Baseball for the Thinking Fan

Login | Register | Feedback

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

Wednesday, January 04, 2023

2023 Hall of Merit Election Results

Congratulations to our 2023 Hall of Merit Inductees. We have elected outfielders Carlos Beltran, Lance Berkman and Bobby Bonds.

player name	pts	ballots	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15
Carlos Beltran	606	26	22	2		1		1									
Lance Berkman	314	19		2	5	1	1	3	2	1	1	1			2		
Bobby Bonds	237	17		2	1		4	1	1		2	1	1	2		1	1
-----------------------------------------------------------------------------------------------------------------------------------
Buddy Bell	221	15		4	1	1		2					3	1	1	1	1
David Ortiz	207	13		2	2	2	1		1	2		1	1				1
Sal Bando	178	13		1	2	1	1			1		2	2	1	1		1
Tim Hudson	170	12			2	3	1	1		1				1	1	2	
Tommy John	158	11			2	2	2	1				1	1				2
Bob Johnson	153	11		1			4	1		1		1	1	1	1		
Thurman Munson	142	9		1	3			1		1			1	1			1
Kevin Appier	140	10		1	1		1	2			2	1			1		1
Jason Giambi	140	9		1		2	1	1	2		2						
Phil Rizzuto	122	10		1				2		1	2			2		2	
---------------------------------------------------------------------------------------------------------------------------------
Ben Taylor	105	9		1		1					2	1	1		1		2
Vic Willis	103	8			1		1		1			2	2	1			
Don Newcombe	85	7				1	1		2			1				1	1
Mark Buehrle	78	5			1	1	1		1					1			
Urban Shocker	78	7							1		2	2	1		1		
Tommy Bridges	77	4	1	1			1		1								
John Olerud	77	7						1	2		1				2		1
Wally Schang	74	5			2					1			1			1	
Roy Oswalt	73	7							2	1			1	1		1	1
Jorge Posada	65	4	1			1				1		1					
Heavy Johnson	59	4		1						1	1	1					
Dwight Gooden	58	5						1		1	1			2			
David Wright	54	6				1									2	3	
Hilton Smith	52	3		1		1					1						
Willie Davis	51	4						1		2			1				
Bert Campaneris	50	5								2				1	1	1	
Bucky Walters	47	5							1	1						2	1
Orel Hershiser	47	5									1	2				1	1
Leon Day	45	3			1				1					1			
Dizzy Dean	44	4	1												1		2
Brian Giles	43	4							1	1					2		
Bus Clarkson	43	3		1							1				1		
Jim McCormick	40	2		1		1											
Luke Easter	40	2	1				1										
George Scales	39	3				1				1				1			
Mickey Welch	39	4							1		1					1	1
Babe Adams	38	3				1		1									1
Hugh Duffy	38	3			1									1		1	
Fred McGriff	36	2		1						1							
Tommy Leach	33	2				1	1										
Robin Ventura	32	3									1	1		1			
Tommy Bond	32	2					2										
Hurley McNair	29	2				1					1						
Dale Murphy	28	2			1												1
Gene Tenace	28	2				1						1					
Tony Perez	26	2								2							
Chuck Finley	25	2						1					1				
Jim Sundberg	25	3											1		1	1	
Bob Elliott	24	2						1						1			
Joe Tinker	24	3											1		1		1
Bernie Williams	24	2							1				1				
Lou Brock	23	1		1													
Luis Aparicio	23	2									1	1					
Tony Phillips	23	2								1			1				
Elston Howard	22	2							1						1		
Kirby Puckett	22	2							1						1		
Al Rosen	18	2												2			
Dolf Luque	17	1				1											
Jim Kaat	16	1					1										
No Garciaparra	16	2											1				1
Vern Stephens	16	1					1										
Jose Cruz	16	2												1		1	
Ch Buffinton	15	1						1									
Lee Smith	15	1						1									
Sam McDowell	15	1						1									
Tony Mullane	15	1						1									
Billy Wagner	14	1							1								
Frank Chance	14	1							1								
Pie Traynor	14	2													1		1
Lance Parrish	14	1							1								
Mickey Lolich	13	1								1							
Jim Whitney	13	1								1							
Charlie Smith	12	1									1						
Da Strawberry	12	1									1						
Ron Guidry	12	1									1						
Johnny Evers	11	1										1					
Dave Bancroft	11	1										1					
Junior Gilliam	11	1										1					
Roy White	11	1										1					
Frank Tanana	11	1										1					
Gavvy Cravath	10	1											1				
Tommy Henrich	10	1											1				
Jack Quinn	10	1											1				
Jim Fregosi	9	1												1			
Toby Harrah	9	1												1			
Luis Gonzalez	9	1												1			
Albert Belle	8	1													1		
Bobby Veach	8	1													1		
Cesar Cedeno	8	1													1		
Norm Cash	7	1														1	
Dave Concepcion	7	1														1	
G Van Haltren	7	1														1	
Ron Cey_	7	1														1	
Wilbur Wood	7	1														1	
Eddie Cicotte	7	1														1	
Chet Lemon	6	1															1
Mark Teixeira	6	1															1
Ned Williamson	6	1															1
DL from MN Posted: January 04, 2023 at 07:37 PM | 46 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.

