2005 Results: Boggs Gets 100%, While Browning and Dawson Receive Hall of Merit Honors, Too!
In his first year of eligibility, legendary third baseman Wade Boggs received 100% of all possible points to become the 14th unanimous selection in Hall of Merit history (past unanimous selections include Hank Aaron, Joe DiMaggio, Lou Gehrig, Lefty Grove, Walter Johnson, Mickey Mantle, Stan Musial, Willie Mays, Babe Ruth, Mike Schmidt, Honus Wagner, Ted Williams and Cy Young).
It was a “slightly” longer wait for star batsman Pete Browning as he was inducted into the HoM in his 107th year on the ballot. He received 28% of all possible points.
Last but not least, All-Star outfielder Andre Dawson claimed the final spot for immortality in his 4th year of eligibility, narrowly besting fellow outfielders Bob Johnson and Alejandro Oms by only a handful of points (the latter two appear to be favorites to enter the HoM themselves in 2006). He earned 25% of all possible points.
Rounding out the top-ten were: Reggie Smith (huge jump!), Bucky Walters, Cannonball Dick Redding, Kirby Puckett (surprising finish after his 2004 showing) and Gavvy Cravath.
Thanks to OCF and Ron for their help with the tally.
RK LY Player PTS Bal 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
----------------------------------------------------------------------------------------
1 n/e Wade Boggs 1296 54 54
2 4 Pete Browning 364 22 6 2 2 3 2 1 3 2 1
3 7 Andre Dawson 326 23 3 2 2 1 2 1 2 4 2 2 1 1
----------------------------------------------------------------------------------------
4 6 Bob Johnson 322 25 2 5 2 2 4 2 2 2 1 3
5 8 Alejandro Oms 316 25 3 2 1 1 1 2 3 2 1 4 1 4
6 14 Reggie Smith 279 20 1 2 2 2 3 1 3 1 2 1 1 1
7 10 Bucky Walters 278 18 1 3 3 2 1 2 2 1 1 2
8 9 Cannonball Dick Redding 273 15 5 3 1 1 1 2 1 1
9 5 Kirby Puckett 269 21 2 1 3 1 1 1 3 2 2 1 4
10 13 Gavvy Cravath 261 21 2 1 2 1 3 2 4 2 1 2 1
----------------------------------------------------------------------------------------
11 12 Tony Perez 261 17 1 5 1 3 1 2 1 3
12 11 Hugh Duffy 241 16 1 3 1 1 2 2 1 1 3 1
13 15 Tommy Leach 239 16 1 2 1 3 3 1 2 1 2
14 n/e Bret Saberhagen 237 18 2 1 1 2 2 1 1 2 1 1 2 2
15 17 Luis Tiant 232 20 2 1 1 1 2 4 2 1 4 2
16 18 Graig Nettles 232 19 1 1 2 4 3 2 2 3 1
17 19 Phil Rizzuto 211 15 2 2 2 1 2 1 2 1 1 1
18 22 Ken Singleton 207 18 2 2 1 1 1 2 3 2 2 2
19 21 George Van Haltren 191 12 2 2 1 2 1 1 1 1 1
20 23 Bus Clarkson 189 14 2 1 4 1 1 1 1 3
21 20 Dizzy Dean 189 12 3 1 1 1 1 1 2 2
22 16 John McGraw 187 11 3 1 2 2 1 1 1
23 26 Tommy Bridges 159 10 1 1 4 1 1 1 1
24 24 Mickey Welch 158 11 1 1 1 1 3 1 1 1 1
25 29T Burleigh Grimes 153 13 1 1 3 2 2 2 2
