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

Tuesday, October 31, 2006

1990 Ballot Discussion

1990 (November 27)—elect 3
WS W3 Rookie Name-Pos (Died)

512 156.8 1965 Joe Morgan-2B
312 107.4 1965 Jim Palmer-P
302 86.5 1970 Ken Singleton-RF
286 82.6 1969 Amos Otis-CF
258 64.1 1967 Rick Monday-CF
247 59.3 1971 Greg Luzinski-LF
236 60.8 1968 Bob Watson-1B
181 61.0 1968 Ron Reed-RP
185 54.6 1971 Mickey Rivers-CF
159 56.1 1969 Mike Torrez-P
158 56.1 1965 Tug McGraw-RP (2004)
140 52.4 1971 Paul Splittorff-P
164 43.1 1969 Lou Piniella-LF
122 46.6 1972 Mike Caldwell-P
110 44.8 1968 Tom Burgmeier-RP
112 41.1 1971 Milt May-C
113 39.1 1972 Dick Tidrow-RP
116 37.2 1973 Bucky Dent-SS
122 34.5 1977 Gene Richards-LF

Players Passing Away in 1989

Age Elected

90 1942 Bill Terry-1B
82 1954 Willie Wells-SS

Age Eligible

94 1940 Sparky Adams-2B/3B
90 1937 Bibb Falk-LF
90 1939 Lew Fonseca-1B/2B
89 1941 Jocko Conlan-CF/HOF Umpire
89 1943 Judy Johnson-3B
86 1944 Joe Stripp-3B
85 1945 Fred Frankhouse-P
82 1955 Rip Sewell-P
80 1948 Lefty Gomez-P
78 1953 Skeeter Newsome-SS
73 1953 George Case-LF/RF
73 1958 Ted Wilks-RP
73 1965 Murry Dickson-P
66 1965 Carl Furillo-RF
61 1967 Billy Martin-2B/Mgr
46 1977 Joe Foy-3B

Upcoming Candidate
35 1994 Donnie Moore-RP


Thanks, Dan!

John (You Can Call Me Grandma) Murphy Posted: October 31, 2006 at 07:14 PM | 178 comment(s) Login to Bookmark
  Related News:

Reader Comments and Retorts

Go to end of page

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

Page 2 of 2 pages  < 1 2
   101. Mark Shirk (jsch) Posted: November 14, 2006 at 03:38 PM (#2237479)

you keep stating that 1B in the 1890's was a very tough defensive position despite the evidence that has been shown on these threads. I am talking about the chart someone showed that the plays fielded by 1B were high in the 1880's and high again starting in the mid 1900's, but happen to be low during Beckley prime seasons. I guess that 1890's 1B deserve a magical bump in defense because they weren't fielding more balls, that is for sure. So Dr. C isn't producing analysis that ignores your claim he is producing analysis that is backed up by other analysis that you ignore. And seriously, you think that Beckley deserves a 3-4 WS bump? Maybe 1 but I can't imagine 3-4.

Doc, how does it look with 23 WS?
   102. Mark Donelson Posted: November 14, 2006 at 04:25 PM (#2237521)
Hmmm... it appears I am in the minority about Palmer. I think he was quite over-rated.

In what must be the first instance of agreement on an issue of any controversy between Rusty (who I believe is a reasonably strong career voter) and me (peak peak peak), I'm also in this minority at the moment.

Perhaps I'm relying too much on WARP/PRAA, which may be taking too much away from Palmer's peak seasons (I imagine because of the strong defenses he pitched in front of). I'm not finished crunching numbers yet, but I am finding Palmer is more likely to be low on the ballot than the automatic #2 that I'd assumed he'd be. Presently, I have him behind Jenkins by a small but definite margin.

Still, since it's just me and Rusty so far as the outliers, and it's not like my system and his align all that much anywhere else (well, that's not entirely true: he does vote for Moore and Trouppe), I'm going to be especially careful about this. Any thoughts from other peaksters? Sunny has already weighed in strongly in favor of Palmer, but I'm finding our old alignment is separating more and more in this modern era for some reason. Mark S.? Dr. C?

(All that said, of course, the man they called Cakes will easily make my pHOM immediately; you have to drop to about 23rd at this point not to.)
   103. Dr. Chaleeko Posted: November 14, 2006 at 05:08 PM (#2237562)
It would be silly, of course, to give him 3 WS. That would suggest that he would be as valuable as a very good 3B or as an average or slightly below-average SS. Neither passes the sniff test. Even two WS would be fishy to me, it would more double his fielding in most years, putting him at a level near the average 3B. That's not a scenario I would feel comfortable accepting.

We've been shown ample evidence (especially by CBlau) that 1Bs in the day were not worth twice as much as their WS suggest, but rather MAY (or may not) have been worth some fractional value above their listed FWS value. Maybe or maybe not. I see one WS as possible, especially if you round upward, which I do. So Beckley's best-case scenario is that he's earning one more WS in each season. So 23 tops.

If Beckley is a 23 WS guy instead of a 22 WS guy then 22% of all 100-loss-level teams (.383 pct or lower) are led by a 23-WS or higher guy. Still under 5% of all pennant winners with leaders at 23 WS.

The magnitude is somewhat lessened, but the strength of the trend is abundantly clear. And that's if you think 1 WS is an appropriate increase. I think there are voters who would disagree with that conclusion and would favor a smaller increase or none at all.

I don't have the total distribution of records for teams led by 23 WS guys, but I would unscientifically speculate that those team's aggregate win% would fall around .465. This is based on the fact that the 22-WS led teams aggregated at .452 and the break-even is some fractional value between 25 and 26 WS. That assumes a somewhat linear relationship, of course, of about fifteen pct points per each additional WS from the team leader, so: 23---.465, 24---.480, 25---.495, 25.x---.500. Just a guesstimate with zero supporting data beyond the 22-led aggregate data and the approximate break-even point.
   104. karlmagnus Posted: November 14, 2006 at 05:35 PM (#2237596)
No, DrC, I think most voters would feel 2-3WS was appropriate. I am wholly unconvinced by the theory, backed by very partial data (because we don't have good data) that 1B magically became less valuable during the 1890s and then became more valuable again. There's no reason whatever why this might have been so, and nobody has advanced one.

WS overrates outfielders in 1880-1920, that is generally agreed. The outfielder WS need to go somewhere, and 1B is one of the places they go. ALL infielder positions were far more valuable in the 1890s than they are today, and pitchers and outfielders less valuable because (i) much lower percentage of plays in outfield and (ii) higher error rates (due to gloves, bumpy fields etc.), meaning infielder fielding contributed more to wins, there being more variation between fielding percentages. I do not and never have claimed that 1B was an exceptionally valuable fielding position in the 1890s, I claim only that it was marginally more valuable than a modern CF, somewhere between CF and 3B.

In any case, of Beckley's 19 full seasons, only half fall within the 1890s, but all fall within the era of important infield defense.
   105. DL from MN Posted: November 14, 2006 at 05:35 PM (#2237598)
1990 prelim ballot

1) Joe Morgan - behind Collins, Lajoie and Hornsby in value contributed to era. If I'm allowed to timeline I'll put him even with Collins. Top 50 player.
2) Ferguson Jenkins
3) Jim Palmer
4) Bob Johnson
5) Luis Tiant
6) Norm Cash
7) Jake Beckley
8) Quincy Trouppe
9) Reggie Smith
10) Tommy Bridges
11) Virgil Trucks
12) Jim Wynn
13) Ken Boyer
14) Edd Roush
15) Orlando Cepeda
16-20) Dutch Leonard, Bob Elliott, Jack Quinn, Bus Clarkson, Charlie Keller
21-25) Luke Easter, Dick Redding, Vic Willis, Dave Bancroft, Urban Shocker
26-30) Frank Howard, Bobby Bonds, Gavy Cravath, Hilton Smith, Alejandro Oms
31-35) Johnny Evers, Dizzy Trout, Ken Singleton, Ben Taylor, Boog Powell

Amos Otis - in the 120s
   106. Mark Shirk (jsch) Posted: November 14, 2006 at 05:41 PM (#2237604)

Do you have any data that this is so? The only data that I have seen contradicts your assertion. To disagree with the data simply because there isn't that much of it seems as specious as claiming that 1B in the 1890's was a valuable as in the 1990's.
   107. karlmagnus Posted: November 14, 2006 at 06:18 PM (#2237640)
Data that backs the position of 1B in the 1890s includes infield/outfield hits (doubles, triples and HR vs singles), walks, stolen bases and errors. When you add that data to the written and oral historial record of the trends in the game, the burden of proof is heavily on those who wish to asset a sudden and improbable blip in the way the game was played. I repeat: the staitsical evidence for such a blip is very partial and limited, and nobody has produced a plausible hypothesis of why it might be so.
   108. Mark Shirk (jsch) Posted: November 14, 2006 at 08:14 PM (#2237763)
But 1B, according to CBlau's chart fielded fewer balls per game in the 1890's and early 1900's than the decades immediately preceding and succeeding that period. To me that is data. Did those trends of HR/3B/2B vs. S (if that can even be used for evidence here) stay as true for teh high scoring 1890's as for the lower scoring eras surrounding it? Your evidence is all circumstantial with no hard data. I believe that WS underrates 1B of Beckley's era but I do not think that it does so enough to change seasonal numbers by more than 1 WS.
   109. Mike Emeigh Posted: November 14, 2006 at 08:25 PM (#2237772)
I keep going back and forth on Tinker and Evers, depending on defense vs. offense.