   1. DL from MN Posted: January 04, 2023 at 07:43 PM (#6111882)
Thanks to everyone who continues to vote and support this project.
   2. Jaack Posted: January 04, 2023 at 08:44 PM (#6111892)
A hearty congratulations to Beltran, Berkman, and Bonds! Good alliteration this year. Also a big thanks to DL for keeping everything running smoothly!

As far as caps go, Berkman is obviously an Astro, and Bonds is a Giant. I would say Beltran is a Met, but I think the Royals have a plausible argument at least.
   3. DL from MN Posted: January 04, 2023 at 09:07 PM (#6111897)
Bonds is basically sitting right at my in/out line. Berkman will never make my PHoM.

David Ortiz looks inevitable.
   4. cookiedabookie Posted: January 04, 2023 at 09:08 PM (#6111898)
I would like to pilot the idea I mentioned about this year's voters only ranking the top 15, to see if results would change based on a more concentrated look at those closest to induction, given the diversity in opinion right now. If people are interested, let me know. Here's how I'd rank the top fifteen from this year:

1 Carlos Beltran
2 Bob Johnson
3 Thurman Munson
4 Tim Hudson
5 Bobby Bonds
6 Lance Berkman
7 Buddy Bell
8 David Ortiz
9 Kevin Appier
10 Tommy John
11 Phil Rizzuto
12 Sal Bando
13 Vic Willis
14 Jason Giambi
15 Ben Taylor

   5. DL from MN Posted: January 04, 2023 at 09:12 PM (#6111899)
For the positional rankings I'm leaning toward LF for Berkman. He has more games at 1B than any other position but he also has more games in the outfield than at first.
   6. kcgard2 Posted: January 04, 2023 at 09:15 PM (#6111900)
Bonds vs Bell was a tight race for the entire balloting period. I support both for HOM, congrats to the elder Bonds who edged it out. Ortiz did take a very big jump forward this year. He and Bell look likely for 2024 inductions. My favorite trend of the past two years has been Appier's trajectory into the top runners up.
   7. kcgard2 Posted: January 04, 2023 at 09:20 PM (#6111901)
I agree with Berkman in LF. I count LF/RF as simple corner OF, but HOM keeps those positions distinct. Berkman is a corner OF, and since more of his outfield time came in LF, that's where I count him.
   8. base ball chick Posted: January 04, 2023 at 09:22 PM (#6111903)
berkman was a really great ballplayer who had unfortunately short career. he was not a real too good OF in spite of his famous catch on tals hill when he played CF (????really???) and was an OK 1B. in mah not so umble opinyin he was just really getting good at baserunning when he hurt his knee in 08 (and didn't get put on the DL which was a massive mistake and cost him a lot of playing years)

glad to see him in the HoM getting some recognition

just curious - how many people vote for HOM ballots? who qualifies?
   9. DL from MN Posted: January 04, 2023 at 09:24 PM (#6111904)
1 Carlos Beltran
2 Bob Johnson
3 Phil Rizzuto
4 Kevin Appier
5 Tommy John
6 Tim Hudson
7 Ben Taylor
8 David Ortiz
9 Bobby Bonds
10 Buddy Bell
11 Lance Berkman
12 Thurman Munson
13 Jason Giambi
14 Sal Bando
15 Vic Willis
   10. DL from MN Posted: January 04, 2023 at 09:24 PM (#6111905)
just curious - how many people vote for HOM ballots? who qualifies?


