RC, etc.) fine. They both show a long and productive career if that is your thing. But career rate stats dont' necessarily show this and don't show how a player got those stats. Pennants aren't won over a career but in seasons, if we want to use rate stats we should use both career and in-season. I dont' know if this helps Sewell at all but the fact that his peak isn't that distinguishable from Dave Bancroft (with a defensive adjustment)is one reason he hasn't been in my top 30 for decades.
   189. Mark Shirk (jsch) Posted: August 30, 2006 at 01:53 PM (#2161291)
I think that WARP vastly overrates Sewell's fielding (I don't buy that he had a rate of 111 before being moved to 3B for instance, it just doesn't pass the smell test), what do Total Baseball and WS have to say about this? Any other fielding metrics?
   190. TomH Posted: August 30, 2006 at 02:37 PM (#2161349)
average of best 6, 9, 12, 15 years by Win Shares

player ...top6 top9 top12 top15
Sewell... 25.5 24.0 22.0 19.8
Fregosi.. 27.5 24.1 20.1 17.1
Stephens 27.3 25.0 21.4 18.9
Bancroft. 24.5 22.8 20.3 17.9

I didn't run the Scooter or the other guys who would require speculating how much WWII credit to give.

Stephens had his best and 3rd-best years during WWII; we'd all agree some credit ought to be deducted for this. Sewell and Stephens only played for 14 yrs.
   191. Mark Shirk (jsch) Posted: August 30, 2006 at 03:12 PM (#2161396)
To be fair Tom, I do have Sewell ranked above that group of players but I am not sure if any of those other players are in my top 40 and Sewell isn't that much ahead of them. I just don't see how he seperates himself that much from those guys, enough to have a high ballot spot for a peak/prime voter like myself.

I guess with Rizzuto it comes down to how much WWII and MiL credit you are willing to give. I give him a decent amount and it pushes him ahead of that group and Lundy. Dobie Moore is certainly the best peak candidate at SS.
   192. KJOK Posted: August 30, 2006 at 05:32 PM (#2161577)
Sewell's Runs Created Above Position is 346. Sure, NgLgs players are an issue, but that issue is the same for EVERYONE before 1950.

NAME           RCAP
Gabby Hartnett    364
Frank Baker    364
Al Simmons    356
Hank Greenberg    347
JOE SEWELL    346
Sherry Magee    337
Duke Snider    334
Jackie Robinson    323
Earl Averill    321
Larry Dobby    308
Bill Terry    302
Zach Wheat    300
Billy Herman    298
Frankie Frisch    291
Lou Boudreau    274
Billy Williams    273
Joe Medwick    267
Hughie Jennings    263
Joe Gordon    259
Goose Goslin    257
Roberto Clemente247
Bob Caruthers    243
Enos Slaughter    237
Bobby Doerr    234
PeeWee Reese    223
Ernie Banks    216
Henie Groh    216
Roy Campanella    206
George Sisler    205
Ron Santo    200
Charlie Bennett    196
Bobby Wallace    195
Monte Irvin    151
Jimmy Collins    148
Richie Ashburn    146
Jimmy Sheckard    135
Max Carey    49
   193. Joey Numbaz (Scruff) Posted: August 30, 2006 at 08:03 PM (#2161947)
I think that WARP vastly overrates Sewell's fielding (I don't buy that he had a rate of 111 before being moved to 3B for instance, it just doesn't pass the smell test), what do Total Baseball and WS have to say about this? Any other fielding metrics?


They moved him off SS for a guy that had a 114 rate the year before and a 116 rate the year of the move (Jackie Tavener).

And at the time, 3B was considered a very defensive position - it wasn't like moving a guy from SS to 3B today, it was like moving a guy from SS to 2B, if anything 1925-35 3B hit worse than today's 2B.
   194. Chris Fluit Posted: August 30, 2006 at 08:34 PM (#2161976)
Comparing Hall of Fame and Hall of Merit by position:

<u>Catcher</u>
Hall of Fame/Hall of Merit
Berra, Campanella, Cochrane, Dickey, Ewing, Gibson, Hartnett, Mackey, Santop

Hall of Fame/not Hall of Merit eligible
Bench, Carter, Fisk

Hall of Fame/not Hall of Merit
Bresnahan, Ferrell, Lombardi, Schalk

Hall of Merit/not Hall of Fame
Bennett
   195. jimd Posted: August 30, 2006 at 08:54 PM (#2162007)
(I don't buy that he had a rate of 111 before being moved to 3B for instance, it just doesn't pass the smell test)

I though I explained this before, and you didn't comment on it. I'll repeat it again.

It would appear that Sewell was replaced by a duo with even better fielding stats (according to WARP). Jackie "Rabbit" Tavener (117 rate) and then Ray Gardner (112 rate). Tavener had been Detroit's regular SS of the previous 4 seasons, and was acquired in the offseason along with Ken Holloway (spot starter and reliever) for George Uhle, who had a couple of All-star seasons left (in retrospect, a steal for Detroit).

Tavener hit like Carl Lind, the Cleveland 2b (both had .249 EQA in 1928). Shifting Sewell to 3rd and Johnny Hodapp to 2nd improved the defense at all three positions (Hodapp 91 at 3rd, Lind 92 at 2nd), while not hurting the offense. Or so the Cleveland GM probably thought. It didn't work out well because Tavener didn't hit a lick (.204 EQA in 1929). Neither did Gardner, a minor league callup.
   196. DL from MN Posted: August 30, 2006 at 09:03 PM (#2162017)
Stephens had 7 seasons with >140 games played. Sewell had 9. Stephens has 5 seasons above 7 WARP, Sewell has 8. Stephens wasn't known for in-season durability.

I believe WARP that Sewell was better with the glove.
   197. John (You Can Call Me Grandma) Murphy Posted: August 31, 2006 at 12:47 PM (#2162862)
I have a question about Torre's eventual plaque: what cap should he get? I'm leaning toward the Braves, but the Cardinals have an excellent case themselves. He packed a lot of value in fewer seasons in St. Louis. Of course, he was hardly catching by then.

Any feedback would be greatly appreciated. This might be the toughest cap to figure out.
   198. sunnyday2 Posted: August 31, 2006 at 01:14 PM (#2162887)
Yankees, of course ;-)
   199. John (You Can Call Me Grandma) Murphy Posted: August 31, 2006 at 02:06 PM (#2162948)
Yankees, of course ;-)

Heh.

BTW, I'll tally up serious (ahem! :-) and unbiased choices on Tuesday and go with the consensus. Will it be Milwaukee or St. Louis?
   200. DanG Posted: August 31, 2006 at 03:12 PM (#2163030)
Torre should have a Braves cap. By WARP, his best year was as a Brave in 1966. Win shares prefers 1971, but he still had more total WS as a Brave than a Cardinal (155 to 142). He was on more all-star teams as a Brave (5 to 4). He's not a clear HoMer if he wasn't a catcher, and he didn't catch much for StL.

His raw stats with the Braves would look much more impressive but for the era (1963-68) and the ballpark (1961-65). That's what really kept him out of the Coop as a player.
Page 2 of 3 pages  < 1 2 3 > 

You must be Registered and Logged In to post comments.

 

 

<< Back to main

BBTF Partner

Support BBTF

donate

Thanks to
Kiko Sakata
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Syndicate

Page rendered in 1.3915 seconds
49 querie(s) executed