I would suggest that you look very carefully at what happened when Evers moved from Chicago to Boston in 1914, if you're going to give Tinker an advantage over Evers based on defensive value.

-- MWE
   110. karlmagnus Posted: November 14, 2006 at 08:27 PM (#2237777)
That's the point at which I don't believe the data. We don't have that information for those years in any kind of accurate or complete form, and I've yet to see from CBLau or anyone else any convincing sourcing of his charts.

The proposition that hitters suddenly stopped hitting towards first base for a decade and then started again is extremely unlikely unless there is any remotely plausible explanation for why it might be true -- and I've yet to hear one. You are a very long way from refuting the null hypothesis, which is that 1B was similar in the 1890s to the 1880s and the 1900s, with the game changing in known fashions that don't involve 1B fielding before 1920, though they do include the difficulty of playing the position, which gradually declined, as it did for all infield fielding, as fields got less bumpy and gloves got better. There will have been some effect from the change in pitcher distance in 1893, but nobody's explained why it should produce fewer chances to 1B -- rather the opposite, in fact.
   111. Dr. Chaleeko Posted: November 14, 2006 at 08:52 PM (#2237797)
Let's look at reasonableness.

Short-form WS assigns 1 fielding WS to each position based on the following guidelines:
C: 1 per 24 games
1B: 1 per 76 games
2B: 1 per 28 games
3B: 1 per 38 games
SS: 1 per 25 games
OF: 1 per 48 games.

SFWS presumably uses the average defensive value of a position over time and so I'm using it as a proxy for average.

What's that translate to in terms of /162 value?
C: 6.75 FWS
1B: 2.13 FWS
2B: 5.79 FWS
3B: 4.26 FWS
SS: 6.48 FWS
OF: 3.38 FWS

Is it reasonable to suggest that a 1B could get 3 WS more credit per year in the 1890s? A 241% increase in FWS? I don't see it, and Karl, I don't think it's defensible. That would mean that the average 1B was 20% more valuable than the average historical 3B. That doesn't pass the reasonableness test. The league's batters would need to be much more left-handed than they were for that to be reasonable. It would further mean that the average 1B would be be nearly equivalent to the average historical 2B. Again, that doesn't seem reasonable to me at all.

If you add merely two FWS, you are doubling the value of a 1B in the period. Again, I'm not thinking that reasonable. Was a 1B doubling up the number of plays he was making or the number of runs he was saving? In that scenario the 1B becomes as valuable as the average historical 3B. Again, I think that's pretty dubious.

Now add a single WS (46% more value per 162), and you're talking about being as valuable, on average, as the average OF (making no allowances for corner versus center). OK, I can buy that as a best-case scenario, though it still feels high to me.

OK, so let's turn our attention to Beckley himself. What do his 162-adjusted FWS look like?

In the decade in question, 1890-1900, here's his FWS adjusted to a 162 G schedule:

1890: 2.4* (3)
1891: 2.3* (3)
1892: 2.1 (5)
1893: 3.2* (1)
1894: 2.8* (2)
1895: 3.3* (1)
1896: 1.8 (4)
1897: 1.8
1898: 1.9 (4)
1899: 3.2* (1)
1900: 2.4* (1)

Beckley had some very nice defensive years: the asterisks indicate years where he's above average and the parentheticals show when he was in the top five FWS for his position. He did both a lot. But adding two or three WS to his totals in 1893, for instance, is making him as valuable as an average historical SS (per 162). It would be very hard to believe that a 1B in any era would be that valuable (maybe in the early, early days, maybe), particularly once gloves are gaining wider acceptance.

So, what kind of adjustment would I make to his totals based on the best-case scenario? I'd add 1 WS/162 to him as the highest adjustment. I'd prorate it based on his playing time in-season, translated to a 162 schedule. Here's what that would look like:

1890: 3.35
1891: 3.26
1892: 3.08
1893: 4.18
1894: 3.82
1895: 4.29
1896: 2.64
1897: 2.70
1898: 2.66
1899: 4.03
1900: 3.44

This would represent a total increase in his FWS for the period of 37% from 27.3 to 37.5. Which strikes me as emminently more fair than piling on 30.4 WS at a rate of 3/162 and upping his FWS output by 211%.

At these kinds of rates, Beckley is getting credit but he's not, for instance, placing among the fielding leaders at more valuable defensive positions. Which it seems to me is appropriate.

Anyway, you can take all the batted-ball data you want, but those hits don't all go to first basemen, in fact, without a surfeit of lefty hitters, most of those in-play balls should head toward 3B/SS/LF/CF, where Beckley has little or no influence on the play. Plus we have data from cblau which strongly suggest that Beckley was not fielding as many bunts as his deadball successors did.

You don't need to do mental gymnastics to puzzle this out. If you think a 37% increase in Beckley's FWS is reasonable (or a 46% increase in FWS for the average 1B playing all 162 games), then 1 per 162 is going to work for you. If you think a 211% increase in his FWS is reasonable, then 3 per 162 is going to work for you. Given that the evidence is mixed, not clear, on the question of 1B defensive value in the era, the 37% figure seems like a logical and reasonable increment to me.

Which leads back to the point earlier. 23 WS still leaves Beckley leading fewer than 5% of all pennant-winning teams and leading a substantially larger number of really poor teams.
   112. karlmagnus Posted: November 14, 2006 at 09:14 PM (#2237816)
WS has 1B 2/3 as valuable as an outfielder; that's clearly not right for the 1890s. Also, it ascribes too many WS to pitching and not enough to fielding for such an error-rich period. Your total fielding WS is 35.55; it should be more like 50 for the 1890s, with pitching and pitcher fielding reduced commensurately from 208 to 193 (underlying assumption of WS is 3 per game and offense = defense, right?). This would increase all fielders by around 40%. Instead of increasing OF, reduce them to 2.5, reflecting the mcuh lower level of OF chances. Then there are roughly 17 WS to split between catcher (about the same as today, say+1) and 4 infielders, which split evenly would be 4 each. Give Beckley somewhat fewer, to load the SS, or 3 extra WS per annum.

Your shortstops would be getting 4-5 extra WS per annum, so giving Beckley an extra 3 doesn't put him ahead of them.

That's assuming you rate WS at all, which I don't really -- the underlying assumptions are gross and make it dubious (halve WS for pre-1893 pitchers -- WHY -- it's pulled out of a hat!)
   113. karlmagnus Posted: November 14, 2006 at 09:17 PM (#2237817)
If as is possible I'm out by a factor of 2 on the total WS available, doesn't divide fielding adjustment by 2, just reduces pitchers from 86 to 71.5, which still looks reasonable for the perriod.

Damn silly system.
   114. Daryn Posted: November 14, 2006 at 09:30 PM (#2237825)
What's that translate to in terms of /162 value?
C: 6.75 FWS
1B: 2.13 FWS
2B: 5.79 FWS
3B: 4.26 FWS
SS: 6.48 FWS
OF: 3.38 FWS

karl says he values 1890 1b as somewhere between 3b and outfield. So that would be something like 1.50 to 1.75 extra WS per 162 games.
   115. Mike Emeigh Posted: November 14, 2006 at 09:40 PM (#2237835)
Which leads back to the point earlier. 23 WS still leaves Beckley leading fewer than 5% of all pennant-winning teams and leading a substantially larger number of really poor teams.

Stating that "Beckley's WS is not typical of the best player on a pennant-winning team" is IMO just another way of saying "Beckley didn't have much of a peak" - which I think everyone on both sides of the discussion has already acknowledged.

I'm still going to argue that, if the first baseman is not typically the best player on a pennant-winning team of Beckley's era, treating him as though he is supposed to be the best player on the team penalizes him for not having better teammates at the positions that are supposed to pick up Win Shares. If pennant-winning teams typically had 30 WS starting pitchers, shortstops, or center fielders, it's hardly Beckley's fault that his team wouldn't have one of those.

-- MWE
   116. Mark Shirk (jsch) Posted: November 14, 2006 at 09:48 PM (#2237844)

If this is an are where all fielders should be given a boost then why are you degreding OF WS? Shouldn't they are least stay the same from where they are in Bill James original setting?
   117. Mark Shirk (jsch) Posted: November 14, 2006 at 10:48 PM (#2237895)

Yes it is a way of saying that he didn't have a great peak. But it is also a way of demonstrating how much peak he lacked. On the 1991 thread someone compared Staub to Beckley. Staub has six seasons better than Beckley best accordint to WS, four if you add one WS per season for defensive credit. Staub's best seasons are form best to wrost, 32, 30, 28,28,27,26,25 and Beckley's best season is 26 with two 25's when adjusted for schedule and defense. So it is not like Beckley ahs a low peak but that his peak is nearly non existent. Do we want someone like that in the HOM?

Also, checkign if 1B should be leading pennant winners during Beckley's era doesn't really work becuase 1B was not a very competitive position during Beckley's time. Should Fred Tenney, Dan McGann, and Harry Davis be leading their teams? I would think that even Beckley supporters aren't supporting the trio above at all. A lack of good 1B means htat 1B won't be leading teams in the 1890's.
   118. DL from MN Posted: November 14, 2006 at 11:13 PM (#2237908)
George and Alfred Rawlings announce opening of retail sporting goods store in downtown St. Louis

The adoption of the baseball glove by baseball star Albert Spalding when he began playing first base influenced more infielders to begin using gloves. By the mid 1890s, it was the norm for players to wear gloves in the field.