26 people voted this year. Anyone who wants to join the discussion is welcome.
   11. Jaack Posted: January 04, 2023 at 09:36 PM (#6111908)
Just pulling straight from my rankings I'd have the top 15 as:
1. Carlos Beltran
2. Lance Berkman
3. Tommy John
4. Bob Johnson
5. Bobby Bonds
6. Tim Hudson
7. Buddy Bell
8. David Ortiz
9. Kevin Appier
10. Jason Giambi
11. Thurman Munson
12. Vic Willis
13. Sal Bando
14. Phil Rizzuto
15. Ben Taylor


I also have Berkman as a LF - 2000 more total OF innings, and generally in his best season he was somewhere in the OF.
   12. Chris Cobb Posted: January 04, 2023 at 10:04 PM (#6111911)
My ranking of the Top 15 would be

1. Carlos Beltran
2. Buddy Bell
3. Tim Hudson
4. Jason Giambi
5. Bobby Bonds
6. Kevin Appier
7. Lance Berkman
8. Vic Willis
9. Sal Bando
10. Phil Rizzuto
11. David Ortiz
12. Thurman Munson
13. Bob Johnson
14. Tommy John
15. Ben Taylor

In my own record keeping, I have Berkman as LF/1B, but for a single positional designation I'd go with left field.
   13. cookiedabookie Posted: January 04, 2023 at 10:24 PM (#6111914)
I'm Team LF for Berkman
   14. Michael J. Binkley's anxiety closet Posted: January 05, 2023 at 07:59 AM (#6111930)
My ranking of the top 15:

1. Carlos Beltran
2. Jason Giambi
3. David Ortiz
4. Lance Berkman
5. Kevin Appier
6. Thurman Munson
7. Bobby Bonds
8. Vic Willis
9. Buddy Bell
10. Tim Hudson
11. Tommy John
12. Sal Bando
13. Ben Taylor
14. Bob Johnson
15. Phil Rizzuto

I had Berkman listed as a 1b, but I am more than willing to move him to LF since I currently have fewer LF in my PHoM than any other position.
   15. Howie Menckel Posted: January 05, 2023 at 12:45 PM (#6111990)

1. CARLOS BELTRAN
2. LANCE BERKMAN
3. BOB JOHNSON
4. TIM HUDSON
5. KEVIN APPIER
6. DAVID ORTIZ
7. SAL BANDO
8. BUDDY BELL
9. BEN TAYLOR
10. PHIL RIZZUTO

11. THURMAN MUNSON
12. BOBBY BONDS
13. VIC WILLIS
14. TOMMY JOHN
15. JASON GIAMBI
   16. kcgard2 Posted: January 05, 2023 at 04:41 PM (#6112048)
1. Carlos Beltran
2. Buddy Bell
3. Sal Bando
4. Tommy John
5. Bobby Bonds
6. Lance Berkman
7. Kevin Appier
8. Bob Johnson
9. Jason Giambi
10. Tim Hudson
11. Thurman Munson
12. Vic Willis
13. David Ortiz
14. Ben Taylor
15. Phil Rizzuto
   17. Brent Posted: January 05, 2023 at 04:41 PM (#6112049)
My ranking of the top 15:

1. Carlos Beltrán
2. Bobby Bonds
3. Sal Bando
4. Buddy Bell
5. Kevin Appier
6. Jason Giambi
7. Lance Berkman
8. Phil Rizzuto
9. David Ortiz
10. Tim Hudson
11. Thurman Munson
12. Vic Willis
[gap]
13. Bob Johnson
14. Ben Taylor
15. Tommy John
   18. The Honorable Ardo Posted: January 05, 2023 at 07:26 PM (#6112082)
Hello all, I sincerely apologize for not voting. I'm dealing with financial and personal issues at the moment and didn't have enough time to put together a ballot. Congratulations to the inductees!
   19. The Honorable Ardo Posted: January 05, 2023 at 08:11 PM (#6112086)
It's also time for one of my favorite annual rituals: the One-Vote team!

