Baseball for the Thinking Fan

Login | Register | Feedback

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

Sunday, August 21, 2005

1959 Ballot Discussion

1959 (September 5)—elect 2
WS W3 Rookie Name-Pos (Died)

338 98.9 1936 Johnny Mize-1B (1993)
287 91.0 1940 Bob Elliott-3B (1966)
233 89.2 1934 Dutch Leonard-P (1983)
237 80.9 1934 Bobo Newsom-P (1962)
223 73.7 1939 Bill Nicholson-RF (1996)
191 66.7 1943 Eddie Stanky-2B (1999)
173 63.7 1943 Harry Brecheen-P (2004)
147 52.8 1939 Joe Dobson-P (1994)
145 49.2 1941 Jerry Priddy-2B (1980)
146 43.9 1939 Barney McCosky-CF/LF (1996)
132 46.2 1938 Max Lanier-P (living)
103 42.4 1946 Ewell Blackwell-P (1996)
114 33.9 1943 Connie Ryan-2B (1996)
107 33.7 1942 Johnny Lindell-CF/LF (1985)
102 32.0 1943 Billy Johnson-3B (living)
086 24.3 1941 Pat Mullin-RF/LF (1999)
063 15.0 1915 Luke Easter-1B (1979)

1959 (August 28)—elect 2
HF% Career Name-pos (born) BJ – MVP - All-Star

HF 27-53 Satchel Paige-P (1906) 2 – 1*
HF 33-53 Ray Dandridge-3B (1913) #1 3b 0 - 1*
52% 32-56 Wild Bill Wright-OF (1914) #4 rf 0 - 4*
00% 37-53 Dave Barnhill-P (1914)1 - 1*

Players Passing Away in 1958
Age Elected

70 1934 Tris Speaker-CF
49 1952 Mel Ott-RF

Age Eligible

89 1910 Jack Doyle-1b
83 1914 Luther “Dummy” Taylor-P
74 1921 Red Murray-RF
61 1936 Ike Boone-RF
61 1938 Lu Blue-1B
59 1939 Bernie Friberg-3B/2B
53 1949 Chuck Klein-RF
48 1950 Frank Demaree-RF/CF
45 1953 Mort Cooper-P
39 1957 Snuffy Stirnweiss-2B

My compliments to Dan and Chris for the lists!

John (You Can Call Me Grandma) Murphy Posted: August 21, 2005 at 11:18 PM | 161 comment(s) Login to Bookmark
  Related News:

Reader Comments and Retorts

Go to end of page

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

Page 2 of 2 pages  < 1 2
   101. Dr. Chaleeko Posted: August 26, 2005 at 08:38 PM (#1575814)
One man's preliminary thoughts and agonies.... Yes, I'm strongly considering a vote for both Easter and Trouppe, and I'd like feedback on it if anyone has reactions.

1. Paige
2. Mize
3. Duffy
4. Oms
5. Medwick
6. Mendez
7. Cravath
8. Matlock
9. Walters
10. Ferrell
11. Burns
12. Browning
(13a. Hack)
(13b. Herman)
(13c. Lyons)
13. W Brown
14. LUKE EASTER: See, depending on how I'm viewing Easter's pre-NNL credit scenario, I'm voting for him anywhere between 3 and off-ballot. If, in fact, careers are in essence parabolic in nature, then Luke Easter's 1941-1946 seasons (ages 25-30) would amount to something like 170 WS of value...oh, and it would include his baseball peak too. In other words, if true to Bill James's wisdom that the vertex of the parabola that is Easter's career falls on his age-27 season, then my initial estimation has Easter showing up this way from 1941-1946

1941 25  15 (half-season for injury)
1942 26  32
1943 27  33
1944 28  32
1945 29  30
1946 30  27
total   169

Add that to the combination of the standard MLEs and the MLB WS (243 WS) and he's coming out around 410 WS.

Now maybe that's too rosey-eyed. Let's say that he earns half that total, that's still 243+85=328 WS.

I think the truth is probably in between those two perspectives, something like 375 career WS. A bigger question would be would he earn them in batches (for peaksters) or in a spread?

We've already accounted for 1941's injury. 1946, Gadfly tells us, was a big year for Easter. So let's say that's a full credit year based on the parabolic model.

STatistically, a player is most likely to peak at 27, so let's assume Easter does: he hits his estimated peak at 27 and the remainder of the xCR is divided roughly equally among the remaining three seasons, thus

1941 25  15 (half-season for injury)
1942 26  20
1943 27  33
1944 28  20
1945 29  19
1946 30  27
total   134 

Career: 134 + 243 = 377
Any 3 peak: 40, 33, 27 = 100
Any 5 peak: 40, 33, 27, 23, 23 = 146
Any 10 prime: 40, 33, 27, 23, 23, 22, 21, 20, 20, 20 = 249

This is not adjusted for 162.

Once I adjust Easter to 162, he slots in a little behind Suttles and McGwire, ahead of Tony Perez. I know it's a TON of speculation, but in a way it isn't. He was constrained in 1941 and 1946 by his geographic location, but was playing on highly competitive teams (per Gadfly), and in war, he was essentially bound to working in essential industry.

15. QUINCY TROUPPE: Same thing that went for Easter goes for Trouppe. Those 1933-1937 seasons are up in the air, yet we know he was playing ball at the highest level because not only was he in ND with Satch and the gang, but he was also playing with the Monarchs aside from that. He was a roughly league-average hitter in 1932 at age 19 and in 1938 at age 25. If he's around 10-15 WS for those five years, he's adding bulk to his caeer and probably ending up around 350-360 WS (adj to 162). That would put him a little behind Gary Carter on my catching depth chart, but ahead of Joe Torre.

16. Ruffing
17. Jennings
18. Averill
19. Rixey
20. Van Haltren

DAVE BARNHILLL: He's a wildcard and could yet appear on my ballot. He has really good numbers and deserves a long look.

LEON DAY: I didn't like him the first time around when he was named Hilton Smith.... but I'm open to more information.

BOB ELLIOT: I like him, but the peak looks a little soft to me. He's not too different valuewise than Bobby Bonilla, though a very different player. He's down in the 50s. But he's a lot better than Ray Dandridge.

RAY DANDRIDGE: Based on the available data and the MLEs, he may be one of the HOFs biggest misses. I just don't see how he could have been a productive major-leaguer at the plate without more speed, more power, or more walks. And at 3B, his glove, no matter how good, can't carry him enough to make him worth the HOM's while. Outside the consideration set.

BURNIS WRIGHT: Gary Matthews, per SDY2, looks like a great comp. Outside the consideration set.
   102. Chris Cobb Posted: August 26, 2005 at 08:44 PM (#1575837)
BOB ELLIOT: I like him, but the peak looks a little soft to me. He's not too different valuewise than Bobby Bonilla,

WARP sees the matter quite differently. He's a lot better than Bonilla as WARP sees it, and very close to Stan Hack. Elliott's plate discipline was somewhat better (hence a better EQA), and his defense was a lot better.
   103. Dr. Chaleeko Posted: August 26, 2005 at 08:47 PM (#1575842)
I should have said in my Elliot comments, that he's in the cluster that includes Gordon, Stephens, Sewell, and the like.
   104. Trevor P. Posted: August 26, 2005 at 08:53 PM (#1575858)

Thanks for reminding me about Elliott's outfield play.