By the early 1900's the glove developed full fingers and looked like an open hand and felt like an oven mitt. Heels or padding on the base of the glove appeared to give a pocket so the ball would not fall out.

Rawlings introduces historic Bill Doak glove. This revolutionary glove featured a multi-thong web laced into the first finger and thumb with a natural formed deep pocket. Becomes the prototype for all-purpose gloves of the future.

Rawlings hears fielders' plea. Develops the Trapper - a three-fingered, deep well pocket glove that changes the design of first base mitts and sets the standard for Major League specifications.

The Rawlings glove line expands to include a glove for each position - with a name, position and number.

Rawlings introduces The Playmaker, a new three-fingered glove.

By the 1950's actual deep "pockets" were being developed into the gloves with lacing and sewing that sculptured the padding. "Heels" were losing padding.

It wasn't until 1957 when the fielder's glove added a hinge that spawned additional modification such as closed backs and checkerboard webs in the 1960's. Not much changed in the next three decades.

stolen from various sources
   119. KJOK Posted: November 15, 2006 at 12:04 AM (#2237935)
What's that translate to in terms of /162 value?
C: 6.75 FWS
1B: 2.13 FWS
2B: 5.79 FWS
3B: 4.26 FWS
SS: 6.48 FWS
OF: 3.38 FWS

Is it reasonable to suggest that a 1B could get 3 WS more credit per year in the 1890s? A 241% increase in FWS? I don't see it, and Karl, I don't think it's defensible.

I'm going to try to defend Karl's position, although I'm not sure I agree with it..

Those Per 162 figures Eric posted equal 28.79 Win Shares per 162 on a team level.
I THINK we all agree that in the 19th century, Win Shares has an 'incorrect' distribution between Pitchers and Fielders.

IF the fielding win shares should be something more like 50 per 162 per team, and IF 1st base should have more of that pie, then you could have something like:

C:    11.5    FWS
1B:    4.5    FWS
2B:    9.5    FWS
3B:    9.5    FWS
SS:    11.5    FWS
OF:    3.5    FWS
   120. Mike Emeigh Posted: November 15, 2006 at 12:33 AM (#2237948)
Also, checkign if 1B should be leading pennant winners during Beckley's era doesn't really work becuase 1B was not a very competitive position during Beckley's time.

I'm well aware of that. Hence my point that it's unfair to compare Beckley's total to that of other team leaders of his era, because almost no teams expected the 1B to be the best player on the team.

-- MWE
   121. Howie Menckel Posted: November 15, 2006 at 01:11 AM (#2237971)
I don't mind the questions about Beckley's fielding bonus, although the evidence is mixed.

But this is a guy who was effective - and 120 OPS+ at a weak-hitting position is quite effective - for more than a dozen seasons. And there is not a single player from his era with that some of longevity/durability combo who is NOT a HOMer.

He also finished in the top 10 in RBI TWELVE times. Not sure that anyone else on the entire ballot can say that.

Does that mean he's a HOMer? Debatable, obviously.
Does that mean he belongs inside someone's top 100, maybe top 50, top 25?
I'd say yeah.

If your system can't put Beckley - or Dobie Moore, a very different case - in the top 100, I suggest nicely it might be time to rethink that system.
Merit comes in all sorts of forms. Beckley was effective longer - a lot longer - than his peers, and that has merit.

It's a little weird that Beckley is constantly derided for not having a peak, but does anyone get deride for not having a long career? I realize they get downgraded for the latter, but I just don't see the vitriol.
   122. karlmagnus Posted: November 15, 2006 at 01:22 AM (#2237981)
Also as a final point, if you must use WS and Beckely's 3WS short per year, then his career total is 57WS below what it should be. That puts him in much more exalted company career-wise, and perhaps makes the view of WS-users more in line with that of stats-users.

Apart from those, a substantial minority, who simply find me irritating, Beckley-haters seem to fall into two camps, those who demand a peak (for whom I can do little; his peak is indeed hard to find and not very impressive -- 1900 seems to me to have the best case) and those biblically devoted to Win Shares, for whom Beckley's career is artificially diminished by WS's errors in the pre-1920 period. The latter group need to adjust their WS numbers as discussed here to see Beckley's career in a reasonable light.
   123. Mark Shirk (jsch) Posted: November 15, 2006 at 03:45 AM (#2238179)

I am not sure it really matters what teams expected a certain position to provide in terms of total value. I mean no one expects Middle relievers to be the best players on a given team but then again we aren't electing any middle relievers. Now obivously Beckley was WAY better than any middle reliever but he should still be put up against other players who are of HOM level and most of them were good enough to be the best player on a series of pennant winners, Beckley doesn't seem to be.


I have two responses, I think that there has been a candidate that attracted a bunch of vitriol and was the exact opposite of Beckley and that was Hughie Jennings, who happens to have gotten elected anyways. And I must admit htat it is highly likely that we will one day elect Jake Beckley. But I would say that anyone with a high peak is going to have decent career numbers, with the exception of someone like Al Rosen maybe. High peak years carry a lot of weight and it takes fewer of them to get to a certain level of career value. I think that what most career guys value is not number of seasons played but instead total value given am I right? To use WS as a proxy, three 33 WS seasons is pretty much the same as five 20 WS seasons. Same career value. Charlie Keller has somewhere between 260-300 career WS depending on the credit given, even Jennings had about 250 schedule adjsuted WS I believe. I know you don't like WS, but you can use it as a prxy measurement right? On the flip side a lack of peak cannot be remedied by a longer prime or a longer career. So Sam Rice and Jake Beckley cannot have many 30 to 35 WS seasons no mater how many 18-22 WS seasons they have. So peak guys get decent career numbers, usually, by having their peak seasons but career guys don't get peak by having many decent seasons. You either have or do not have peak.


Of course if you arent' giving Beckley 3 more WS per seasons for 19 seasons (this is even more ridiculous than givin ghim 3 WS in his 3 or four best seasons) you aren't going to give him 57 more career WS. I certainly fall into the first camp of voters that demand a peak of anyone I am going to put into my PHOM but I could be accused of the second as well and I am not sure that the charge is completely fair. I think a lot of vters use WS in meta studies because it is a decent amalgamation of everything a player can do on the field. It is not perfect and none of us think it is. I use it heavily and still adjust for 3b/2b, the fact that WS likes CFers a little too much, 19th century pitcher innings, etc. However, to use something like OPS+ or hits tells you only part of what a player does and in a larger study like the one Doc did, it really tells you pretty much nothing. Therefore, I don't think there are tons of voters who are 'biblically devoted' to WS, only a group of voters who choose WS as a way to encapsulate everything a player does in one easy to use number. That said, I don't think that WS is nearly as inaccurate on Beckley as you think . WARP does have the fielding adjustment that you think WS lacks and it still doesn't like Beckley very much.
   124. Howie Menckel Posted: November 15, 2006 at 04:08 AM (#2238211)
Mark S,
That's an interesting response, but you accounting for what more than a dozen 120 OPS+ seasons were worth back then relative to position?
I'll grant you there was some Jennings vitriol, yes. But SRice didn't exceed his contemporaries by even nearly as much as Beckley did.

Again I'll say, any system that says "who cares?" for example, to a 13th-best 120 OPS+ full-time season at a position of weak offense for the era, well, you have to jump that credibility hurdle...
   125. JoeD has the Imperial March Stuck in His Head Posted: November 15, 2006 at 04:14 AM (#2238221)
I still hold that Beckley is basically Palmeiro. WARP agrees, if you adjust Beckley for schedule length.

Howie summed it up well, there is more than one way to skin a cat. The no-brainers are the guys that have the peak and the longevity.

But I still think the very good for very long time path is just as valid/valuable as the short career but huge season path.

Mark S. says that three 33's shouldn't equal five 20's, and that is true, first you have to adjust for replacement level - let's say that's 10 WS per full season. So really you should be comparing the three 33's (69 WSaR) to five 24's (70 WSaR).

But the difference is along the magnitude of a 10-15% bump for the 33 WS seasons, not along the lines of the 33's being all valuable and the 24's being useless. The three 33's are directly comparable to five 26's (80 WSaR), for example, even after adjusting for the 'big year' bump.

Once you adjust Beckley's WS for team decisions (not games, WS is based on decisions), give him a slight bonus for WS underrating 1B (either because it was more defensively important at the time, or because no one else that played 1B in his time hit very well either, compared to the era WS was calibrated for - either are valid reasons - the correct answer is probably a little of both), he ends up coming out pretty well - basically the Raffy Palmeiro of 100 years before. Never the best in the league, but generally a very good player who lasted forever. In modern times he certainly would have been a 3000 hit, 500 HR guy.
   126. Rob_Wood Posted: November 15, 2006 at 04:17 AM (#2238223)
On behalf of the pro-Beckley camp, I think it should be kept in mind that defensive stats (and especially meta-stats)
are not as accurate or precise as offensive stats. Also, as others have pointed out, the strength of Win Shares is
not in its treatment of the 19th century players, and especially not in its 19th century defensive values.