Rules: 25-man roster, received exactly one Hall of Merit vote.

Catchers: Lance Parrish, Frank Chance (I know... Chance caught 187 games which is > 15% of his career appearances)
Infielders: Mark Teixiera, Norm Cash, Johnny Evers, Dave Bancroft, Ron Cey, Vern Stephens, Junior Gilliam
Outfielders: Albert Belle, Darryl Strawberry, Lou Brock, Cesar Cedeno, George Van Haltren, Chet Lemon
Pitchers: Dolf Luque, Jim Kaat, Tony Mullane, Ron Guidry, Wilbur Wood, Mickey Lolich, Eddie Cicotte, Sam McDowell, Lee Smith, Billy Wagner

Given a 162-game season and these players performing at an "average" level (not their absolute peak, but not their decline phase either), how many games does it win?
   20. DL from MN Posted: January 06, 2023 at 10:24 AM (#6112125)
The Honorable Ardo Posted: January 05, 2023 at 07:26 PM (#6112082)
Hello all, I sincerely apologize for not voting


We moved your prelim to the ballot thread. You voted.
   21. John (You Can Call Me Grandma) Murphy Posted: January 06, 2023 at 10:28 AM (#6112126)
Not bad results, IMO. The top two were on my ballot, while the last one was on my ballot for years (#16 for this election. Congrats to the newest members of the HoM!
   22. bachslunch Posted: January 06, 2023 at 11:54 AM (#6112149)
My ranking of the top 15. Seven were on my ballot, four were in my 16-42 ranking, four were below this (thus the gap):

1 Carlos Beltran
2 Buddy Bell
3 Bob Johnson
4 Bobby Bonds
5 Vic Willis
6 Sal Bando
7 Tommy John
8 David Ortiz
9 Thurman Munson
10 Lance Berkman
11 Tim Hudson
[gap]
12 Jason Giambi
13 Kevin Appier
14 Phil Rizzuto
15 Ben Taylor
   23. cookiedabookie Posted: January 06, 2023 at 12:23 PM (#6112154)
Thanks for everyone who has shared their re-rank of the top 15. We've got a bit more than a third of the voters who have responded, the top of the ballot is slightly different, but very close at the top
   24. kcgard2 Posted: January 06, 2023 at 03:29 PM (#6112196)
2023 ballot consensus scores (lower number = greater consensus):

cookiedabookie 11.64
Rob Wood 12.91
Mike Webber 13.66
Al Peterson 13.91
Howie Menckel 14.35
Chris Cobb 14.70
kcgard2 14.78
Devin (fuzzy socks) 14.86
Bleed the Freak 15.23
Esteban Rivera 16.05
Jaack 16.24
Brent 16.36
rwargo 16.52
James Newburg 16.53
The Honorable Ardo 17.23
bachslunch 17.55
theorioleway 18.41
DL from MN 18.55
Dr. Chaleeko 19.11
Kiko Sakata 19.19
Mark A Shirk 20.50
progrockfan 20.62
Michael J. Binkley 20.83
Patrick W 21.10
John Murphy 21.58
bjhanke 21.85

   25. cookiedabookie Posted: January 06, 2023 at 03:31 PM (#6112197)
Nice :)
   26. theorioleway Posted: January 06, 2023 at 04:29 PM (#6112207)
Beltran
Appier
Hudson
Willis
Taylor
Bell
Berkman
Rizzuto
Ortiz
Bonds
Munson
Giambi
Johnson
John
Bando
   27. kcgard2 Posted: January 06, 2023 at 05:02 PM (#6112213)
This was the highest finish historically for Bell, Appier, Hudson, Giambi, Ortiz, Bando, John, Buehrle, Olerud, Oswalt, Bridges, Gooden, Hershiser, Finley, Ventura, Willie Davis, Scales, and Day.
This was the lowest finish historically for Leach, Posada, Stephens, Cicotte, González, and Teixeira.
These were the first votes ever for Sam McDowell (really?!), Wilbur Wood, and Chet Lemon.
   28. kcgard2 Posted: January 06, 2023 at 05:10 PM (#6112217)
Bonds' lowest finish in an election was 38th. Only 8 players in HOM have ever finished lower, 4 of them non-NgL players.
   29. The Honorable Ardo Posted: January 06, 2023 at 07:03 PM (#6112244)
Very kind of you all. Thank you, DL and the entire community.