26 28 Dave Concepción 140 11 1 1 1 1 2 1 1 1 1 1
27 32 Larry Doyle 137 10 2 2 1 1 1 3
28 27 Vic Willis 135 9 1 1 1 1 1 1 2 1
29 31 Dale Murphy 134 12 2 3 2 1 1 1 2
30 29T Orlando Cepeda 133 11 3 2 1 1 2 2
31 34 Elston Howard 130 12 3 2 2 3 1 1
32 25 Lou Brock 130 9 1 1 1 1 2 1 1 1
33 33 Rusty Staub 126 10 1 2 1 2 3 1
34 37 Tommy John 125 8 3 1 1 2 1
35 36 Bobby Bonds 121 10 1 2 1 1 2 1 1 1
36 38 Bob Elliott 114 10 1 2 1 2 1 2 1
37 35 Norm Cash 113 9 1 1 3 1 2 1
38 39 Ben Taylor 102 8 1 1 2 1 1 1 1
39 48 Pie Traynor 98 10 1 1 1 1 1 4 1
40 41 Carl Mays 90 8 2 1 1 1 1 1 1
41 42 Wally Schang 87 6 1 1 1 1 1 1
42 45 Don Newcombe 86 8 2 1 1 1 2 1
43 50T Lee Smith 82 5 2 1 1 1
44 40 Dave Bancroft 77 7 2 1 1 2 1
45 47 Vern Stephens 70 6 1 1 2 1 1
46T 46 Chuck Klein 67 5 1 1 2 1
46T 49 Rick Reuschel 67 5 1 1 2 1
48 44 Bill Monroe 65 5 2 1 1 1
49 95T Bert Campaneris 64 5 2 1 1 1
50 54 Ed Williamson 64 4 1 1 1 1
51 52 Sal Bando 62 5 2 2 1
52 50T Frank Tanana 61 4 1 2 1
53 62T Urban Shocker 60 6 1 2 1 2
54 43 Don Mattingly 59 5 1 1 2 1
55 61 Johnny Pesky 56 6 1 1 1 2 1
56 55 Addie Joss 52 4 1 1 1 1
57 53 Thurman Munson 49 5 1 2 2
58 69T Leroy Matlock 49 4 2 1 1
59 71 Jack Quinn 48 4 1 1 2
60 60 Ernie Lombardi 46 4 1 1 1 1
61 59 Wilbur Cooper 46 3 1 1 1
62 62T George J. Burns 45 4 1 2 1
63 58 Tony Oliva 45 3 1 1 1
64 65T Lance Parrish 44 4 1 1 1 1
65T 56T Frank Chance 40 4 1 1 1 1
65T 72T Al Rosen 40 4 2 2
67 67 Tony Mullane 38 3 1 1 1
68T 56T Buddy Bell 36 3 1 1 1
68T 69T Rabbit Maranville 36 3 1 1 1
70 64 Lefty Gomez 35 4 1 1 2
71 81 Fred Dunlap 33 4 1 1 2
72 75 Bruce Sutter 32 3 1 1 1
73 74 Frank Howard 30 3 1 1 1
74T 65T Ed Cicotte 30 2 1 1
74T 76 Jimmy Ryan 30 2 1 1
76 77 Bobby Veach 29 3 2 1
77 68 Ron Cey 28 4 1 3
78 78 Jim Kaat 25 2 1 1
79 84T Jim Rice 24 3 1 1 1
80 72T Jack Clark 22 2 1 1
81 82T Luke Easter 22 1 1
82 80 Sam Rice 21 2 1 1
83 84T Dave Parker 18 2 2
84 82T Brian Downing 17 1 1
85 86 Luis Aparicio 16 1 1
86 79 Bill Mazeroski 15 1 1
87T 93T Tommy Bond 13 1 1
87T 87 Sam Leever 13 1 1
87T 88T Carlos Morán 13 1 1
90T 91T Tony Lazzeri 12 1 1
90T n/e Virgil Trucks 12 1 1
92T 91T Fielder Jones 11 1 1
92T 88T Hack Wilson 11 1 1
92T 88T Dizzy Trout 11 1 1
95T 100 Dick Lundy 10 1 1
95T 95T Jack Morris 10 1 1
97T 93T Brett Butler 9 1 1
97T 95T Mickey Vernon 9 1 1
99T 99 Elmer Smith 8 1 1
99T 95T Jim Fregosi 8 1 1
99T 100 George Kell 8 1 1
99T n/e Dolf Luque 8 1 1
103T 100 Charlie Hough 7 1 1
103T 104 Bill Madlock 7 1 1
103T 100 Gene Tenace 7 1 1
103T n/e Jim Whitney 7 1 1
107T 104 Dutch Leonard 6 1 1
107T 104 Dennis Martinez 6 1 1
107T 104 Al Oliver 6 1 1
Ballots Cast: 54
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.