Here's W/L records and winning percentages for the eleven seasons when both Elliot and Hack were full time players:

Elliott (40-51): 951-888, .517
Hack (34-45): 999-843, .542

With Hack's '32 and '33 thrown in, that's eight straight winning seasons for Hack to start his career. Lucky bastid.

Right now, I tend to buy WARP's view of Elliott's hitting (since looking at the raw stats I find it hard to account for a 13 Win Share difference), and I'm on the fence as to whether or not WARP or Win Shares is right about their fielding.

With respect to the winning percentages, how many win shares do you think that might account for in Hack's favor, over the course of his career?
   105. Howie Menckel Posted: August 26, 2005 at 10:34 PM (#1576007)
I used to vote for Roush sometimes, but even with the WW I adjustment, he missed an annoying number of games. I like OPS+ a lot, but you have to beware the high-end and low-end playing-time guys.

Lots of good stuff on Browning, pro and con. Thanks, fellas!

I also tend to see Elliott as quite similar to Hack, but have to factor in the OF issue.
   106. Chris Cobb Posted: August 27, 2005 at 02:19 AM (#1576586)
With respect to the winning percentages, how many win shares do you think that might account for in Hack's favor, over the course of his career?

If I'm running the numbers correctly, it's a negligible amount: 3 win shares. There's very little distorting effect for teams near .500. Hack's teams get an extra 30 win shares over these seasons, of which Hack gets about 10%.

Unless there's something weird happening in the interaction between Hack's and Elliott's big seasons and extreme team seasons, I think we can say that win shares' difference of opinion with WARP on Elliott's value relative to Hack is only very slightly influenced by team effects.

I just noticed another piece of the minutiae that may be influencing the difference between WARP and WS on the fielding side. The end of WWII is a shift-point for the win shares formula for 2B/3B win shares. Depression-era 3B, of which Hack is one, earn an average of 3.86 ws/1000 def innings, while post-war third basemen, of which Elliott is mostly one (a lot of his outfield play was in his early seasons) earn 3.27 ws/1000 def innings.

So even though Elliott's seasons as a regular only trail Hack's by 6 years, Hack as a B- third baseman earns 4.01 ws/1000 innings while Elliott as a B- third baseman earns 3.47 ws/1000 innings.

The change in the win share the formulas might account for about 5 win shares difference in the fielding evaluations of the two players.
   107. Kelly in SD Posted: August 27, 2005 at 10:25 AM (#1577469)
I give WWII credit based on a wighted average of the 3 seasons before and 3 seasons after. I give credit for blacklisting (C Jones) / minor league captivity (Averill, Cravath, Fournier) / skin color banning. I do not timeline. I think most league quality comparisons (AL vs. NL, NL vs. AA) are based on too small a comparison base so generally disregard those, though in extreme cases I do apply them. Win Shares voter and I consider the opinions of the people who saw them play when they are confirmed by the numbers and vice versa. Long careers that are slightly above average do the worst in my system. Catchers receive a bonus also.

1. Satchel Paige, long career at a high level against the highest quality opposing starters. He faced the opponent's best pitcher and he still won over 60%.
2. Johnny Mize, especially with the WWII credit for 3 years in the heart of his career.
3. Mickey Welch.
4. Luke Easter - that's giving credit from 1941 on through 1959. Unique player, unique circumstances. I haven't seen a reason to not give such credit.
5. Charley Jones. Top 5 player in NL before blacklisting. Top 5/10 player in AA after blackslisting, why should I discount?
6. Pete Browning. See previous posts this thread.
7. Charlie Keller. Top 5 player in AL in 1941, 42, 43, 46, plus WWII credit in 1944, 45. All-Star level in 2 other years. Jennings with 2 more MVP level years and 1 more All-Star year.
8. Wes Ferrell. Best AL pitcher of 1930s, but for Grove. Did not have the offensive or defensive support that Ruffing had.
9. Qunicy Trouppe. Based on 1938-1952 numbers and credit for starting ML career late in 1934, full time play in 1935-36. No credit for year off to box. (Like Too Tall Jones in 1979?). Bill James mentions he was an on all-star teams 23 different times between all the different leagues he played in.
10. Hugh Duffy. High peak and prime. Great defense. Underrepresented era.
11. Bucky Walters. No eligible pitcher was best in his league 3 times. Yes, one was in WWII NL and I discount that, but he was still the best that year.
12. Earl Averill. One year of minor league credit. Very good defender. Best CF b/t Speaker and DiMaggio. Perfect example of a player that my system loves: Long high prime (lots of 25+ WS seasons.)
13. Vic Willis. Forgotten pitcher of the Oughts.
14. Alejandro Ohms.
15. Willard Brown.
16. Burleigh Grimes.
17. Dobie Moore.
18. Cupid Childs.
19. Wilbur Cooper.
20 - 26. GVH / Jennings / Gordon / Burns / Mackey / Medwick /Leach. All these players are within a hair's bredth of each other.

Elliott is around the top 30. Third basemen (pre-Mathews) are the one position that my ranking formula has major discrepancies with the HoM. For every other position, it has identified every HoMer as all the top eligible performers (except those on my ballot.)

Ruffing is nowhere the pitcher Ferrell is. Had tremendous support (offensively, defensively, and home park) with the Yankees. Was used in a consistent fashion by the Yankees - never out of the bullpen during their big years. Compare that to Ferrell's use pattern. No better than Gomez during their time together.

Dandridge is nowhere near the ballot per the numbers from his thread, yet James has him the best third baseman in the Negro Leagues, the American Assoc voted him MVP his first full year there, and there is a lot of anecdotal/personal opinion about his greatness... what to do...
   108. TomH Posted: August 27, 2005 at 10:54 AM (#1577473)
career RCAA best seasons RCAA
Averill 391 .. 72 58 53 41
Roush.. 294 .. 43 40 38 31

And of course this doesn't include minor league credit for EA.
   109. sunnyday2 Posted: August 27, 2005 at 02:36 PM (#1577559)
I agree that Luke Easter is a unique player with unique circumstances. Given all of that and all of the discussion on his thread, I ended up with him at #78 and Kelly has him at #4!

I don't think there is another player for whom we have created quite as "fantastical" or "hypothetical" career, unless it's Buzz Arlett. But then you've got guys like Gavy Cravath, too, and, well, every NeLer has been given a hypothetical career under the heading MLE.

Generally I agree with the MLE conversions that Chris and Doc and David are providing. Generally I agree that if a player was on a baseball diamond regularly and routinely during some interregnum from an otherwise elite career, then credit is appropriate--e.g. Dobie Moore with the Wreckers, Cravath, etc.

The problem I have with Easter is that in order to make a candidate out of him we are not filling in the blanks in an otherwise HoM-like career, we are creating his peaks as well. I don't think there's another player, except Buzz Arlett, for whom we have done that.

(And Arlett is not in my top 100.)

Seeing Easter at #14 on Doc's ballot was one thing, seeing him at #4 is another. I just had to say, "Wow!"
   110. Mark Shirk (jsch) Posted: August 27, 2005 at 03:14 PM (#1577598)
Win Shares doesn't distort player's totals due to that player being on a winning or losing team. There may be a small strength of schedule adjustment (though for all non-Yankees and non- 1930's Phillies this is negligible and I don't use it anyway), but the distortion comes when a team is consistently above or below their pythag.