Let's not kid ourselves that the Defensive Win Shares for 19th century players are more than approximations that
are often inaccurate and sometimes wildly inaccurate. Beckley has a fairly stong offensive resume that should not
be dismissed by an over-reliance on defensive stats such as Win Shares.

In sum, as a career voter I heartily embrace Beckley's candidacy. Does he have his "cons"? Of course. But he also
has his "pros". Let's try to be fair to both sides of the argument.
   127. Howie Menckel Posted: November 15, 2006 at 04:25 AM (#2238240)
My guess is that we most us us agree that Beckley is borderline HOM.
But it may be that by happenstance, the amount of peak voters happened to outnumber the amount of career voters.
Flip it, and he may have gotten in decades ago, while Sisler roasts on the spit, ironically....
   128. mulder & scully Posted: November 15, 2006 at 09:41 AM (#2238383)
Some comments, none Win Shares related.

Beckley had 13 or 14 years over 120 in his career which is a very good total. But, my problem with this argument is that in 8 of those years, he is between 122 and 128, or he barely makes it in.
Comparing him to others:
Ed Delahanty has 11 full seasons of 120 OPS+. 10 of the 11 are over 156.
Jesse Burkett has 13 years and 8 are 140 or higher.
Dan Brouthers has 14 years of 120 OPS+ and all are over 132.
Fred Clarke has 14 years and 10 are over 130.
Roger Connor has 15 years and 13 are over 140.
Sam Crawford (who has Beckley has been favorably compared to) had 15 such years and 14 are over 130.
Joe Kelley had 12 such years and 9 were over 130.
Cap Anson had 16 such years (after 1875) and 14 are over 130.

If you instead looked at 130+ seasons, Beckley only has 5 or 6 (depending on how you count 1888).
To me, this qualification of Beckley's is similar to when someone makes the 2500 hit/300 homerun/200 steals list (for example) and 4 of 5 players are HoFers and player 5 has 2501 hits/306 homers/201 steals. Yes Beckley is one of the few players of his era with so many 120 plus seasons, but he barely meets the mark in the majority of the years.

Also, for all of Beckley's slugging prowess as demonstrated in the OPS+, he made the top 10 in his league all of 4 times. Every time in a 2 or 3 league year.
   129. DL from MN Posted: November 15, 2006 at 02:41 PM (#2238456)
> WS's errors in the pre-1920 period

Which correlate nicely to the introduction of the Bill Doak glove.
   130. Mark Shirk (jsch) Posted: November 15, 2006 at 03:53 PM (#2238533)

I am not so sure that 1B durign Beckley's time was a 'weak hitting' position. I am pretty sure that it was above average, which is far from weak. Was it what it woudl become after 1920? No, of course not but it's not like it was catcher or SS. I see three reasons for this, two of which are connected. First, Beckley's contemporaries by and large weren't very good. This he doe snot deserve credit for. Second, 1B was a different position then, one with slightly more defensive responsibility, this he shoudl get soem credit for. Third, 1B was wrongly percieved to a degree to be a defensive position when in fact it was not, or at least was not as much of a defensive position as managers thought. I wouldn't gtive credit for this. Either way, I am nto sure that calling 1B a 'weak hitting' position is entirely accurate.

When you say a 120 OPS+ seasons, exactly how much value is that above average or replacement?

Finally, while I am a peak voter, I was never a big supporter or George Sisler. I actually think that he peak wasn't that high. If he were stille eligible he would be on the edge of my ballot below guy slike Keller, Moore, and Walters.


Defensive stats may be murky but they are all we have. I don't recall hearing that Beckley was anything like Keith Hernandez or Hal Chase out there, merely a good defensive players. The stats, IIRC, say the same thing. So we can't necessarily use 'Beckley was super valuable defensively' as the null hypothesis with little evidence like karl wants to do. Also, his strong offensive resume is not so strong as to make him leap out.
   131. Dr. Chaleeko Posted: November 15, 2006 at 04:02 PM (#2238542)
120 OPS+ years aren't necessarily meaningful by themselves. Reggie Smith's full seasons, for instance:


He absolutely creams Beckley: 10 full years above 130 to five for Beckley, though Beckley, in fairness, does have three more full years over 100 than Smith does. Smith has his own issues, of course, just as Beckley does....

Howie and Rob are exactly correct that just about everyone knows/believes that Beckley is a borderline candidate. I don't think anyone hates Jake Beckley or that he's reviled in any quarters. I think the differences of opinion about him are directly related to Mark Schmeagol's assertion of no-peak, no-soup---because an individual's POV on the peak/career question is perhaps the most important pathway to defining every voter's vision of the HOM. And I think we all care A LOT about the institution, which is why all the hand-wringing over probably the most extreme guy we're going to take a serious look at. (And I'm not really a hardcore peak guy either!)

Is there room for peak and no-peak? With 8-10 backloggers, there may not be in my vision of the HOM. Which is mine only, of course! To use a bad analogy, who do I want in my pRRHOM (personal Rock 'n Roll Hall of Merit), John Mellancamp or Nirvana? Do I think ~20 albums and 30 years of better than average rock (with some nice highlights) is preferable to three or four albums' worth of truly outstanding, landscape-changing music? I'll always take Nirvana. Or the Velvets and their four records. Is My Bloody Valentine (and their two records) too little output to warrant an induction? Probably, yeah. That's even less output than Al Rosen probably. But this is the level of extremity we're talking about here when we're talking about so few slots left and given Beckley's extreme career/peak tilt.

Take the musical analogy for the less-than-perfect argument tool it is, and I'm sorry if you like Mellancamp and hate Nirvana or the VU. I admit to having a soft spot for "Little Pink Houses" and I did say he was better than average.


Now to inquire into another point of discussion. I used the SBE and bb-ref to generate a BIP/game estimate for MLB 1890-1900 and for MLB 1950-1960. I figured the 1950s would be a decent in-between point since it wasn't as pitching dominated as the 1960s and 1970s and was a fairly normative run environment, running from around 4.15 to 4.75 but usually in between (plus no DH).

From the ABs, I subtracted the league's Ks and 2/3s of the league's homers for the 1890s (figuring a much higher pct of in-the-parkers back then, but all homers for the 1950s), and added the sacs. I estimated Ks and SACs for years in which they were unavailable by using a simple average of the other years in the sample for which I had data.

For 1890-1900, I'm seeing 33.074 BIPs/g (in 20052 games)
For 1950-1960, I'm seeing 29.252 BIPs/g (in 27220 games)

That's not insubstantial, but it's not large either. The pct difference is 13% more BIPs per game. If we raised the SFWS average across the board by that same amount, then 1B would get 2.41 WS/162 for 162 g. Raising the generic average FWS listed above by a total of 13% across the board would yield 40.2 FWS/team, up from 35.55 (which includes all three OFs).

Now, what I don't know is whether or not a 13% increase in BIP leads directly to a 13% increase in FWS. I doubt it, but I don't have much speculation on it. Some items that might help: depending on whether you include the SACs in your BIP for the purposes of BIPAVG, you get .292 or .298 for the 1890s and .302 or .307 for the 1950s. ROE/BIP: I took a SWAG at ROEs in the 1890s by taking the total league ERR and mulitplying by .67. I have no idea if that's too high or too low, but given that the oral tradition suggests more running all around, it's pretty likely a lot errors were committed on base-running plays and weren't BIP-type events. (There is data available for 1957-1960 on ROE and it generally confirms .67 as a decent estimator, at least for the late 1950s).

ROE/BIP 1890s: .058
ROE/BIP 1950s: .021

If you pull the ROE out of both eras, the number of bip/game is

1890s: 31.17
1950s: 28.65

The percentage difference in BIPs is now 9%. So errors alone are accounting for about 4% of the BIP difference. That's pretty interesting I suppose. So sure-handed fielder will probably do a little better in the 1890s since he'll be a bit more better than average than the typical 1950s guy.

I'm assuming that the 13% increase in BIP provides a sense of the range for the transfer of WS from pitchers to fielders. Not the exact amount, but a point of reference. As in is it reasonable to think that the average position player will gain 3, 2, or 1 WS? This data suggests 1 WS or less at all the positions is very reasonable and that the 37% increase represented by adding 1 WS per 162 to the average 1B FWS is somewhat likley within the range of reasonable. Like I said, I don't have the exact number, I'm just trying to offer some thoughts about the magnitude of transfer we might be talking about.
   132. Mark Shirk (jsch) Posted: November 15, 2006 at 04:12 PM (#2238557)

1. good study

2. I am not a mellencamp fan at all! I used to be agnostic toward him, mostly because he was before my time. But with that 'This is my Country' song all over TV I have decided to hate him!

By the way, I am a big Nirvana fan.
   133. dan b Posted: November 15, 2006 at 05:47 PM (#2238704)
I suspect some voters are overstating the value of 19th century 1B play because of the inferior equipment. The onus would have been on the other IF to make a throw that could be caught, most likely with 2 hands. Many of the plays we now see as routine would have been infield hits with no throw attempted.
   134. DL from MN Posted: November 15, 2006 at 06:10 PM (#2238740)
Doesn't "that could be caught" depend greatly on the quality of the first baseman?
   135. Mike Emeigh Posted: November 15, 2006 at 06:34 PM (#2238767)
Now, what I don't know is whether or not a 13% increase in BIP leads directly to a 13% increase in FWS.