Quick ranking of top 15:

1. Beltran
2. Bell
3. Ortiz
4. Taylor
5. Munson
6. Berkman
7. Rizzuto
8. Bonds
9. Bando
10. John
11. Hudson
12. Willis
13. Appier
14. Giambi
15. Johnson

These aren't sharp distinctions. I have a bigger gap between #1 Beltran and #2 Bell than between Bell and Bob Johnson.
   30. Bleed the Freak Posted: January 06, 2023 at 08:58 PM (#6112259)
1 Carlos Beltran

2 Lance Berkman

3 Thurman Munson

4 David Ortiz

5 Tommy John

6 Tim Hudson

7 Jason Giambi

8 Sal Bando

9 Bobby Bonds

10 Kevin Appier

11 Bob Johnson

12 Vic Willis

13 Buddy Bell

14 Ben Taylor

15 Phil Rizzuto
   31. kcgard2 Posted: January 06, 2023 at 09:38 PM (#6112265)
The plaque room and HOM site have been updated.
   32. bjhanke Posted: January 07, 2023 at 02:29 AM (#6112283)
Huh. I may have actually had an effect on this election. I made the decisions to get my ballot done by force of will (I've had medical problems), because I noted that Bobby Bonds was near to being elected, and I wanted to vote for him. And he got in. And I think my vote was worth about 16 points. Which is the difference between Bonds and Buddy Bell. Huh. See,it does pay to vote. Huh.
   33. John (You Can Call Me Grandma) Murphy Posted: January 07, 2023 at 07:48 AM (#6112288)
2023 ballot consensus scores (lower number = greater consensus):

cookiedabookie 11.64
Rob Wood 12.91
Mike Webber 13.66
Al Peterson 13.91
Howie Menckel 14.35
Chris Cobb 14.70
kcgard2 14.78
Devin (fuzzy socks) 14.86
Bleed the Freak 15.23
Esteban Rivera 16.05
Jaack 16.24
Brent 16.36
rwargo 16.52
James Newburg 16.53
The Honorable Ardo 17.23
bachslunch 17.55
theorioleway 18.41
DL from MN 18.55
Dr. Chaleeko 19.11
Kiko Sakata 19.19
Mark A Shirk 20.50
progrockfan 20.62
Michael J. Binkley 20.83
Patrick W 21.10
John Murphy 21.58
bjhanke 21.85


I'm never going to have a high consensus number anymore, since I'm not a WAR guy and most here are wedded to it. I lean more towards Win Shares, but none of these systems are good enough for me to use without doing some analysis on my own.
   34. John (You Can Call Me Grandma) Murphy Posted: January 07, 2023 at 08:10 AM (#6112289)
My top 15:

1) Carlos Beltran

2) Lance Berkman

3) Jason Giambi

4) Vic Willis

5) Bobby Bonds

6) Bob Johnson

7) Kevin Appier

8) Sal Bando

9) Tim Hudson

10) Buddy Bell

11) Thurman Munson

12) Phil Rizzuto

13) Ben Taylor

14) Tommy John

15) David Ortiz
   35. Rob_Wood Posted: January 07, 2023 at 06:51 PM (#6112365)
My top 15:

1. Carlos Beltran
2. Buddy Bell
3. Tim Hudson
4. Bob Johnson
5. Kevin Appier
6. Bobby Bonds
7. Sal Bando
8. Vic Willis
9. Phil Rizzuto
10. Lance Berkman
11. Thurman Munson
12. David Ortiz
13. Ben Taylor
14. Jason Giambi
15. Tommy John
   36. cookiedabookie Posted: January 08, 2023 at 01:55 PM (#6112436)
More than half the voters have re-ranked the top 15, and Beltran sails in unanimously so far. Only one of the other two we elected is in the top three right now. Love to get the rest, although I may have to try and sleuth out from those voters posts
   37. bjhanke Posted: January 09, 2023 at 04:17 AM (#6112512)
RE: Consensus scores. I'm essentially a New Historical Player Ranking system guy, which is based on Win Shares. So, it's no surprise that it's you and me, Grandma. On the other hand, if any of the newer raw systems (like WAR or Win Shares) had as good a derivative system as the New Historical, maybe I'd change my tune. But the other derivative systems are SO bad thatt there's no chance of my taking them seriously. I mean, JAWS? Really? Compared to Career / 3-yearPeak / 5-year Prime / Career Rate per 162 Games / Extras (postseason, etc)? There is literally no derivative system that I know of that even deserves any respect.
   38. Chris Cobb Posted: January 09, 2023 at 09:04 AM (#6112521)
On the other hand, if any of the newer raw systems (like WAR or Win Shares) had as good a derivative system as the New Historical, maybe I'd change my tune. But the other derivative systems are SO bad thatt there's no chance of my taking them seriously. I mean, JAWS? Really? Compared to Career / 3-yearPeak / 5-year Prime / Career Rate per 162 Games / Extras (postseason, etc)? There is literally no derivative system that I know of that even deserves any respect.

It doesn't seem to me that there is any necessary relationship between a comprehensive metric and an approach to evaluating merit for the purpose of determining player excellence. Bill James presents both a comprehensive metric and an approach to evaluating players in the NBJHBA. The evaluative approach requires a comprehensive metric to work, but it doesn't depend on Win Shares: any comprehensive metric could be used as the base source of information for the Bill James ranking method. (Indeed, I would say that, historically, the components of the James ranking method pre-date the Win Shares metric, rather than being derivative of it. James just shifted the inputs to his ranking system and determined a way to scale them once he had invented a comprehensive metric that he could use.) The timeline component would have to be re-scaled by dividing by 3 for a system denominated in wins instead of win shares, but that's about it. The comprehensive metric determines the number of wins a player should be credited for their career on a seasonal basis: the ranking system (as interpreted by the judgment of the person applying it) determines how much merit those achievements should be awarded. In the case of the NBJHBA, the same person has created the metric and the evaluative system, but there's no reason the two need to be tied together. You could start feeding wins-above-replacement data into the Bill James evaluative system tomorrow if you wanted to, or Jay Jaffe could start feeding Win Shares instead of Baseball Reference WAR into his JAWS system tomorrow, if either of you thought that using a different comprehensive metric would yield better results.

One way to look at what most Hall of Merit voters are doing is trying to figure out both sides of the evaluative problem: what comprehensive metric (or mix of metrics) provides the most accurate base information about player performance, and how should that that information be processed by an evaluative system in order to establish merit meaningfully? Some voters find it preferable to use a single comprehensive metric; some voters find it preferable to combine them; a few voters have the skill, intrepidity, and determination to develop their own. Voters then develop an evaluative system that processes the base information about player performance. Bill James' emphasis on the distinction between peak and career value, on the need to combine them, and on the challenge of integrating players from different periods into a single ranking have been of foundational importance to the whole HoM project, even though most voters (and for period integration, the HoM's constitutional design) have not adopted James' specific ways of dealing with these issues. My system, for example, uses peak/prime/career inputs that were derived from James's system, but it modifies those inputs to shift the balance between peak and career a bit toward career and to eliminate the career rate measure. Originally, my system used Win Shares as its data source, then transitioned to a combination of Win Shares and Clay Davenport's WARP, then dropped WARP and combined Win Shares and Baseball Reference's WAR, then dropped Win Shares and added Fangraphs WAR for position players, then added DRA for fielding, then dropped it back out, as I've tried to identify and use the best comprehensive metrics as they have evolved. But the basic system, peak ratex5 + seasonal wins above average + career WAR has remained the same. My approach to integrating periods has evolved along different lines, not directly inspired by any prior published method, and it is still evolving.