It would be much more constructive to see if Hack's or Elliot's teams consistently performed above or below their pythag, much more constructive than simply looking at whether or not they were winning or losing.
   111. Brent Posted: August 27, 2005 at 03:57 PM (#1577637)
I'm sorry Sunnnyday2, but I've got to respond to your swipes at Buzz Arlett. Given that you routinely accept and use NeL MLEs, I just don't understand how you can characterize the MLEs I've calculated for Arlett as "fantastical" or "hypothetical."

Unlike Luke Easter, Arlett's 19-year career is fully documented. All but a few games of Arlett's career were either in the highest minor leagues or the majors. The statistics for the top minor leagues such as the PCL are much more complete and reliable than anything available for the NeL (with the exception of the handful of seasons for which GaryA or others have compiled complete statistics). Furthermore, since players from the PCL, AA, and IL routinely moved to and from the majors, we have had ample opportunities to test the MLEs with major league players. I don't recall that anyone saying that my MLEs for DiMaggio, Lombardi, Averill, Berger, or Bob Johnson were fantastical or hypothetical.

It is well documented that Arlett's services were repeatedly sought by major league teams. The reasons that he didn't make it to the majors before 1931 are fully discussed on the Arlett thread, but the main reason is quite simple--Oakland didn't want to sell him, and as an independent team, they didn't have to. We as HoM voters are vigilant in treating NeL players fairly. How about a similar vigilance in providing fair treatment to a player who was kept from the majors only because he happened grew up in California and signed on with his local PCL team?

BTW, I share your concerns about Easter. If we based his case solely on his recorded statistics, they are very good for a player in his mid to late 30s and early 40s, but compared to a similar player like Ralph Kiner, he clearly falls short. As a peak voter myself, Easter's HoM case would have to depend on assuming a hypothetical peak, which I am very reluctant to do. It's sort of like Cecil Travis - if Travis had had seasons in 1942-44 like his age 27 season in 1941, he'd be a HoMer in my book. But while I do give Travis war credit, I'm not willing to give him that much credit.
   112. Ivan Grushenko of Hong Kong Posted: August 27, 2005 at 04:03 PM (#1577646)
John, why not a thread for Latino candidates generally or else title it as Cepeda, Coimbre, Vargas, Bragana, Duany and Garcia, or whatever. I agree that we ought to not let Latinos who didn't nec. play in the U.S. be missed.

Or, John, should we just go read the Latino thread on BP? Is there stuff there that would be helpful? What's the link?

You guys are probably way ahead of me on this but if anyone cares to comment on any of these guys who have not already been discussed, it would be interesting.
   113. sunnyday2 Posted: August 27, 2005 at 07:46 PM (#1577986)
Brent, my use of the terms fantastical or hypothetical just means were are projecting MLEs for players/seasons when they were not in the MLs.

My main point is to differentiate between hypothetical shoulder seasons versus hypothetical peaks. Adding shoulder seasons for Cravath and Averill is different than hypotheticizing peak seasons for Arlett or Easter.

The big problem of course is that we are hypothesizing peaks for all of the NeLers (when we create MLEs). But with NeLers there is no other alternative. We would be disregarding a whole class of player otherwise. With Arlett, he is the one and only and sole victim of my "rule" about how far to go with MLEs.

Except for Easter, then, whose hypothetical peak didn't even occur in the NeLs.

I doubt that that clears it up, of course....
   114. Chris Cobb Posted: August 27, 2005 at 08:09 PM (#1578026)
Win Shares doesn't distort player's totals due to that player being on a winning or losing team. There may be a small strength of schedule adjustment (though for all non-Yankees and non- 1930's Phillies this is negligible and I don't use it anyway), but the distortion comes when a team is consistently above or below their pythag.

As I and others have said and documented before, this is not true. Extreme team performance, whether good or bad, can influence player win share totals as much as large swings above or below Pythagorean projections.

Bob Johnson is the current poster-child for this effect. Some time in the next couple weeks I'll have a chance to run the numbers on him to document this rigorously.

It would be much more constructive to see if Hack's or Elliot's teams consistently performed above or below their pythag, much more constructive than simply looking at whether or not they were winning or losing.

I did that, actually, at the same time I did the calculation of the effect of team winning percentage, but I didn't post them because the results didn't matter much to the comparison.

Both Hack and Elliott played for teams that didn't do very well vs. Pythagoras while they were in their primes.

Hack's teams were 23 wins below, 1934-45.
Elliott's teams were 14 wins below, 1940-51.

This probably cost Hack about 7 win shares.
This probably cost Elliott about 4 win shares.

The biggest seasonal effects appear in Hack's career in 1940-41. Each year the Cubs missed their projection by 7 wins.

According to WARP1, Hack had an excellent year in 1938, an off year in 1939, and very good years 1940-42: 9.4, 5.4, 8.6, 8.3, 7.9

Win shares agrees on 38 and 39, but it handles 40-42 differently: 33, 23, 25, 30, 26.
   115. Dr. Chaleeko Posted: August 27, 2005 at 08:18 PM (#1578036)
Naturally, I posted Easter and Trouppe at 14 15 on my prelim in part because I think it's possible that's where they belong. But I'm hoping also to sharpen the discussion of them.

Is it best practice to give a 100% "bullshirt dump" to Easter's 1941-1946 seasons and only take his 1947-1959 MLE at its face value?

Is it best practice to give him some kind of career value without considering the missing peak value?

Is it best practice to ignore the ample research that suggests players peak around age 27, assume that Easter peaked in his 30s and let his case lie?
   116. Chris Cobb Posted: August 27, 2005 at 08:27 PM (#1578043)
I haven't given Easter much thought yet, but the player whose situation is closest to Easter (although there's a lot more data for him) is Gavvy Cravath.

Is Cravath a late bloomer, or is his peak concealed by league changes, uncertainties about the meaning of minor league stats, and poor judgment by major-league management?

The two cases are in that respect quite similar.
   117. Kelly in SD Posted: August 27, 2005 at 09:51 PM (#1578265)
Well, that is why I posted the prelim ballot.

I am trying to take all the information into account that our various researchers have provided regarding Luke Easter. There are a huge number of uncertainties about Easter and his career. His candidacy involves most of the controversial ways we give credit.

1. NeL credit and at what level - 1947, 48.
2. Barnstorming credit - 1946.
3. Minor league credit - 1954-1959.
4. Dobie Moore Army credit / viable alternative to organized ball / better pay / consistent pay / good team that defeated NeL teams - 1937-1941.
5. World War II credit - 1942 - 1945.

Easter and Cravath may have the most unique careers of any candidates we deal with. Easter is not going into the Hall of Merit on this election nor, presumably, in the near future so we have time for debate. The circumstances of his career force us to examine the various methods by which we give credit and the assumptions that lay behind the types of credit we give.

Similar to way that Pete Browning requires us to decide how much we trust league quality comparisons and modern defensive metrics that may have no application to his time period, Easter's career requires us to examine what we give credit for.

Is giving credit in all five of the methods listed above something we want to do for a single player? Each voter has to decide that. The lack of documentation for the St.Louis industry team and the barmstorming years is problematical. How good was he and at what age? Since we don't have that information currently, how should WWII credit be given? Do we assume his career path was the 1 hitter in several hundred (or thousand) to peak in his thirties - Cravath, Wheat (but change in style of play had its effect there.) (are there any others.) Or do we assume that his career follows a more typical career path and the numbers we have are the downside of his career?