It doesn't. The pitching/fielding ratios - by design - vary little across eras, regardless of BIP.

From 1890-1900, it appears that pitchers accounted for 67.3% of the defensive win shares. From 1950-1960, it appears that pitchers accounted for 67.6% of the defensive win shares. I'm using the team WS files on the baseballgrahps site for these.

-- MWE
   136. jhwinfrey Posted: November 15, 2006 at 07:04 PM (#2238795)
1990 Preliminary Ballot:

1. Joe Morgan (ne): Easily in the top 5 all-time at his position.
2. Jim Palmer (ne): Racked up a lot of Gray and Black Ink in a fairly long career. A great pitcher.
3. Burleigh Grimes (4): One of the top pitchers of his era, with a long career--A sure HoMer for me.
4. Ferguson Jenkins (5): Very similar to Grimes for me--Grimes had slightly higher peak value, so he gets the edge.
5. Orlando Cepeda (6): Do I have to turn in my career-voter card for having him ahead of Beckley? Cepeda still had 14 good years (and 4 great years) so I'm not worried.
6. Jake Beckley (7): In comparison to Cepeda, Eagle Eye had 18 good seasons, 3 of them great.
7. Charley Jones (8): Continuing the comparison, Jones had 10 good seasons, 8 of which were great. Few players produce 300 win shares in 10 seasons.
8. Dick Redding (9): By my reckoning, Redding had a slightly higher peak value than Palmer, but was less league-dominant and had a shorter, less productive career.
9. Edd Roush (10): I have him as the 4th-best fielder on my ballot, behind Trouppe, Oms, and Fox.
10. Quincy Trouppe (11): Now the best catcher not elected--I don't think he's as good as Ted Simmons, though.
11. Pete Browning (12): Like Jones, a short-career player with a peak that can't be ignored.
12. Nellie Fox (13): A great gloveman with a long career and good WS total. Rabbit Maranville with more ink.
13. Reggie Smith (14): A better peak than Roush, but not quite as long a career or as good a glove.
14. Alejandro Oms (15): I'm still looking for a reason to rank him with Trouppe and Redding, but he was clearly a great player.
15. Jim Kaat (16): Kaat sneaks on the ballot with only two big-name new eligibles. It's his longevity that puts him ahead of Bucky Walters and Carl Mays for me.

Other guys from last year's top ten:
22. Ken Boyer--Boyer actually ranks higher than Wynn overall, but a positional bonus gives him the edge. I have Tommy Leach ranked ahead of him.
25. Jimmy Wynn--Very close to Amos Otis, actually. Otis has the edge with the glove, Wynn with the bat.
27. Dobie Moore--Definitely a great hitter, but not for long. Two or three more seasons and he'd be on my ballot.
   137. Mark Donelson Posted: November 15, 2006 at 08:20 PM (#2238854)
If your system can't put Beckley - or Dobie Moore, a very different case - in the top 100, I suggest nicely it might be time to rethink that system.
Merit comes in all sorts of forms.

That's very moderate of you, Howie, and overall I know the electorate agrees.

As far as merit coming in all sorts of forms, well, sure. You could say that Doug Glanville appears to be a nice, charming fellow, and that's merit of a kind, so he deserves election. Obviously that's absurd, but my point is that just because something a player does can be described as "meritorious" in some general sense doesn't mean we all have to vote for him.

Our mission, as described in the constitution, reads: "Our goal is to identify the best players of each era and elect them to the Hall of Merit." To me, the best players of each era are the ones who were the most dominant, even if was only for fairly short periods. And those who were never dominant players--and Beckley most certainly falls into this category, in my opinion--are more likely to be Hall of Very Good types to me. Not just in my system, mind you, but in my definition of "the best players of each era."

Now, I realize that I have a fairly extreme viewpoint on this subject, but it's not a rash or ill-considered one, I assure you. So no, I don't think it's time to revisit my system because Beckley's not in my top 100, any more than you should think it's time to revisit yours because you're not as high on Win Shares as most of the rest of us, say. In fact, since Beckley is precisely the kind of player I don't want to see in the HOM, it seems to make a lot of sense that he's not in my top 100.

That's not to say I'll be upset when Beckley is inevitably elected, or that I'll argue that anyone with a less peak-centric philosophy (read: just about everyone) shouldn't vote for him. There are many players already elected who are also precisely the kinds of players I don't want to see elected, and who are/were not in my top 100. And I think one of the best things about the way the HOM works is the consensus aspect. The final results are an amalgam of all our opinions--the radicals on both ends, the moderates in the middle, and everything in between.

And I don't see a problem with that. Those who vote for Beckley and don't have Moore on the radar are fine too--we just have different idea of what "the best players" means, and it all comes out in the wash when we all vote. 'Cause just as merit comes in all forms, so does personal philosophy about what makes the best ballplayers.
   138. Juan V Posted: November 15, 2006 at 08:53 PM (#2238888)
Very prelim ballot:

1) Hmmmmm... let me think about this.... maybe this Morgan dude? :)
2) TBA
3) TBA
4) TBA
5) Ale Oms
6) Gavvy Cravath
7) Jimmy Ryan
8) Ken Boyer
9) Jimmy Wynn
10) Charley Jones
11) Tony Lazzeri
12) Bob Johnson
13) Luis Tiant
14) TBA
15) TBA

2-3-4: I´m inclined towards Palmer over Jenkins, and both belong in this group. But where does Quincy Trouppe rank among them is the question I need to solve
14-15: As I said in the last ballot thread, I´m taking another look at Tenace, maybe I´m overdoing it with the catcher bonus? Fregosi will probably be the one who jumps to the ballot.
   139. TomH Posted: November 15, 2006 at 09:55 PM (#2238943)
Question for Gavy Cravath voters:

When projecting Gavy's missing years, do you account for the fact that he took extreme advantage of the tiny gimme-a-home-run Baker Bowl to generate his impressive power numbers? That we ought to agree that in another envirnoment, his home run stats likely would have been greatly diminished?

Anyone have full home/road numbers for Cravath? The only item I recall is that one year he led the league in dingers while hitting ALL of his home runs at his home park.
   140. jimd Posted: November 16, 2006 at 01:51 AM (#2239111)
I'm using the team WS files on the baseballgrahps site for these.

Interesting site. (I love graphs.)

How does one access the "team WS files"?

   141. Mike Emeigh Posted: November 16, 2006 at 02:19 AM (#2239129)
How does one access the "team WS files"? - there's a ZIP file containing historical WS through 2005, and one of the spreadsheets is historical WS for teams.

-- MWE
   142. Mike Emeigh Posted: November 16, 2006 at 02:25 AM (#2239134)
Anyone have full home/road numbers for Cravath?

You can get the breakdowns from the SABR Encyclopedia on the member site; find the player, click on Batting to get his individual totals, and then click on any HR total for any year to get into the HR log.

Cravath home/away totals:

1908: 1H
1909: 1A
1912: 6H, 5A
1913: 14H, 5A
1914: 19H, 0A
1915: 19H, 5A
1916: 8H, 3A
1917: 8H, 4A
1918: 8H, 0A
1919: 10H, 2A
1920: 1A

-- MWE
   143. Mike Emeigh Posted: November 16, 2006 at 02:26 AM (#2239137)
You can get the breakdowns from the SABR Encyclopedia on the member site

...if you are a SABR member, that is.

-- MWE
   144. TomH Posted: November 16, 2006 at 02:42 AM (#2239147)
Thanks Mike. I shoulda known that.

So, Mr Cravath in his MLB career hit 26 home runs in road games. Given a typical home-field advantage, he might have hit 60 dingers all told, instead of his actual 119. That's a whole lotta black ink that would get wiped away. Would we then be projecting a HoM career?
   145. jimd Posted: November 16, 2006 at 03:06 AM (#2239159)
Cravath had the ability to take advantage of a unique park, and real value resulted.

I see no reason to penalize him for that.

OTOH, I see no reason to give him extra credit for any value that was never actualized due to the fact that it took awhile for player and park to come together.
   146. jimd Posted: November 16, 2006 at 03:07 AM (#2239160)

Thanks Mike.
   147. JoeD has the Imperial March Stuck in His Head Posted: November 16, 2006 at 03:22 AM (#2239170)
TomH - I don't even look at black ink. I can't imagine dropping off 60 HR (many of which would have been 2B instead) is going to take that much luster off of Cravath's OPS+, etc.. In the grand scheme of things it's 5 or 6 HR a year, and that's already been accounted for in his park adjusted numbers anyway.
   148. Brent Posted: November 16, 2006 at 05:11 AM (#2239212)
According to the article in Deadball Stars of the National League, Cravath learned to hit to the opposite field to take advantage of the short porch while playing in Minneapolis, which had a ballpark (Nicollet Park) configured a lot like the Baker Bowl (279' down the right-field line with a 30' fence). When he hit 29 home runs for the Millers in 1911, it set the season record for organized baseball.