In a sense, then, each voter at the Hall of Merit is operating their own "derivative system," and I think that many of them are, taken just on their own, better than what James did in the NBJHBA or what Jaffe is doing with JAWS. Put together into the Hall of Merit, the aggregate results are much better. So in another sense, we are all participating in a "derivative system," an evaluative system that uses the comprehensive metrics and other data sources as inputs for a collective system for evaluating merit. Because we vote together, the Hall of Merit presents itself as an alternative to the Hall of Fame, but the HoM is equally a continually updated alternative to some parts of the NBJHBA. It's surely not a perfect macro-system, but it works pretty well, and it's pretty resilient, too, which is a credit both to the system and to the participants.



   39. DL from MN Posted: January 09, 2023 at 01:41 PM (#6112553)
Creating a derivative system based on an uber-stat is just simple addition. The hard part is finding an uber-stat that is consistent across baseball history.
   40. Jaack Posted: January 09, 2023 at 02:26 PM (#6112566)
I think any derivative system is prone to some biases somewhere or another, regardless of the inputs, or if it's a fairly comprehensive one like James or something simplistic like JAWS.

For example, James defines a peak as 3 years. JAWS has it at 7 years. Both choices seem valid enough, while also being quite arbitrary. Realistically, when designing any sort of derivative system, you will make arbitrary choices. The HoM aggregate voting is going to smooth out those arbitrary choices very well - we will identify great peak candidates better than any singular derivative system ever could.
   41. Chris Cobb Posted: January 09, 2023 at 02:42 PM (#6112567)
Creating a derivative system based on an uber-stat is just simple addition.

Mathematically, yes, for the most part, but making sound judgments about how much to add in of what in order to get results that provide a satisfying balance between different aspects of value is just as important to a system and isn't as straightforward as addition.

The hard part is finding an uber-stat that is consistent across baseball history.

Amen, although I'd qualify that a little. My sense is that the comprehensive metrics are all designed to establish a player's value accurately in the context of each baseball season. To succeed in that goal, they can't be consistent across baseball history, because of changes in the dynamics of play and of the level of competition involved. Therefore, we have to find ways to work with the comprehensive metrics to achieve consistency in our evaluations. Would your aim be to have a comprehensive metric for which we could say "a win is a win is a win" for any season and league within that season from 1871 to 2022? I'd like to know more about your sense of what consistency across baseball history entails.
   42. Mike Webber Posted: January 09, 2023 at 07:09 PM (#6112606)
1 Carlos Beltran
2 Sal Bando
3 David Ortiz
4 Lance Berkman
5 Jason Giambi
6 Phil Rizzuto
7 Bobby Bonds
8 Buddy Bell
9 Tim Hudson
10 Bob Johnson
11 Kevin Appier
12 Vic Willis
13 Ben Taylor
14 Tommy John
15 Thurman Munson
   43. Eric J can SABER all he wants to Posted: January 09, 2023 at 07:44 PM (#6112611)
For example, James defines a peak as 3 years. JAWS has it at 7 years. Both choices seem valid enough, while also being quite arbitrary. Realistically, when designing any sort of derivative system, you will make arbitrary choices. The HoM aggregate voting is going to smooth out those arbitrary choices very well - we will identify great peak candidates better than any singular derivative system ever could.

Any arbitrary choice of this type has the additional effect of inherently biasing a system in favor of certain players. For example, Dizzy Dean has been brought up in this context before; James ranked him as the #25 pitcher ever in the NBJHBA, and it's largely because his system is perfect for Dean. 3-year peak? Dean's 3 best seasons are excellent. 5-year consecutive prime? The bulk of Dean's value was accrued in a consecutive 5.5-year span. Value per game (phrased as value per season, but it's done on a per-162 basis, or per 43 starts for pitchers)? Dean's injury kept him off the mound entirely, rather than allowing him to pitch at a reduced effectiveness, so his rate stats are great. If the system was best 5 seasons and best 8 consecutive years, or if the rate measure penalized injury seasons, Dean would fare far worse.