Luke Easter has a huge numbers of unknowns; unknowns that we have tackled for many players previously to general agreement. Just not all the unknowns have been present in one career.

Some voters may reach their "unknown quota." There are a lot. Some voters may be comfortable with all the unknowns and give credit for them. But if you have given credit for the various circumstances in the past, why would you not here?
   118. EricC Posted: August 27, 2005 at 09:56 PM (#1578272)
Is Cravath a late bloomer, or is his peak concealed by league changes, uncertainties about the meaning of minor league stats, and poor judgment by major-league management?

Or the Baker Bowl?
   119. sunnyday2 Posted: August 27, 2005 at 10:25 PM (#1578324)
Here in more detail is my thought process on Easter, which in fairness should be similar to the thought process for other players.

The Gold Standard

1. The NL and AL (and for a short time the NA and PL) represented the best baseball played anywhere on the planet. They are the gold standard. Performance records from those leagues retain 100% of their value.


2. Performance records from all other leagues--including the other nominally major leagues--are discounted because they were never the best baseball. The discount rates vary.

3. NeL records are discounted but they--and they alone--are also regarded as a gold standard of sorts because they were the highest level of play available to an entire cohort of players.

4. The one exception to rules #1 and #2 is the WWII era when play did not get better, it got significantly worse, even though for a short time. There is a discount for that even thought it was ML.

Extra Credit

5. Players miss out on gold standard-level play for lots of reasons. Many affect only individual players--they get hurt or ill, they are held back in the MiLs, etc. etc. There is no XC for that.

6. On rare occasions, however, entire cohorts of players are held back--as in blacks before 1947 and players who fought in WWII. The day they were born, they were doomed to miss time when they otherwise had gold standard-level skills. There is XC for that.

Peaks and Careers

7. As a peak voter, peaks matter to me a lot more than careers. But I do not create peaks made out of XC. A peak has to be real and it has to have been created at a gold standard-level of play, in the ML or NeLs. This is too large a leap of faith.

8. I will cheerfully add career value on as XC to players who have demonstrated that they can play at a high enough gold standard peak. This is an acceptable leap of faith to me.

That is my thinking at the present time on XC and etc. There are some leaps of faith that are just too big. One is that because most players peak at 27 that therefore Luke Easter (or Luke Skywalker or the Easter Bunny or anybody else) also did peak at 27.

But based on all these rules, I can add career value for Gavy Cravath but I can't create a hypothetical peak for Buzz Arlett. I have no misgivings about this. And it is easy within my system to add career value for Dobie Moore's play with the Wreckers. He proved beyond any doubt that he was a world-class player and he did it at a gold standard-level. All I'm doing is filling out the career of a gold standard player. But I would never add value for him after he broke his leg.

With Easter you're asking me to make the leap of faith that he is a gold standard player, and I've never done that with any other player. OK, I have done it. We did it with Frank Grant and Home Run Johnson and Rube Foster (and some of you did it with Pete Hill). There was no gold standard against which to measure them.

There is a gold standard against which to measure Luke Easter and he is one of the very very very few 27-28 year old men that we have considered who was not playing at a gold standard-level at that age.

That is not something that happened to a whole cohort of players (rule 6), it is something that happened almost uniquely to Luke Easter. So to me it is more like George Sisler's eye or Dobie Moore's broken leg or Ross Young's or Addie Joss' death. Easter didn't fail to play in the NeL because he was black, he failed to play in the NeL for reasons all his own.

It's not just that he had a huge number of unknowns, per se, but because they were unknowns that are unique to him. So I can't quite make the leap of faith in this case.

Having said all of that, I will admit that the period of 1947-1955(ish) is problematic for black players, in that many who would have had their gold standard years in the NeL during that period nevertheless didn't get to play in the MLs (e.g. Willard Brown). Maybe Brown and Easter and others really are more comparable to HR Johnson and Frank Grant--maybe they were deprived of their gold standard opportunity. But given the age of Easter and Brown, right now I don't think that's true. I might feel differently about Leon Day or somebody, but right now I don't think that is true for the players eligible to date.
   120. karlmagnus Posted: August 27, 2005 at 11:35 PM (#1578434)
Sunnyday2, I have 2 comments:

1) I am DELIGHTED that others are taking up my long term advocacy of a Gold Standrad :-)

2) I think you have very well defined a rigorous and sensible approach to the difficult questions of demarcation involved in those who split their careers between different formats, not all of them top quality.

To take another difficult case (though in my view not truly of even borderline quality) your approach does mean, doesn't it that those like e.g. Bell who had their most spectacular year in Mexico, not "gold standard" by any definition should not be allowed to count it as a peak, because of the uncertainties involved.

Your approach does serve in a number of cases as a very helpful "Ockham's Razor" and I would hope that it could be generally accepted, even though others will continue to differ with you or me about its application in individual cases.
   121. sunnyday2 Posted: August 28, 2005 at 12:19 AM (#1578516)

Yes, right at the present time I don't regard Cuba or Mexico as gold standard. The one player whom I do regard as a gold standard-caliber player based essentially on Latin American play is Jose Mendez. But while his best play was in the Cuban League, there are two mitigating circumstances. 1) Like Johnson and Grant and Foster, he played at a time when there was no gold standard opportunity for blacks and 2) his play in the Cuban League is supplemented by a truly sterling record against barnstorming ML teams, most of them among the better or best ML teams.

So, yes, there are occasionally mitigating circumstances on this or that resume, but these are truly special cases.

As I said, I could be persuaded that some blacks circa 1947-60 might not have had a gold standard opportunity available to them, but it would have to be borne out by the particulars of the case.

I would also point out that my rules are, as karl says, meant to be pretty rigorous, and part of that is that no player could really establish himself as a playing at the gold standard strictly by virtue of his play in 1943-'44-'45.
   122. Brent Posted: August 28, 2005 at 01:32 AM (#1578652)
My, all this talk about gold standards and who you're not going to give extra credit to --- you'd think the PCL and the Mexican League were playing slowpitch softball on the moon!

All I want to know (and what I thought this project was all about) is to figure out who the best baseball players were. If they happened to have been playing in Cuba or Minneapolis or Oakland I don't see why that should threaten us. I think we should simply do our best to figure out how good they were and evaluate them accordingly.
   123. Dr. Chaleeko Posted: August 28, 2005 at 02:42 AM (#1578849)
I find myself in persuaded by Sunnyday/Karl as well as by Brent.

-On one hand, I think the former present a reasoned and rigorous approach to the question at hand. I don't really think it's a great idea for me to create a peak out of thin air for Luke Easter and tout it as evidence of his greatness. I don't think it's problematic to use his MLEs, some mathematics, and age-27 research as the basis for generating a career shape and wonder how likely it was for him to achieve it, but calling it evidence is another matter.

-On the other hand, Luke Easter's career is simply not like anyone else's.

What I think is needed, therefore, is more information about
1) his baseball activities from 1937-1941
2) his wartime job and specifics about how it effected where/when he could play
3) more specific information about his 1947 year in Cincy.

We need to answer the question: why couldn't/didn't Luke Easter find himself on a top-flight NgL team from 1937-1946...or was he on top teams, and we don't recognize them because they weren't League teams?
   124. Dr. Chaleeko Posted: August 28, 2005 at 03:57 AM (#1578986)
Hey, everyone, quick inquiry.