I agree with the idea that players should be evaluated based on the actual environment they played in - if they can learn to take advantage of it, real wins result.
   149. rawagman Posted: November 16, 2006 at 08:39 AM (#2239315)
OPS+ is supposed to be park-adjusted, so I give credit according to career rate stats, without much thought to ensuing black ink.
   150. TomH Posted: November 16, 2006 at 12:46 PM (#2239341)
Let me try to state it more clearly

1. Yes, OPS+ is park adjusted
2. Yes, many of his HR would have been doubels (although some would have been outs) in other parks.
3. Yes, while Cravath took more-than-normal advantage of his park, this resulted in real wins.
4. NO, park adjsutments do NOT cpature how much Cravath in particular was heled. A typical park factor for Philly was 107. If you take 3 Gavy home runs and turn them into 2 outs and 1 double, you reduce his rc/g by 7%. I hope no one will argue that Cravath would only have lost that amount of offense had his home games been elsewhere. If you take away 6 HR a year and turn them into 3 2B and 3 outs, in 500 PA, you take away 46 OPS from Cravath, and 11% of his run production.

most importantly -

4. Cravath's case depends on projecting a non-MLB career into perceived value for the years he missed. You can't simply put him in the Baker Bowl for those years and assume he would have cranked out a 150+ OPS. Because in virtually every other MLB park, he would have been less valuable. Project him in a neutral park, and imagine what numbers he would have put up. Use his MLB road stats to project him. Do SOMETHING other than imagining he would have been would have been one of the best hitters in the game; because when away from the tin box, he was not.
   151. John (You Can Call Me Grandma) Murphy Posted: November 16, 2006 at 12:56 PM (#2239345)
You can't simply put him in the Baker Bowl for those years and assume he would have cranked out a 150+ OPS.

I agree with that, Tom, and take that into account.
   152. Howie Menckel Posted: November 16, 2006 at 01:35 PM (#2239358)
Wow, 38-5 for home/road HRs in 1914-15.
I was factoring in a benefit of Baker Bowl, but not THAT much.
I'll have to adjust accordingly...
   153. rawagman Posted: November 16, 2006 at 02:04 PM (#2239373)
My 1990 prelim

Use a sort of peak-over career number that measures ink by playing time. Combined with rate stats and a glove measurement, I feel this gives me both context for what the player actually achieved versus what the league around him was able to do.
Morgan and Palmer go straight to PHOM, taking Fergie along for the ride. I liked Amos Otis. He debuts at 44. Probably not enough to ever make my PHOM or ballot. Bob Watson, Greg Luzinski and Ken Singleton make the consideration set, but not by as much as I might have thought. Singleton is between Bobby Bonds and Bobby Murcer. Both outside of my top 75. Watson looks like a poor man's Roy Sievers. Greg Luzinski look slike something between Sid Gordon and Lefty O'Doul.
Then there's Jimmy Wynn - I'm just not impressed. Not in my top 75.

1)Joe Morgan (PHOM)
2)Jim Palmer (PHOM)
3)Hugh Duffy (PHOM)
4)Fergie Jenkins (PHOM)
5)Ben Taylor (PHOM)
6)Gavvy Cravath (PHOM)
7)Lefty Gomez (PHOM)
8)Edd Roush (PHOM)
9)Nellie Fox (PHOM)
10)Quincy Trouppe (PHOM)
11)Tommy Bridges
12)Vern Stephens
((12a)Bill Freehan))
((12b)Biz Mackey))
13)Bobby Veach
((13a)Willie Stargell))
14)Orlando Cepeda
15)Ken Boyer

Next in line
16)Wally Berger
17)Reggie Smith
18)Dizzy Dean
((18a)Juan Marichal))
19)Bus Clarkson
20)Ernie Lombardi
21)Roger Bresnahan
22)Al Rosen
23)Mickey Welch
((23a)Jim Bunning))
((23b)Billy Pierce))
24)Dick Redding (PHOM)
25)Chuck Klein
26)Tony Oliva
27)Charley Jones
28)Jim Bottomley
((28a)Joe Gordon))
29)Dobie Moore
30)Addie Joss
((30a)Cupid Childs))
   154. DavidFoss Posted: November 16, 2006 at 02:37 PM (#2239412)
You can't simply put him in the Baker Bowl for those years and assume he would have cranked out a 150+ OPS

I didn't know anyone was doing this. Calling out a 'strawman' sounds harsh, but I guess that's what I'm doing. :-)

Check out Brent's MLE's from the Cravath Thread

1. Play from the AA and PCL is used to create the MLE's -- its not an extrapolation of his Phillie numbers
2. His MLE's don't have him at a 150 OPS+ level except for 1911. His 1907,1909-1911 level is nice (129,132,142,161) and complement his peak nicely but that's a bit lower (140-ish) than his career MLB rate.
   155. rawagman Posted: November 16, 2006 at 02:49 PM (#2239423)
I guess I didn't adequately explain myself.
I don't really use numbers where credit is concerned. His career is his career. I am essentially a peak voter who has a hankering for a consistent level of production. I look at career length as a measure of peak arc. When a player deserves credit (I like to think that I am fairly liberal in that regards, especially concerning old time labour practices) I add the credit time to help "fix" my arc.
   156. Mark Shirk (jsch) Posted: November 16, 2006 at 03:06 PM (#2239433)
Most of our MLE's are based on the information that we have on that player for that year, though many NeL MLE's are regressed. They weren't made out of thin air. And if what Brent said above is true, that Minneapolis had a park very similar to the Baker Bowl, I see no reason to not give him teh same kind of credit there as well.
   157. DL from MN Posted: November 16, 2006 at 04:13 PM (#2239483)
"Nicollet [Park] is best remembered for its short right-field fence, only 279 feet, 10
inches from home plate."

This is the same park where Buzz Arlett had his career year. Willie Mays hit .477 with eight home runs, 30 RBIs and 38 runs scored in 35 games. Ted Williams hit .366 with 43 home runs and 142 RBIs at age 19.

Other HoM Millers: Carl Yastrzemski, Hoyt Wilhelm, Zack Wheat, Rube Waddell, Monte Irvin, Billy Herman, Red Faber and Jimmy Collins.

Roger Bresnahan, Orlando Cepeda and Dave Bancroft also spent time in Minneapolis.

There are good photos of the park here:

The fence was not as tall for Cravath as it was in the 1954 photo. Lots of foul territory, they probably could have backed home plate up some.
   158. TomH Posted: November 16, 2006 at 09:23 PM (#2239780)
I didn't know anyone was doing this. Calling out a 'strawman' sounds harsh, but I guess that's what I'm doing. :-)

Play from the AA and PCL is used to create the MLE's -- its not an extrapolation of his Phillie numbers
His [MLEs] 1907,1909-1911 level is nice (129,132,142,161) and complement his peak nicely but that's a bit lower (140-ish) than his career MLB rate.

Good gentle shot, David. I am glad to hear no one is doing this.

My point is still that his MLEs are being projected from numbers he generated in a great power park; and maybe if he were in Boston or Chicago in 1907-1911, his MLB stats would have been much different, if we had been projecting numbers from some non-HR friendly minor league stadium.

What I would like to see is for a FO G Cravath to lay out what they think the man's MLB stats would have been, assuming a 1907-19 MLB career in a neutral park, and convince me that it is deserving of my ballot. I'm not saying this to be taunting; maybe I CAN be convinced! The MLB OPS in that period was a mere 649, so the numbers won't have to look all THAT good, althogh we'll have to remember speed and defense will not help Gavy. Compare him to contemps Sisler and Doyle and (catcher bonus) Bresnahan and Schang, and tell me he belongs.
   159. Mark Shirk (jsch) Posted: November 16, 2006 at 10:25 PM (#2239833)
I am not 100% sure exactly why his numbers should be translated into a neutral park in the way you are talking about. Sure, they should be downgraded in teh suual way, but I am not convinced that one should take away the extra value he gave his teams by taking extra advantage of the park in Minneapolis.

Could it be possible that the reason the Phillies finally signed him was that they believed he could do the same this in the Baker Bowl?
   160. TomH Posted: November 17, 2006 at 02:50 AM (#2240014)
Maybe the Phils WERE smart this way. But why, if a player was in the minors, would you use a 'tainted' sample (if indeed Minneapolis is that) of his minor league stats to project his MLB value?
   161. Chris Cobb Posted: November 17, 2006 at 03:32 AM (#2240042)

It's not a "tainted sample." It's what he actually did: the goal of MLEs is to translate what a player actually did to an equivalent performance against a different level of competition. It shows what a player's value would likely have been if the American Association or the Negro National League had actually been a major league.

To turn the question around, why would we assess a player's value in a given season by using something other than his actual statistics??

Using his Minneapolis statistics to generate his MLEs is no different than using his major-league statistics to generate his OPS+. If he gained a competitive advantage from his home ballpark that significantly exceeds both the park effect and the normal home-field advantage that players enjoy, then one is certainly entitled to argue that the player's OPS+ or MLE is not a fair measure of his ability, because he was not competing on an even playing field with other hitters. But the MLE or the OPS+ nevertheless accurately reflects the player's contextual value.

So whence merit? Does it derive from value, or ability? Do we call Cravath's unusual home run totals an accident of circumstance or a product of skill? Different voters will reach different conclusions on these questions.

I will say this: I am inclined to believe that Cravath's anomalously late peak arises when it does partly because Cravath's use of the home run as an offensive tool is so unusual in his context that it offsets the declines in skills and durability that ordinarily diminish a hitter's effectiveness as he moves through his 30s.

In other words, I don't believe that Cravath was "a freak of nature," and I don't believe the case advanced by Gadfly that Cravath had a Ruthian peak that was obscured and diminished by the long periods of adjustment that hitters need when they change leagues.