The attempt to reach consensus tends to produce a "consensus backlog" dominated by good-but-not-great players with a 1-dimensional feel to them. I would suggest that the last few HOM members should be players who combined skills, without necessarily floating to the top in heavily quantitative systems that measure 1 dimenion. Thus hitters at glove positions (Lombardi, Schang, Stephens)players who hit and pitched (Smith, Van Haltren, maybe someone off our radar screens like Kid Gleason), pitchers who could hit (Mays) or players whose career was cut short by FATE (Joss, Cicotte, Leever, FATE in three different guises) are the ones we have overlooked, because their abilities don't pop out in statisitical studies focused on one dimesnsion. As Bill James said, low similarity scores are an indication of quality.
Fate dealt me a stinky hand by holding back the talent that I needed. John Murphy for the Hall of Merit!
As for pitchers who could hit, I tossed Newcombe a vote last election thinking precisely of his superb hitting (along with the low pitching stdev of the 50s NL and his being blocked, which you could consider "fate" as well).
Cicotte's case wasn't fate--he chose to throw the World Series! (That said, I've voted for Cicotte). How do you determine when to invoke your "fate" clause and when not to? John Murphy is right--fate determined that none of us were major league baseball players; that doesn't make us Meritorious. Even limiting it to guys who did reach the majors, guys' careers have been cut short for God knows how many reasons...why Sam Leever and not Tony Conigliaro? His most similar player at age 21 was Mickey Mantle and at 22 was Frank Robinson...
It should be pointed out that, from the perspective of real science (formally known as the natural sciences), social science as a whole is an interesting hypothesis with some empirical support.
1. Ty Cobb, 29.2
2. Mel Ott, 25.0
3. Ted Williams, 24.2
4. Arky Vaughan, 23.0
5. Joe Jackson, 21.7
6. Mickey Mantle, 20.7
7. Alex Rodríguez, 20.5
8. Jimmie Foxx, 20.3
9. Sherry Magee, 19.4
10. Ken Griffey, Jr., 18.2
11. Rogers Hornsby, 17.5
12. Eddie Mathews, 17.4
13. Stan Musial, 17.2
14. Tris Speaker, 17.1
15. Vada Pinson, 16.9
16. Al Kaline, 16.6
17. Jimmy Sheckard, 15.9
18. Rickey Henderson, 15.7
19. César Cedeño, 15.5
20. John McGraw, 15.5 (would be higher if his 1891 and 92 were included)
21. Cal Ripken, 15.1
22. Johnny Bench, 14.8
23. Frank Robinson, 14.7
24. Fred Lindstrom, 14.5
25. Albert Pujols, 13.8
26. Donie Bush, 13.7
27. Andruw Jones, 13.6
28. Stuffy McInnis, 13.5
29. Robin Yount, 13.4
30. Hank Aaron, 13.0
31. Jim Fregosi, 12.5
32. Joe DiMaggio, 12.3
33. Eddie Collins, 12.2
34. Sam Crawford, 12.2
35. Tony Conigliaro, 12.0
36. Joe Kelley, 11.8
37. Buddy Lewis, 10.7
38. Chris Speier, 10.6
39. Alan Trammell, 10.4
40. Bob Horner, 10.3
41. Joe Morgan, 10.3
42. Ross Youngs, 10.0
Some interesting names. Wow, through age 26 (1965) Pinson sure looked like HoM'er...what happened to him? He fell off a cliff just as he should have hit his peak. Cedeño's case is famous--shooting someone, even accidentally, is bad for your HoM prospects. Did McGraw quit by age 30 just so he could dedicate himself to managing full-time, or could his body just not take it anymore? As far as I'm concerned, he'd have been an *inner circle* HoM'er if he had played into his 30s at anything like the level of his 20s. Lindstrom seems to have been very up-and-down, two big years in '28 and '30 and then not much more to write home about. Donie Bush seems like a dead ringer for many great-glove, average-bat shortstops I think are Meritorious really throughout his 20s, but he fell apart with the live ball era--was it injuries or could he not adjust to the new game?
Conigliaro had a great start to his career, no doubt, but it's not like he was a LOCK for a superstar career based on what he had accomplished up to that point.
I had never heard of Buddy Lewis before he popped up on this list.
After major injuries in 1900 and 1901 his legs were in bad shape. In some respects he couldn't have been the same player; a clear case where Addie Joss is murky.
I rate players in a top-down way. The "replacement level" concept plays no role in my system.