I'm working on Silvio Garcia for 1960, and I've got a lot of information on him, but I'm looking for a couple quick things:

1) Anyone got his Santo Domingo numbers for 1937 or know a website or book that has full S.D. numbers? (Holway only lists avg)

2) Anyone got 1940 Cuban Stars stats? Which league, if any, were they in?

3) Anyone got any idea what the leaguewide stats for the 1939 PRWL were?

   125. Paul Wendt Posted: August 28, 2005 at 05:30 AM (#1579181)
Kelly #117
peak in his thirties - Cravath, Wheat (but change in style of play had its effect there.) (are there any others.)

Clemente (a strong case)

sunnyday2 #119
The one exception to rules #1 and #2 is the WWII era when play did not get better, it got significantly worse, even though for a short time. There is a discount for that even thought it was ML.

Well, the NL was weak in 1884, 1890 and 1902 compared to 1881, 1889 or 1892, and 1899. Right?

to ChrisJ, et al, mainly on the preceding page:
Prorating WWI to 154 games, Edd Roush played 136 138 146 149 games 1917-1920, when he achieved #1 #2 #4 #6 on ChrisJ's list of top ten seasons by OPS+. That is a peak.

Otherwise, I agree with explanations by andrew siegel, et al, for ranking Averill significantly higher than Roush. One point not mentioned by others, and a companion to Averill's three good and one great PCL season: #10 on that OPS+ list for Roush is his first full season, in the Federal League.

andrew siegel (i hope i have this right):
How does George Van Haltren survive the timeline to rank #3 to #5 on late 1950s ballots?
   126. Kelly in SD Posted: August 28, 2005 at 06:12 AM (#1579234)
Your examples prove the point about how rare it is for a very good/great player to peak in his 30s. Even for the 5 players named there are unique circumstances associated with them. Cravath's career had several severe interruptions so that we don't know what his real peak was. Zack Wheat had the luck to have the game to change to meet his skills perfectly. Barry Bonds - I won't touch that with a ten-googleplex foot pole. Schmidt's 3 year peak, if you adjust for the 1981 strike, is from ages 29 to 31 barely over ages 24 to 26. Clemente I definitely agree with.
So if I am trying to construct an approximate value for Easter, I would probably assume that it followed the typical peak at 26 to 28 and decline from there, rather than have the 1 in hundreds exception.

Good post. My trouble with the "gold standard" is it imposes our present-day judgment on situations that were radically different than today. I have great difficulty applying that standard to non-white baseball, especially in the 1930s. It is my understanding that the Negro Leagues were not stable throughout this period. Mexican leagues became a viable alternative because they offered a more stable economic environment compared with US opportunities.

To me, it is a legitimate choice for a baseball player to decide to play for an industrial league team where he knows that he would receive a consistent paycheck and would not have to spend months out on the road barmstorming.
I just watched a Simpson's episode so this is the way I am thinking. You are a good baseball player. There is a Depression. You cannot play in the best league for some reason - place of birth, skin color, zodiac sign, etc.
You have two options: 1. You can play in "organized, major" leagues that will admit you, but these leagues are on shaky financial ground and have recently disbanded before reforming; Or 2. You can go to work for Mr. Burns which entails mainly playing baseball for his factory team. This factory team has other players would eventually make the best leagues, as would you, and these players would be good players in this best league. This factory team will sometimes play and beat teams in the "organized leagues" that are open to you.
I know what league I would join.

We need to know more about Luke Easter. What opportunities presented themselves? Are there any records of his games in the black press from St.Louis? If there are records, are they available somewhere?

Is it accurate to describe any singular option for a non-white player as the gold standard in the 1930s when there is the Mexican League, the NAL, and the NNL, plus the various winter leagues, plus barnstorming options, plus a semi-pro Dakota League?

I have more to say, but I cannot formulate my words to get my point across.
   127. sunnyday2 Posted: August 28, 2005 at 02:17 PM (#1579504)
Brent, of course we're trying to identify and honor the best baseball players. But doesn't it help to have a method for doing so?

I mean, just taking Luke Easter and all other eligible 1Bs, how do you compare Easter to Jake Beckley or Frank Chance or George Sisler or Dave Orr?

What it comes down to for me is that you start with their real record. What did they do? Against what level of competition? I'm certainly not giving him a 100% discount (that is, ignoring) his industrial league play, but OTOH I'm not willing to pretend that his numbers there should be taken at 100% of their nominal value.

Still we are all asking, what special circumstances are there that might merit some departure in our analysis from the pure record?

As I said above, for me it is easy to depart from the real record when a player is affected by circumstances that also affected a lot of other players, and so I treat the NeLs like the gold standard that they are and I award XC to players who fought during WWII, etc.

But when it comes to players affected by special circumstances that are peculiar to that one player only, it gets a lot harder, and of course Luke Easter is a lot harder. But if I give him special credit, then why not George Sisler? Why not pretend that he never had sinusitis and just project 7 more seasons for him like the first 7? Why not give Sisler credit for his time at the University of Michigan?

Easter's case is complicated by the fact that his special circumstances include the fact that he was black, and those things deserve XC. But he is a special case among his black cohort, too, and so we need to separate out those two things.

James once said that giving credit for WWII years for players who served in the military is appropriate (though he gives that credit not in his numerical analysis but in his bullshirt dump). Because, it's not a question of imagining how good they were. We already know how good they were based on what they did pre- and post-.

With Easter, unfortunately, we are trying to imagine how good he was. I just think this is an area where we have to be careful, though we've done it before, as I said, with the pre-1920 black players. And I could be convinced that we should do it for Silvio Garcia, e.g. But, carefully.

It's one thing to be fair to Luke Easter and Silvio Garcia. It's another to bend over backwards to the extent that we are not being fair to other players. (And I'm the one who asked for a Luke Easter thread because he deserved a fair hearing, which I think he is getting. Anyone who really likes Easter, BTW, should also spend some time studying Bobby Estalella, another of those extremely odd cases affected by segregation in a completely unique way.)

In summary the relevant "rules" that I think are fair to all players and therefore to Luke Easter are:

1. XC for circumstances that affect a whole cohort, but only in rare cases for players who are uniquely affected.

2. Per James, once we have solid knowledge of how good a player was, we can easily fill in the blanks in a career (per rule #1). But imagining how good a player was--what an imaginary peak might have looked like--is a different and harder task, and one that should only be done in rare cases and on a case by case basis.
   128. Paul Wendt Posted: August 28, 2005 at 03:36 PM (#1579577)
Your examples prove the point about how rare it is for a very good/great player to peak in his 30s.

I don't think so. Schmidt enjoyed his best season at age 31.6 to 31.11, his three best batting seasons at 30.6 to 33.0 (years.months). I didn't search a database, where I think you'll find numerous great players with peak years centered at 23 or 31. What you won't find in numbers is great players arriving at age 31, after a few years of full-time major league play below league average, like Clemente apparently "arrived" at age 25.8 (if you say April 1960).