I believe that his offensive skill and style was unique in his time, and uniquely valuable, and its effect is so much greater than the other factors that it overwhelms the effects of aging.

I don't hold the uniqueness of this skill, nor its limited context, against Cravath. The context in which the skill was valuable was his context, and the skill was his skill, and I give him full credit for its value.

Does that get him onto my ballot? No, but it keeps him close.
   162. Brent Posted: November 17, 2006 at 04:09 AM (#2240064)
"Nicollet [Park] is best remembered for its short right-field fence, only 279 feet, 10
inches from home plate."

This is the same park where Buzz Arlett had his career year. Willie Mays hit .477 with eight home runs, 30 RBIs and 38 runs scored in 35 games. Ted Williams hit .366 with 43 home runs and 142 RBIs at age 19.

I'd like to remind anyone who uses or is considering using my MLEs for Cravath (or Arlett) that I did make an effort to adjust for park effects. Lacking home/road splits, I wasn't able to calculate true park factors, but I did take account of runs scored and allowed by each team relative to the rest of the league and its W-L record, which did pick up the fact that Nicollet Park was hitter-friendly.
   163. Howie Menckel Posted: November 17, 2006 at 04:10 AM (#2240065)
Chris Cobb,
That's a great note, almost like a whodunit - I don't know your final response until the last graph.

And for those newcomers, well that, to me, is how Chris C works - not with an agenda for a player and then a rationalization to support it, but following where the evidence takes him.

I'm sure he'd be the first to say that is much too insufficient to lead one's vote - but I'd also caution one would do well not to ignore it....
   164. TomH Posted: November 17, 2006 at 04:11 AM (#2240067)
why would we assess a player's value in a given season by using something other than his actual statistics?

Mythical pitcher Smokey tosses 4 fine minor league years, ages 22-25, not called up by a club that has too many riches, has no clue, whatever reason. He puts up an ERA+ of 155 in AAA, equiv to a 130 ERA+ in MLB. His main weapon is a great fastball, which rings up a ton of Ks.

Especially at home. His home minor league park is notorious for bad visibility, making it in general friendly to pitching (lowers ERAs by 20% in all home games). But Smokey has an ERA FIFTY percent lower at home those years. He is an average pitcher on the road.

He finally gets to the majors, and becomes an average MLB pithcer for a year. Then gets hurt. And gets traded. To a team with... a notroiously poor visibility park. Once again, he turns into a monster. His wise manager uses him at home as often as possible, and in twilight games, whenever baters can't see as well. He returns real value to his MLB team, putting up a 130 ERA+ for 8 years.

Now, we realize he was 'stuck' in the minors. We wish to estimate his MLB value for those years. Do we

a. simply translate his real, actual, minor league stats, or

b. do we realize that we can't pretend he would have been in a park in the majors so good for him, and instead project what he would have done in a typical MLB stadium?

I think b. is the answer, and this is why we would use something other than a player's actual statistics to assess a player's value. And I would be surprised if many people, once they think about it, would not agree with me.

We had this discussion in another form 100 'years' ago with Ross Barnes. We found that Barnes, in the end, likely would have been a very god player even without the fair-foul bunt. But most of us would not put him in the upper half of the HoM, which is where his stats would place him.
   165. Mark Shirk (jsch) Posted: November 17, 2006 at 04:36 AM (#2240096)
But our MLE's are about what a player did, not what a player would have done. Same with our NeL MLE's we translated the value that a player had into a tougher context not the value a player would have had. I see no reason not to give a player credit for value that he had.
   166. Brent Posted: November 17, 2006 at 05:15 AM (#2240138)
Do we

a. simply translate his real, actual, minor league stats, or

b. do we realize that we can't pretend he would have been in a park in the majors so good for him, and instead project what he would have done in a typical MLB stadium?

For Cravath's era, I don't agree with the premise that minor league games were meaningless and that the minor leaguers stats were only meaningful if we "pretend" that they had been played in the majors. An AA championship for the Millers was just as important to the residents of Minneapolis in 1911 as an AL division title would be for the Twins would be in 2006. I want to evaluate those seasons based on what actually happened, then adjust for context, which includes adjustments for offensive context, park effects, and quality of play. But for me, making those adjustments doesn't involve projecting him into an entirely different environment - if he was able to take special advantage of his playing environment, so be it - I think he deserves the credit.

However, if you do want to project his performance to a neutral ballpark, Cravath's statistics for Boston in 1908 and for Los Angeles in 1906-07 suggest that he still would have hit with considerable power, though he might have lost 10 to 15 points in OPS+. A more exact estimate will have to wait until retrosheet extends its coverage to the teens.
   167. DL from MN Posted: November 17, 2006 at 01:59 PM (#2240273)
All evidence points to tremendous (relative to era) opposite field power for Cravath. He hit a ton of doubles also.
   168. Dr. Chaleeko Posted: November 20, 2006 at 11:01 PM (#2242568)
I think there's another question about Cravath, call it the Coors Field Cwestion.

When Coors was still Coors, and the team's hitters go on the road, they have a much bigger than normal H/R split. Could the same have happened to Cravath? Or to the Phils as a team?
   169. Chris Cobb Posted: November 21, 2006 at 04:10 AM (#2242749)
Over on the ballot thread, I was asked (1) why I have only one pitcher aside from the two who will be elected this year on my ballot and (2) why I have Tommy Bridges so low.


(1) Positional balance in the HoM received quite a bit of discussion a year or two ago, and there I opined that we had done pretty well in selecting pitchers. Our numbers are in line with my sense of positional balance. We have elected a lot of pitchers from the backlog recently, new and from the backlog, and I have supported them. Counting backwards, Pierce, Waddell, and Mendez were all on my ballot. Before that, we elected Gibson, Marichal, Wilhelm, Bunning, Drysdale, some of whom are towards the lower end of pitchers in the HoM, on the first ballot. All were on my ballot. So I don't think there is evidence that the shortage of pitchers on my ballot has to do with my not being willing to support pitchers. Having 3 pitchers on the ballot in 1990, with 2 of them at #2 and #3, doesn't suggest that I am lowballing pitchers as a group.

So the question is, then, why do I rate this or that particular pitcher lower than some others do? Daryn (after checking his 1989 ballot) appears to have the following 5 pitchers on ballot that I don't: Mickey Welch, Dick Redding, Burleigh Grimes, Addie Joss. I was also asked about Tommy Bridges, so let's take each in turn.

Mickey Welch: I agree that by RSI, he looks pretty good, but looked at in terms of component stats, he doesn't look so good. His career is fairly comparable to those of many 1880s elected pitchers, but also to some unelecteds who are almost entirely off the radar--Jim McCormick and Tony Mullane--both of whom I prefer to Welch. His peak, however, is definitely weaker than any of the elected 1880s pitchers. Ranking pre-1893 pitchers is one of the most difficult tasks we face, but I don't see what would make Welch the best of those still eligible, and I don't see him as having been that much of an impact player at his position. Show me that Welch is really the best member of this trio, and then I'm ready to talk about whether he or Charley Jones deserves most support among remaining 1880s stars.

Dick Redding: I supported Redding for a long time, but, in my view, his case depends on his performance in the early 1920s. If he was still an outstanding pitcher for the Bacharach Giants during those seasons, then he would have a prime of 10+ years, which together with his reputed peak in the teens, would justify his election. When the most complete numbers we had for those seasons were Holway's, Redding's case looked strong. The HoF numbers, corroborated in part by the even more reliable data from Gary A., just don't support the view of those seasons created by Holway's data. Redding doesn't look like he was much better than average after 1920, and his pre-1920 numbers, limited though they are, show a few probably brilliant seasons, but others not so much. He didn't demonstrate the sustained peak that made Mendez electable, and without the early 1920s to pad his resume, I don't see him as having enough. My policy in dealing with NeL candidates throughout has been to go with numbers over reputation, and at this point I don't believe Redding has the numbers, though I am keeping him close, because the case isn't clear cut. What numbers _should_ be drawn on to evaluate Redding?

Burleigh Grimes: He's not very different from Eppa Rixey, but at this point small differences lead to many ballot spots' difference. Grimes is _just_ outside the HoM as I see it. If the HoM's profile (and my own profile on pitchers) offered persuasive evidence that they were being underrated, Grimes is one of five pitchers who would contend for ballot spots, along with Redding, Kaat, Newcombe, and Shocker. Among remaining 1920s stars, I think Roush, Oms, and Bancroft had more valuable careers. Grimes blew hot and cold. Peak rate over a consecutive set of at least five seasons is one of my three measures, and Grimes' failure to string together a sequence of strong seasons without mediocre or poor seasons mixed in brings him down just a little bit. I've supported Grimes in the past, and he's not far off my ballot, but I would need a good reason to reconsider my system's evaluation of pitchers to move him back into contention. I think so far we have too many outfielders, too few infielders, and about the right number of pitchers. What's the evidence that pitchers are underrated as a group?

Addie Joss: He is not close to my ballot, and I have never supported him. I use DERA, not ERA+, which overrates Joss, so I take the ERA+ 142 with a big grain of salt. His IP just aren't special for his era, so he doesn't have all that many big years. I can't see changing my mind on this one: he's a long way out of contention.