I use Win Shares per plate appearance. The system needs to use a comprehensive rate stat. In spite of the flaws of WS, at least the methodology is published in full, unlike that of WARP, which may be superior, but I can't tell.
Anyway, #211 is an attempt to put rates of performance across seasons on the same scale. To compare careers, consider a plot where the x-axis is playing time in units of full seasons, and the y-axis is the average performance over time as rated by the rate of performance metric. All consecutive season streaks of all players can be represented as points on this plot. Careers are compared, finally, by compressing the two-dimensional plot to a one-dimensional career rating, and taking the optimum result for each player.
The advantages of the system are that the units of the y-axis (i.e. replacement level) don't matter, that bad years at the beginning or end of a career won't hurt a player, that the system could be applied to other sports (think for example of golf: to compile a quantitative list of greatest golfers, one would probably not try to define a "replacement level golfer"), and that the system, imo, captures the way in which HoF voters "think".
The disadvantage, aside from the numerical uncertainties, etc,, that apply to all systems, is that the way the two-dimensional career data strength and length is projected to a single rating is arbitrary. I tried to see if HoF voting patterns defined a sharp in/out line, but, not surprisingly, the HoF doesn't define a clear "peak vs. career" criterion. An empirical but perhaps not arbitrary criterion that I could use would be to set the rating so that, e.g., the 100th best performance of all time over each length interval get the same career rating. The problem with this is that it leads to results that appear much too peak-extreme and subject to fluke excellent seasons. In practice, I'm using a compromise between the voting patterns as best defined by the HoF and the rarity of performance rate over the length interval.
Bill James is very smart, but advanced math is not one of his strong points. If Win Shares did not have a replacement level, it would be IMPOSSIBLE to get negative Win Shares (well, it is impossible, but only because James ROUNDS THEM UP TO ZERO). I think it's been mathematically shown that James used MARGINAL Wins, which by definition will have some replacement level.
Win Shares asserts that replacement players have value. That value is determined by however many wins a team of replacement players would win in a given era (it might be the same across baseball eras, it might not; James makes no comment about that). This "value below replacement" is awarded to the players who actually played; those who didn't play can't receive that value, though they could have if only the manager had played them instead.
If you believe that a replacement team should be able to play .250 ball, then that has a consequence in Win Shares. It implies that 50% of the league total Win Shares are awarded based on playing time, the other 50% is awarded based on value added over and above replacement.
Let's see:
a) Replacement level is the zero level in some baseball measurement systems, and value below replacement level is possible
b) Win shares also has a zero level, and negative win shares are also possible (before adjustment).
c) Therefore, the zero level of win shares is replacement level.
I don't know much about advanced math, but something seems to be missing in this "proof."
Yeah, Brent. Win Shares sets its offensive zero point at 52% of league average. Everything above that counts towards Batting Win Shares, everything below that is given 0 (see Bill Bergen, who of course should have negative Batting WS but doesn't, leading his teammates to be incorrectly penalized). As the great Bill James himself would say, if that's not a replacement level, I'm a lug nut.
Rather than tell us what the great Bill James would say, why don't we read what he actually did say (p. 107 of Win Shares):
"For many years, I have argued that 'value' for a baseball player consists of being better than a replacement-level player. In fact...I hope I am not confusing the record here, but I think this was one of my ideas, this 'replacement level' concept for measuring a player's value. If somebody else was writing about this first, I apologize, but I think it was mine.
"Anyway, the Win Shares system operates on an entirelydifferent theory of value—that value consists in winning games. Two distinct theories:
"1. That value consists in being better than replacement level,
"2. That value consists in winning games for your team.
"Because I have in the past advocated theory (1), or because there is a 'marginal' concept within Win Shares, or for some other reason, some people have assumed, incorrectly, that Win Shares are a way of measuring marginal win contributions, or contributions above replacement level. Not isolated people—lots of people. There has been discussion of Win Shares on bulletin boards and in similar forums, much of which assumes that Win Shares represent marginal win contributions, and debates the system based on that assumption.
"No, no, no...this doesn't have anything to do with marginal win contributions. There is no 'replacement level' contemplated within this system. Our theory here is that value consists in winning games."
And on p. 109:
"Look, I acknowledge that there is some difference between "Wins" (or Win Shares) and "Value." In the other book, the New Historical Abstract, I didn't always rate the players according to their Career Win Shares. Win Shares are like Wins. Paul Splittorff won more games in his career than Sandy Koufax. Nobody believes that this makes Paul Splittorff a greater pitcher than Sandy Koufax. You have to consider Win Shares, Loss Shares, big-impact seasons, impact on pennant races, special contributions, etc."