The age 27 generalization isn't strong enough to imply that Cravath was on the downside as a ballplayer at age 32 (1913, his best mlb season) or even that he was as good age 28-30 as he was age 34-36.
   129. John (You Can Call Me Grandma) Murphy Posted: August 28, 2005 at 05:30 PM (#1579765)
I'm finally able to set up those threads now. Jim Furtado was able to correct the problems that I was still facing. The threads will be up within the hour.
   130. Brent Posted: August 28, 2005 at 05:51 PM (#1579831)

I don't think we're actually that far apart in the case of Easter. There are conflicting principles at work in his case. On the one hand I do give WWII credit, and I also give credit for major-league quality play for teams without statistical documentation (for example, Dihigo in Venezuela). But I have never given a player credit for performance at a level higher than that demonstrated in the seasons for which we do have statistics. And by demonstrated level of performance, I don't mean the player's best season (so Cecil Travis doesn't get WWII credit for 4 seasons at 34 WS). So even though the majority of players peak at a level higher than their age 32-36 level of performance, I'm not willing to extrapolate a peak for Easter higher than what we can observe. As a peak voter, that leaves Easter falling short of my ballot. (However, if new information becomes available on his pre-1947 performance, I will certainly reconsider.)

Where I disagree with you is with your premise that crediting players for playing baseball in places other than your "gold standard" leagues is somehow giving them "extra credit." Extra credit is credit given for seasons not played (military, blacklisted) or for opportunities not available to all players (performance in post-season play or winter leagues). But the guys playing in the minor leagues or Mexican League were just playing baseball in the best places avilable to them, and our project is to try to figure out how good they were at it. There's no extra credit involved.

And Sisler is a red herring. We're trying to measure how good he was as a player, and after he got sinusitis he wasn't a very good player. That's entirely different from Buzz Arlett, who demonstrably was a near-HoM quality player for many seasons in the PCL, or from Cool Papa Bell's 1940 season in Mexico, where we have statistical evidence that he really was very good. (On the other hand, if you're able to put together convincing evidence that Sisler was playing at the level of a major league star while he was attending University of Michigan, I'll be happy to take account of that information in my assessment of him.)
   131. sunnyday2 Posted: August 28, 2005 at 08:27 PM (#1580267)

I was using the term "extra credit (XC)" as synonymous with "major league equivalent (MLE) credit." Your point that they are not always the same thing is well taken, though I'm also not sure it changes the evaluation of any player. I'll have to think about it.

As for Sisler versus Arlett, yes, they're different circumstances. And I am certainly not saying that Sisler was playing at a "star" MLE level at Michigan, I do wonder if he wasn't above replacement, which, if so, would be worth a little bit of career value.

I agree we're not that far apart after reading your comments about a player's peak. I also agree that new info could always result in a new evaluation.

I don't agree that guys playing in the minor leagues were playing at the highest level available. That is where my distinction between individual players versus cohorts is helpful. Arlett could have played in the MLs, seen from a macro perspective. Joe Rogan, taking another pitcher-hitter as an example, could not.

Though the real point is of course that Arlett's MiL numbers need to be discounted. It's just a question of how much and what is left. I guess I just see a little less of a remainder than you do.

It is these special cases that make this fun and equally importantly cause me to think about how and why I evaluate players the way I do, and whether what I do is fair not only to the special cases but to all.
   132. yest Posted: August 28, 2005 at 09:20 PM (#1580389)
for the past 2 days I have been missing the date markers on all the posts for examples
# Posted by Trevor P. on at 04:05 PM (#1575731)
# Posted by andrew siegel on at 04:25 PM (#1575772)Posted by Chris Cobb on at 04:31 PM (#1575790)

does anyone know how I can fix that
   133. John (You Can Call Me Grandma) Murphy Posted: August 28, 2005 at 09:50 PM (#1580438)
does anyone know how I can fix that

Joe or I can't do anything about that, yest, but Jim Furtado and Dan Szymborski are aware of the problem. The site is undergoing renovations, so things might go a little haywire until things are completed.

BTW, I have changed our home page. From now on, the Important Links and Negro League threads wont show up at the top anymore. It has been replaced by the Welcome to the Hall of Merit thread. Don't fret, you can connect to all of the threads by clicking on Important Links within the new thread.
   134. sunnyday2 Posted: August 28, 2005 at 09:54 PM (#1580445)
But I think the new thread will disappear from Hot Topics since there is no comment box associated with it. The Hall of Merit under Blogs used to serve that function and it never went away...?
   135. John (You Can Call Me Grandma) Murphy Posted: August 28, 2005 at 09:57 PM (#1580456)
But I think the new thread will disappear from Hot Topics since there is no comment box associated with it. The Hall of Merit under Blogs used to serve that function and it never went away...?

As I just e-mailed to you, Marc, the new thread will always be at the top of our home page because I have made it "sticky." Yes, it will fall off Hot Topics, but Important Links never was there, either.
   136. karlmagnus Posted: August 28, 2005 at 10:22 PM (#1580479)
We should therefore bookmark "Important links" as our HOM home page, otherwise it will disappear, right?
   137. John (You Can Call Me Grandma) Murphy Posted: August 28, 2005 at 10:58 PM (#1580521)
We should therefore bookmark "Important links" as our HOM home page, otherwise it will disappear, right?

Depends, karlmagnus. If you bookmarked the page that had Important Links, Negro Leaguers' Home Page, plus all the latest pages following it (Hall of Merit Home Page), your bookmark will be fine. If not, then you will have to update your bookmark.
   138. karlmagnus Posted: August 29, 2005 at 02:31 AM (#1580900)
Without wishing to be difficult "Important Links" should be above the Negro leaguers on the home page, as they are just one set of players being linked to. Currently, you have to scroll down 2-3 times to get to "Important links" which gives the NEL players an "affirmative action" boost that seems uncalled for.

Personally I preferred the way it was, but there we are; this is true in a lot of areas of human activity, I find!
   139. Sean Gilman Posted: August 29, 2005 at 07:07 AM (#1581421)
"Important Links" is above the Negro Leaguers, Karl, and it has been all day. . .
   140. andrew siegel Posted: August 29, 2005 at 01:28 PM (#1581606)
What to do with George Van Haltren?

Paul asks why George Van Haltren retains his high place on my ballot, despite timelining considerations.

The literal answer to that is easy--once we get past 1868 or so, I don't timeline much. I dock players for low league quality compared to other leagues of their time, I deflate 1870s numbers because the distribution of talent was more uneven creating larger standard deviations, I am skeptical about giving much credit for huge IP numbers when those numbers are the norm, etc.--but I don't timeline per se.

But Paul's question is a bigger one? Am I holding on to Van Haltren (and some other 1890s players) b/c/ I think they should have been elected years ago and not b/c/ I think they are better than the modern ballot candidates?

The short answer is "I don't know."

Re: Van Haltren I *think* I am following the numbers, not my stubborn heart, for the following reasons:

The case against Van Haltren is that he was an OF playing in an era with lots of big hitting OF's and only managed to crack the league leaders in OPS+ and similar stats a few times.

The case for Van Haltren is, well, everything else. His OPS+ was OBP heavy, he stole lots of bases, his peripheral offensive stats are good enough to give him a nice lead in offensive winning percentage over roughly similar guys like Ryan, he was very consistent, he was very durable, he was a good defenisve OF who played a higher percentage of his games in CF than most of his similar hitting contemporaries, he was versatile enough to play some SS (though not so well), he was a decent pitcher for a few years, he arguably deserves a little minor league credit on both ends of his career, and his best seasons were in the heart of the one-league 1890s (unlike Ryan or Beckley).