Tommy Bridges: The case against him is similar to the case against Joss. Yes, he was a highly effective pitcher, but he didn't throw all that many innings, either for his career (though his career is quite substantial compared to Joss's) or in any given season. I think in-season durability is quite important to a pitcher's value, and Bridges doesn't show that quality much. I give him a little war credit, but I am not convinced that he would have keep pitching at his 1943 level for two more years, and he would need to do that to make my ballot. He is closer than Joss to my ballot, but I'm a long way from being convinced here, also.

I hope that clarifies my position on these pitchers. I would be happy to go into more detail on the numbers, if desired. I think it is good to look at numbers, but as we don't all look at the same ones, I tend not to advance them in an argument unless what they show is very clear-cut even in fairly uncomplicated metrics. I will note, that, insofar as these questions are being asked as examples of ways in which my ballot might be subjected to questioning similar to the questions I put to yest's ballot, I would add that my position on none of these pitchers is anomalous, as far as I know.
   170. sunnyday2 Posted: November 21, 2006 at 12:04 PM (#2242815)
Chris, how effective would Joss or Bridges have to be to be HoMers with their IP totals? Or can they not get there from here?
   171. DL from MN Posted: November 21, 2006 at 02:25 PM (#2242845)
Thanks, I understand why we differ on Bridges now.
   172. Chris Cobb Posted: November 21, 2006 at 04:48 PM (#2242956)
Chris, how effective would Joss or Bridges have to be to be HoMers with their IP totals? Or can they not get there from here?

I'm not sure whether you are asking for my standards or my sense of the Hall of Merit's collective standards. I don't believe that either has too few innings to be electable, simply because of shortness of career, _if_ they were very highly effective and if they had big seasons.

To speak of the Hall's standards first. We elected Koufax, who pitched about the same number of innings as Joss. We also elected Wes Ferrell, who has fewer IP than Bridges, even if Bridges gets no war credit. I supported Koufax and Ferrell for election, so my standards are pretty similar to the HoM's on this point. So the scope of their careers isn't itself a deal breaker.

But with a short career, the effectiveness bar has to be set pretty high. Joss would need to be as good as Koufax, I think. His ERA+ looks better, but (1) he wasn't the workhorse that Koufax was during his peak, (2) before you adjust for fielding support, his effectiveness looks about the same as Koufax's, but after you adjust for fielding support, he doesn't, and (3) there's the whole issue of whether an IP in the deadball era is as valuable as an IP in the live-ball era.

I think everyone agrees on (1) (though some don't care), and some people agree on (2) and (3), though some disagree. But I think Joss would have to have bigger seasons and more effectiveness to be HoM-worthy with his IP totals, and even in that case, his era would be against him.

Bridges is closer, I think. Speaking now about my system, I think that if his IP, run support, and defensive support were just as they are now, if he a record of 206-126 rather than 194-138, that is, if he added twelve wins to his teams in his career from his own accomplishments as a player, then he would be a HoMer. In terms of DERA, I think that if his career DERA had been about 3.60 or so rather than 3.76, he would be a HoMer. He wouldn't need big seasons, and he wouldn't need to be as effective as Koufax (DERA 3.50), but, given his in-season durability and career length, he would need to be closer to Koufax in effectiveness than he is in order to be worthy.
   173. Daryn Posted: November 22, 2006 at 04:09 AM (#2243617)
Thanks, Chris. I know you acknowledge that the differences are minute, but I still don't see the differences others see between Rixey and Faber on the one hand, and Grimes on the other.
   174. fra paolo Posted: November 27, 2006 at 12:25 AM (#2246206)
I vote on the basis of achievements during prime, working on the basis of an average season during the prime, and defining prime differently for pitchers, hitting positions (1b+OF) and fielding positions (remaining IF). I also prefer, where possible, to use OPS+ relative to position, than to overall League.
1 Joe Morgan The Best Secondbaseman Ever? Maybe. The Worst Commentator Ever? No.
2 Jim Palmer He's got two of those peak seasons I like, 1973 and 1975, and packs a lot of value in a 10-year prime.
3 Jimmy Wynn Way underrated. I think he provides an excellent combination of batting and fielding production.
4 Ken Boyer Boyer is an excellent third baseman in an 11-year prime, and is only a little less valuable than Childs. Another underrated candidate.
5 Thurman Munson Closer to Freehan over prime than people seem to think, I rate these two as of equal value. If you voted for Freehan, but have not voted for Munson, I'd urge you to look again.
6 Charley Jones A dominant bat in his era, given a boost because of his missing years owing to a salary dispute.
7 Alejandro Oms Oms beats out a crowded field of outer circle HoMer types because he has got the longest prime.
8 Orlando Cepeda By virtue of height of prime, he's first at first at the moment.
9 Dave Bancroft I think he's the best shortstop in the backlog at the moment. His main drawback is a lack of playing time.
10 Bucky Walters. Equivalent to Pierce over his prime, but offers more high-impact seasons in contrast to greater consistency.
11 Pete Browning He's always been in contention for a place on my ballot, and I've finally decided to rank him ahead of Roush.
12 Bill Mazeroski By the measures I'm using, he adds more value as a fielder than any other player I've ever examined.
13 Elston Howard He is a very good catcher, comparable to Freehan and Munson, mainly owing to his defense.
14 Sal Bando My positional balance fetish returns Bando to my ballot. A very good bat married to a very poor glove. By my preferred batting measure, OPS+ against League position, he scores way ahead of Boyer (201 in 1973!). But his Fielding Runs are atrocious, and I give a lot of weight to fielding at 3b.
15 Tommy Bridges I like him better than Trout, and he's very close to Walters, but only just beats Shocker to a ballot place.
   175. John (You Can Call Me Grandma) Murphy Posted: November 27, 2006 at 01:48 AM (#2246244)
fra paolo, is this your prelim or did you post your ballot here accidently? BTW, did you forget Jenkins?
   176. rawagman Posted: November 27, 2006 at 05:51 AM (#2246331)
from fra paolo's 1989 ballot
Esteemed newcomers:
Gaylord Perry has a lot of value, but I'm getting mixed readings on how he stacks up against ballot incumbents Walters and Bridges. So I'm giving the incumbents the benefit of the doubt pending further review if Perry remains unelected this year. Ferguson Jenkins isn't really peak enough for me.
   177. fra paolo Posted: November 27, 2006 at 09:46 AM (#2246366)
John, It's a preliminary. No, I didn't forget Jenkins, as rawagman has noted.
   178. DL from MN Posted: November 27, 2006 at 09:57 PM (#2246780)
2007 HoF ballot announced (years on ballot):

• Harold Baines (1)
• Albert Belle (2)
• Dante Bichette (1)
• Bert Blyleven (10)
• Bobby Bonilla (1)
• Scott Brosius (1)
• Jay Buhner (1)
• Ken Caminiti (1)
• Jose Canseco (1)
• Dave Concepcion (14)
• Eric Davis (1)
• Andre Dawson (6)
• Tony Fernandez (1)
• Steve Garvey (15)
• Rich Gossage (8)
• Tony Gwynn (1)
• Orel Hershiser (2)
• Tommy John (13)
• Wally Joyner (1)
• Don Mattingly (7)
• Mark McGwire (1)
• Jack Morris (8)
• Dale Murphy (9)
• Paul O'Neill (1)
• Dave Parker (11)
• Jim Rice (13)
• Cal Ripken (1)
• Bret Saberhagen (1)
• Lee Smith (5)
• Alan Trammell (6)
• Devon White (1)
• Bobby Witt (1)
Page 2 of 2 pages  < 1 2

You must be Registered and Logged In to post comments.



<< Back to main

BBTF Partner

Support BBTF


Thanks to
Harry Balsagne
for his generous support.


You must be logged in to view your Bookmarks.

Hot Topics

Latin American Players and Pitchers Home Page
(60 - 4:05pm, Feb 17)
Last: Dr. Chaleeko

Luke Easter
(111 - 4:02pm, Feb 17)
Last: Dr. Chaleeko

Elston Howard
(48 - 9:47pm, Feb 13)
Last: Dr. Chaleeko

Bruce Petway
(19 - 9:46pm, Feb 13)
Last: Dr. Chaleeko

John Donaldson
(13 - 9:19pm, Feb 13)
Last: Dr. Chaleeko

Bill Byrd
(25 - 9:18pm, Feb 13)
Last: Dr. Chaleeko

Dave Brown
(15 - 9:17pm, Feb 13)
Last: Dr. Chaleeko

Most Meritorious Player: 1940 Discussion
(26 - 12:19pm, Feb 12)
Last: DL from MN

2019 Hall of Merit Ballot Discussion
(103 - 5:49pm, Feb 09)
Last: Dr. Chaleeko

Most Meritorious Player: 2011 Results
(8 - 10:26am, Feb 08)
Last: DL from MN

Most Meritorious Player: Trophy Case
(39 - 8:06pm, Feb 07)
Last: Chris Fluit

Most Meritorious Player: 2011 Ballot
(23 - 7:47pm, Feb 07)
Last: Chris Fluit

2018 Hall of Merit Election Results: Chipper, Thome, Rolen and Vlad Elected!
(32 - 4:48pm, Feb 07)
Last: DL from MN

Andy Pettitte
(51 - 4:27pm, Feb 02)
Last: Carl Goetz

Most Meritorious Player: 2011 Discussion
(90 - 1:22pm, Jan 22)
Last: DL from MN


Page rendered in 1.0423 seconds
42 querie(s) executed