There's more, but it's clear to me that (a) James understands replacement level, (b) is aware that it's possible to interpret WS as a value-above-replacement-level system with extremely low replacement level, (c) rejects that interpretation because he did not intend for replacement level to be the zero point in his system, and (d) recognizes that simply using raw WS would overstate the career value of long-career, low-value players and makes adjustments for this factor in actually evaluating players.
James sets out to justify the discrepancy between the concept (value consists in winning games) and the execution (winning games below a certain rate doesn't count) by admitting the system breaks down at the extremes but that it does a good-enough job of allocating responsibility for wins in the vast majority of actual cases.
The more I look at how win shares has worked in a lot of actual cases, the less I am convinced that it does a "good-enough" job for our purposes.
The catch is that there are systems out there that claim greater perfection, such that in theory one can vote the numbers straight up. This is where I get off. I think one needs a bullshirt dump no matter what statistical system you use as your main statistical tool and the vast majority of us have one. My sense is those who claim not to have a bullshirt dump have just buried it in the numbers in the form of this assumption or that.
Now I think James himself could do better. We've improved on WS by the simple expedient of adjusting for season length and filling in missing seasons for certain circumstances. But. The fact that James' bullshirt dump is out there in the open rather than hidden like the lunatic uncle is a good thing. It is a strength rather than a weakness. How one constructs one's own bullshirt dump of course can be a different strength or a different weakness. But when James calls attention to the fact that there is a bullshirt dump, he is doing us a favor.
And of course the actual math is transparent and the numbers don't change every quarter.
So WS may not be "good enough," but it is "as good as."
Which is one reason why James was at one point working on "NEW" Win Shares, which were to include LOSS Shares. Not sure if he's still working on it though.
If you want an answer, you'll have to read here:
Tango Tiger & Wood Win Shares Critique
and here:
Patriot's Win Shares Walk Thru
Sorry this is so late.
I've got Clarkson with 1589
Career offensive totals for William Aloysius Bergen
Hits: 516
Doubles: 45
Triples: 21
Home runs: 2
Bases on balls: 88
Stolen bases: 23
Runs scored: 138
Runs batted in: 193
Batting Win Shares: 0.0
It's actually 0.1, Dan. :-)
Well, without the 'elimination' of negatives, it's actually negative something or other using the Win Shares calculation.
Obviously there's a "zero point" for batting win shares (and also for pitching and fielding win shares--though one would have to be a pretty awful fielder to get zero fWS). James, however, didn't set the zero points at replacement level. They were set so that the team's win shares would add up to 3 times its wins. James recognizes that if the system were measuring value above replacement level, the zero point would need to be set higher and a team's wins above replacement would add up to fewer than 3 times its wins. He chose not to try to set the zero point at replacement level.
Here's a slightly absurd analogy--would Celsius criticize Fahrenheit for choosing a ridiculously low value for the melting point of ice because the zero point in Fahrenheit's scale is well below the zero point in the Celsius system? Fahrenheit would have simply responded that the zero point in his system was intended for another purpose. (According to Wikipedia, Fahrenheit's zero point may have been determined by the lowest point in a mixture of water, ice, and ammonium chloride.)
As folks over at the Hardball Times have shown, it may be possible to rescale win shares to get "win shares above bench," i.e., a replacement-type concept. (I'm not defending their replacement levels--just mentioning that such a rescaling has been done.)
Every time win shares comes up around here, it seems like we get a data dump of all of the criticism that's ever been written on the subject. I'm not trying to defend win shares, but when Bill James devotes 3 pages to explaining that he didn't design the system as a measure of value above replacement level, it seems to me like an unfair criticism to say "yes he did, and moreover, he blew it in picking the replacement value." Sure, one could interpret win shares as a replacement-value system with an unrealistically low replacement value, but that's not how James designed it and there is no reason to necessarily interpret it that way.
I hope that we don't need to continue this debate, because I think it must fundamentally just be a semantic misunderstanding of some sort. Apparently James (and I) think it's possible to have a system with a zero point that isn't a replacement value and you don't.
You must be Registered and Logged In to post comments.
<< Back to main