Add up all those small plusses and I see him as noticeably better than any of the other 120ish OPS+ long career OF/1B. Now it is possible that all those guys should be out of consideration. But as long as I have (1) lesser contemporaries like Ryan and Beckley and (2) guys with similar offensive and defensive value but less durability or lesser leagues or both (like Roush) hanging around the edge of the ballot, Van Haltren strikes me as deserving at least a mid-ballot slot.
   141. John (You Can Call Me Grandma) Murphy Posted: August 29, 2005 at 02:12 PM (#1581689)
"Important Links" is above the Negro Leaguers, Karl, and it has been all day. . .

Right, Sean. It's within the Welcome to the Hall of Merit thread.

Since the Baseball Hall of Merit's Important Links and the Negro League Home Page are actually new pages that I set up (with all of the old links attached as in the past), those pages will disappear from the top pf our home page when more pages are created. IOW, the Welcome page is where you need to go from now on.

Personally I preferred the way it was, but there we are; this is true in a lot of areas of human activity, I find!

Two things, karlmagnus:

1) We have an introductin to the HoM for lurkers now, while not taking anything away from us.

2) Jim Furtado suggested it to me.

Even though I think it's a good idea anyway, I try to do what the guy who is paying the bills around here thinks is best. :-D
   142. sunnyday2 Posted: August 29, 2005 at 02:15 PM (#1581694)
For some reason the way this new link works changed last night some time. It now seems to be just what is needed.
   143. John (You Can Call Me Grandma) Murphy Posted: August 29, 2005 at 02:22 PM (#1581710)
It now seems to be just what is needed.

   144. TomH Posted: August 29, 2005 at 08:20 PM (#1582294)
A bit of 'clutch' info on Ducky Medwick:

The Cards won a close pennant race in 1934. They did it in part by out-performing their Pythag expected wins; Medwick, who had a fine year, ought to get some share of credit for helping earn a flag. They won a 7-game W.S., and Medwick had a good series, although most games were blowouts; no real clutch hits.

Joe's Dodgers won a close race in 41. They lost a 5-game W.S. to the NYY, mostly close games (margins of 1, 1, 2, and 3 runs). Mediwck hit poorly, driving in 0 runs and scoring one.

The Dodgers lost a close race in 42. They didn't under-pythag, but Medwick had a poor year by his standards (115 OPS+).

All in all, not much + or - credit from me.
   145. OCF Posted: August 29, 2005 at 08:33 PM (#1582312)
On the subject of great players peaking in their 30's (Paul Wendt's posts above), what about Honus Wagner? Not to suggest that he was exactly chopped liver in his 20's, but he didn't have a full-time job until he was 24 and didn't settle at SS until he was 29. His great 1908 season happened at the age of 34.
   146. OCF Posted: August 29, 2005 at 08:53 PM (#1582355)
Nice post in #40 by Andrew. Of course, Van Haltren survived and Ryan didn't, which is harder to explain. (I have trouble enough explaining my own ballots.)

Here's a similar question of ballot survival versus disappearance. Why does John McGraw still live on some people's ballots while Frank Chance has disappeared? Both are essentially peak candidates; both have OBP-rich offensive lines. Both have the same worst flaw: poor durability, including poor in-season durability, resulting in a rather low number of career games played. As a distraction, both were player-managers of successful teams. (McGraw had a much more distinguished post-playing career as a manager; that's irrelevant.) McGraw's raw offensive statistices (including his OBP's) are more eye-catching than Chance's, but he did play when twice as many runs were being scored. But Chance had 300 more games played than McGraw.
   147. Dr. Chaleeko Posted: August 29, 2005 at 08:53 PM (#1582358)
Speaking as a Pennsylvanian at heart, if not at mailbox, and as a man with a lot of PA-Dutch genetic material inside me, Honus's late-peak is the result of his good German stock plus lots of pork. I bet his SCR+ was awesome by comparison to other players of his time.

For those not in the know, SCR+ is Adjusted Scrapple Index, a measure of how much of the yummy but much-maligned pork product a player ingested compared to other players in his league, adjusted for home state.
   148. TomH Posted: August 30, 2005 at 12:56 AM (#1582966)
good Q, OCF. Chance is at #17 on my ballot this week.
And Bresnahan has much the same argument as the other two.
   149. Dr. Chaleeko Posted: August 31, 2005 at 02:17 AM (#1586337)
Gadfly, Christian, Gary A., KJOK, or anyone else,

Do you have any kind of information on the Big Sky League, the ManDak League, the Western League, or the Domincan Summer League in the early 1950s?

In particular I'm looking for number of team games, league average, league slg, league hits, and league bb.

   150. Howie Menckel Posted: September 02, 2005 at 06:16 PM (#1593522)
For amusement purposes only:
Shamelessly cribbing from the works of others, an unofficial rundown of some top candidates per 'year' as we prepare to take on the 1960s. Two listed even when there are fewer than two decent candidates' I probably listed an extra couple that will go nowhere, might miss a decent guy somewhere.

1960 (winners named September 19)—elect 2
WS W3 Rookie Name-Pos (Died)
264 103.5 1940 Hal Newhouser-P (1998)
Negro Lg 1934 Leon Day-P (1995)

1961 (October 3)—elect ONE
WS W3 Rookie Name-Pos (Died)
265 78.8 1942 Vern Stephens-SS (1968)
242 73.1 1946 Ralph Kiner-LF (living)

1962 (October 17)—elect 2
WS W3 Rookie Name-Pos (Died)
292 106.3 1936 Bob Feller-P (living)
257 84.8 1947 Jackie Robinson-2B (1972)
231 72.7 1941 Phil Rizzuto-SS (living)
098 34.3 1940 Monte Irvin-LF (living)

1963 (October 31)—elect 2
WS W3 Rookie Name-Pos (Died)
207 73.6 1948 Roy Campanella-C (1993)
229 66.3 1944 George Kell-3B (living)

1964 (November 14)—elect 2
WS W3 Rookie Name-Pos (Died)
314 95.4 1940 Pee Wee Reese-SS (1999)
232 88.2 1946 Bob Lemon-P (2000)

1965 (November 28)—elect 2
WS W3 Rookie Name-Pos (Died)
323 97.2 1938 Enos Slaughter-RF (2002)
296 77.6 1939 Mickey Vernon-1B (living)
268 77.3 1942 Larry Doby-CF (2003)

1966 (December 12)-elect 2
WS W3 Rookie Name-Pos (Died)
555 169.3 1939 Ted Williams-LF (2002)
226 61.8 1948 Alvin Dark-SS (living)
205 61.8 1947 Bobby Thomson-CF/LF (living)
176 65.0 1949 Don Newcombe-P (living)

1967 (December 26)-elect 2
WS W3 Rookie Name-Pos (Died)
203 48.1 1948 Ted Kluszewski-1B (1988)
187 53.2 1950 Jackie Jensen-RF (1982)

1968 (January 9)-elect 2
WS W3 Rookie Name-Pos (Died)
329 106.0 1948 Richie Ashburn-CF (1997)
262 85.6 1945 Red Schoendienst-2B (living)
267 65.4 1647 Eddie Yost-3B (living)

1969 (January 23)-elect 2
WS W3 Rookie Name-Pos (Died)
604 188.1 1942 Stan Musial-LF/1B (living)
375 119.5 1947 Yogi Berra-C (living)
309 107.9 1941 Early Wynn-P (1999)
263 76.6 1947 Gil Hodges-1B (1972)
   151. Chris Cobb Posted: September 02, 2005 at 07:03 PM (#1593584)
Looks like the next decade is going to be quite interesting, as we'll be electing 5-7 out of the current backlog.
   152. KJOK Posted: September 02, 2005 at 07:27 PM (#1593628)
"an unofficial rundown of some top candidates per 'year' as we prepare to take on the 1960s. Two listed even when there are fewer than two decent candidates' I probably listed an extra couple that will go nowhere, might miss a decent guy somewhere."