JAWS has the same issue for players with exactly 7 good years - say Troy Tulowitzki, who has six seasons of 5-7 WAR plus a seventh at 3.2 and nothing else over 2; Tulo would fare comparatively worse by either top 4 seasons or top 10 seasons. If you compare Tulo's top 7 seasons by bWAR to David Wright's, he leads 40.1 to 38.3. Wright would lead in either top 4 (27.5-26.2) or top 10 (44.7-43.6).

For what it's worth, my preferred method to avert this issue is to take an average of "best N seasons" for multiple values of N, which ends up being a weighted sum of WAR values (or pick your own preferred metric) by rank within the player's career.
   44. Jaack Posted: January 09, 2023 at 08:44 PM (#6112620)
Any arbitrary choice of this type has the additional effect of inherently biasing a system in favor of certain players. For example, Dizzy Dean has been brought up in this context before; James ranked him as the #25 pitcher ever in the NBJHBA, and it's largely because his system is perfect for Dean. 3-year peak? Dean's 3 best seasons are excellent. 5-year consecutive prime? The bulk of Dean's value was accrued in a consecutive 5.5-year span. Value per game (phrased as value per season, but it's done on a per-162 basis, or per 43 starts for pitchers)? Dean's injury kept him off the mound entirely, rather than allowing him to pitch at a reduced effectiveness, so his rate stats are great. If the system was best 5 seasons and best 8 consecutive years, or if the rate measure penalized injury seasons, Dean would fare far worse.

JAWS has the same issue for players with exactly 7 good years - say Troy Tulowitzki, who has six seasons of 5-7 WAR plus a seventh at 3.2 and nothing else over 2; Tulo would fare comparatively worse by either top 4 seasons or top 10 seasons. If you compare Tulo's top 7 seasons by bWAR to David Wright's, he leads 40.1 to 38.3. Wright would lead in either top 4 (27.5-26.2) or top 10 (44.7-43.6).

For what it's worth, my preferred method to avert this issue is to take an average of "best N seasons" for multiple values of N, which ends up being a weighted sum of WAR values (or pick your own preferred metric) by rank within the player's career.


What you've described is pretty functionally similar to how I rate players - everyone's best season gets one multiplier, everyone's second best season gets a slightly smaller multiplier, and so on. This avoids some of the arbitrariness of JAWS type of peak ranking, but has it's own flaws - the multipliers are still arbitrary and there does feel something strange about crediting two players differently for the same season. And you inevitibly end up with biases - this method likes guys like Larry Jackson and Steve Rogers who aren't particularly great peakwise or careerwise.
   45. Eric J can SABER all he wants to Posted: January 09, 2023 at 10:45 PM (#6112631)
And you inevitibly end up with biases - this method likes guys like Larry Jackson and Steve Rogers who aren't particularly great peakwise or careerwise.

Sure, there's no perfect system. (But for what it's worth, if I were a voter, Rogers would have a good chance of ending up on my ballot for reasons that aren't WAR-specific. I may go into more detail in the HoM threads if I can get far enough on my pitcher evaluations.)
   46. Mark A Shirk Posted: January 11, 2023 at 10:53 AM (#6112809)
I am finding this to be a bit of weird discussion. All models have assumptions. All rankings systems have assumptions. There is not way to build one that does not have assumptions. That means you can't have one that is just 'reality'. Sooner or later you have to make theoretical choices about what you value.

I find that I agree with most of the assumptions of WAR so I use it as my base (mix of fangraphs and bbref for pitchers, fangraphs for outfield players). I make adjustments for things like being a catcher, whether I think historical fielding metrics are off, or just quirks in my set of assumptions. But there is no way to 'solve' this problem by rejecting WAR or JAWS (different things as Chris says). It never goes away.

BTW, for peak I do not like to use N years (whatever N is). Instead I use value accumulated over 5 WAR (previously 25 WS). Makes it less dependent on choosing the correct N but can favor players with one amazing season (i.e. Rico Petrocelli) too much. Therefore, I make a subjective adjustment for my ballot/rankings.

You must be Registered and Logged In to post comments.

 

 

<< Back to main

BBTF Partner

Dynasty League Baseball

Support BBTF

donate

Thanks to
cHiEf iMpaCt oFfiCEr JE
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Syndicate

Page rendered in 0.1201 seconds
13 querie(s) executed