Do we have elgibility dates for Artie Wilson or Bus Clarkson??
   153. Chris Cobb Posted: September 02, 2005 at 08:57 PM (#1593797)
Bus Clarkson is eligible in 1962.

Artie Wilson is eligible in 1963.
   154. Paul Wendt Posted: September 03, 2005 at 01:59 AM (#1594453)
Win Shares doesn't distort player's totals due to that player being on a winning or losing team. . . . the distortion comes when a team is consistently above or below their pythag.

Chris Cobb #114
As I and others have said and documented before, this is not true. Extreme team performance, whether good or bad, can influence player win share totals as much as large swings above or below Pythagorean projections.

This as an important point. Has it been covered elsewhere, mainly in one place? systematically?
   155. Dag Nabbit at Posted: September 03, 2005 at 03:44 AM (#1594712)
1960 (winners named September 19)—elect 2
WS W3 Rookie Name-Pos (Died)
264 103.5 1940 Hal Newhouser-P (1998)
Negro Lg 1934 Leon Day-P (1995)

Already? ####

And all this time I've been busily preparing my Allie "no chance I'm getting elected to the HoM" Reynolds page. Crud.
   156. sunnyday2 Posted: September 03, 2005 at 01:26 PM (#1595168)
Paul, there has been intermittent discussion, probably not a systematic discussion, and it seems pretty obvious that there is no consensus on the affect of winning and losing teams on individual player's WS.

James of course says there is no systematic problem.

Examples of course can be cited--and have been cited here--of players over or under-performing on WS due to over- and under-performing pythag. Joe Medwick probably picks up 2-3 extra WS one year and George Sisler probably loses a few, e.g. Wally Berger and Chuck Klein perhaps lose some WS due to team performance though, again, this is exactly the kind of claim for which there is no real consensus.

I would guess that those who feel that this is a big, systematic problem probably don't use WS that much anyway, and most who use WS don't adjust for this sort of thing. So in a way, while there is no consensus, there is also no really big issue.

My own sense is that for most players the ups and downs balance out, and for the few winners and losers the difference is a half a dozen WS in a career, probably not much more. Since I use WS the way James does--which is to say, as a starting point, and then the really dirty work gets done in the bullshirt dump anyway--it is a matter of almost no consequence.

But again, this is not to say there's a consensus on that point.
   157. Chris Cobb Posted: September 03, 2005 at 02:39 PM (#1595199)

To give a somewhat different perspective (as someone who uses ws as a primary metric but is concerned about its handling of extreme teams), it appears to me that there have been two different cases made about the distorting effect of extreme teams.

One has been advanced by jimd. He has argued that because win shares does not adjust (as, bbref OPS+ and WARP) do, for competition differences resulting from the fact that teams do not play themselves, win shares will overrate players on good teams and underrate players on bad ones. He has suggested that the amount of this distortion can be easily estimated by "balancing the schedule": adding to a team's schedule a number of team games equal to the number of games played against each other team. In those games, the team would have the .500 wp they would achieve when playing themselves. With this addition, an adjusted winning percentage can be found, which, applied to the team's actual games, produces a properly competition-adjusted pool of win shares for the team.

The other I have advanced. It is not a theoretical case like jimd's, but an empirical one. I studied all the player seasons between 1905 and 1965 in which a player achieved a 175-185 OPS+. I found the players' bws/g for each of these seasons and grouped them according to team winning percentage. There seemed to be a very strong correlation between highest-winning-percentage teams and the highest bws/g for offensive performances that were measured as approximately equal by a statistical measure not tied to team wins. I also looked at EQA for those seasons: it tracks similarly with OPS+ and very differently from win shares.

I undertook the study to figure out why Joe Medwick's 180 OPS+ season and George Sisler's 181 OPS+ season were valued so differently in win shares when they are almost identical in OPS+ and WARP. The full posts on this study can be found in the 1958 ballot disscussion thread at #87 and #100. There's some discussion from others in the posts between and following. I don't have the statistical tools to do rigorous analysis of the significance of the correlations I found, but they look highly unlikely to be a random distribution.

I find it highly unlikely that in a large majority of cases, in a set of seasons that look identical when analyzed at the level of a player's individual statistics, context-adjusted, all of the seasons that are better, in ways inaccessible to our direct analysis, happened to occur for teams with great records, while all of the seasons that were secretly worse in the group happened to occur for teams with poor records. But, if win shares are to be accepted, this is how these seasons are to be evaluated.

Jimd's point addresses the career win-share effects of playing consistently on very good or very bad teams; my point addresses the seasonal win-share effects on great seasons of win shares being tied as they are to team performance.
   158. Paul Wendt Posted: September 03, 2005 at 02:52 PM (#1595203)
Quoting myself
Prorating WWI to 154 games, Edd Roush played 136 138 146 149 games 1917-1920, when he achieved #1 #2 #4 #6 on ChrisJ's list of top ten seasons by OPS+. That is a peak.

Their career batting patterns suggest that deadball era conditions relatively favored Edd Roush, as 1920s conditions favored Zach Wheat --or, as Bill James put it years ago, Wheat was unusually able to take advantage of a change in conditions. In Deadball Stars of the NL, Wheat is notable for his lusty swing. Perhaps the dirty and doctored baseballs were relatively effective against such a hitter.
   159. Dag Nabbit at Posted: September 04, 2005 at 01:39 AM (#1596271)
Already? ####

And all this time I've been busily preparing my Allie "no chance I'm getting elected to the HoM" Reynolds page. Crud.

Woo-hoo! Finished the notes section! Now all that's left is the AOWP.
   160. Cblau Posted: September 04, 2005 at 01:41 AM (#1596282)
There's an article on the MadDak League in the SABR convention publication, Dominionball.

The league had a 48-game schedule in 1950, according to the article. However, it doens't offer any other statistical information. It quotes claims the league quality was somewhere between AAA and A.
   161. TomH Posted: September 04, 2005 at 07:26 PM (#1597050)
Re: Chris C's summary of the "does Win Sharesd over-benefit those on good teams?":

I made a mistake a few years ago when compiling the results of a study, that may affect his conclusions.

When measuring OPS+ "bins" by team winning percentage, there will be a 'selection effect' occurring; teams with good winning pcts tend in general to have out-performed their Runs Scored / Runs Allowed projection, which is partly how they got to be winning teams. Since Win Shares gives credit to players on teams who win more games than their RS/RA ratio would suggest, naturally similar OPS+ hitters will have more win shares on these teams.

If Chris re-ran his study and defined "good" teams by their ratio of RS/RA instead of WPCT, we would see if this effect persisted, dimished, or disappeared.
Page 2 of 2 pages  < 1 2

You must be Registered and Logged In to post comments.



<< Back to main

BBTF Partner

Dynasty League Baseball

Support BBTF


Thanks to
Marc Sully's not booin'. He's Youkin'.
for his generous support.


You must be logged in to view your Bookmarks.


Page rendered in 0.6891 seconds
41 querie(s) executed