In the process of adding players I found a glitch in how I entered the hitting value for post-2005 seasons. This escalated to an overhaul in my whole spreadsheet regarding pitcher hitting. I also spent some time looking over marginal NGL pitchers since my PHoM criteria is more pitcher heavy than the rest of the group.

You will see some particular areas where I am going to differ from group consensus - I think pitchers are quite under-represented in the Hall of Merit. I give war credit and also credit for minor league seasons played at a level that indicates the player would have been above average in the major leagues. I think the consensus right now is to NOT give war credit which is a shame. I use Dan R's standard deviation adjusted WAR (when available) because he convinced me that runs/win is not a constant 10 across all of baseball history. I like the positional average and positional replacement value calculations in his spreadsheet as well. I also believe a season is a season whether 60 games or 162 and adjust shortened seasons to a 162 game baseline. I think we have enough pre-1890 players and generally identified the correct ones.

I look at two main things - value above replacement and value above average. I'm not as interested in a (theoretical) 65 WAR player with 0 WAA as I am with a 55 WAR 25 WAA player. Therefore I zero out seasons at the beginning and end of a player's career where WAA + WAR < 0. I don't care about peak - consecutive or non-consecutive - and it doesn't really enter into my calculations.

I haven't ever boycotted due to steroid usage. I have penalized players who gambled by applying the lifetime ban for the offense to immediately after the time when it occurred.

2022 Prelim

1) Alex Rodriguez - This is a pretty easy choice. Similar career value to Joe Morgan and Pop Lloyd but behind Lajoie, Collins, Hornsby and far behind Honus Wagner. PHoM 2022

2) Tommy Bridges - I give war credit for two seasons at the level he was pitching 1941-1945. That gives him roughly 60 WAR and 30 WAA. PHoM 1958.

3) Bob Johnson - Giving him 1.5 seasons of credit for minor league performance as he was performing at a level above MLB average. I've been voting for Bob Johnson every single year I have participated in
the project. PHoM 1986.

4) Urban Shocker - giving him 1/2 season credit for WWI. Good bat for a pitcher helped him in my recalculations. PHoM 1968.

5) Phil Rizzuto - gets 3 full seasons of WWII credit at the level of his average performance from the surrounding 9 seasons. That doesn't give him any extra credit for playing 1946 with malaria. Fantastic fielder with just enough bat. PHoM 1967

6) Mark Buehrle - moves down just slightly but my biases explained above are going to like a guy like this. FIP WAR is going to miss his plus defensive contributions. PHoM 2021.

7) Bert Campaneris - If you look at wins above positional average instead of BBREF wins above average you will like 1970s SS more than 1970s 3B. Good fielder and great baserunner - his bat is basically average but his baserunning makes his offensive contribution a positive. PHoM 1991.

8) Dave Bancroft - another mostly glove SS with extra credit (50% for 1919). PHoM 1976

9) Bucky Walters - overlooked pitcher from WWII era. Very good hitter for a pitcher which helped him in my re-evaluation. PHoM 1972

10) Ben Taylor - moves down after latest MLE adjustments. He's Rafael Palmeiro of the deadball era. Also compares well to Keith Hernandez. Better than Mule Suttles. The last obvious Negro League candidate. PHoM 1973.

11) Lazaro Salazar - this is just a prelim but I noticed that I wasn't paying as much attention to the "marginal" Negro League pitchers as I was to the batters. Well, at this point in the project we're selecting from among the margins. I like pitchers better than most, especially if they can hit. Salazar was often called "Dihigo-lite" but Dihigo is in my top 75 while my PHoM is looking for someone around 300th best. All-around player who is in a lot of Latin baseball halls of fame. Contributions as a manager were counted there but are ignored here. This placement is subject to change with discussion.

12) Don Newcombe - gets 2-1/2 years of war credit for serving 1952-54. Gets 1/2 year of minor league credit for being held back due to integration quotas. Good bat. PHoM 2004.

13) Brian Giles - Perhaps the modern day version of Bob Johnson - he was an average contributor right away and provided a ton of value from 2000-2005. Zero out his last season and he's 30WAA. Has an argument for minor league credit but I'm not giving it to him on this placement. PHoM 2020.

14) Norm Cash - haven't had him appear on my ballot in a while now. Terrific fielding 1B with a plus bat and one monster season. PHoM 1997

15) Bus Clarkson - Needs credit for NGL, Mexican League and time missed due to the war as well as minor league credit for integration quotas. He's likely an average fielding 3B or below average at SS but the bat is a plus at either position. PHoM 1967

16) Kevin Appier - PHoM 2009
17) Roy Oswalt - PHoM 2019
18) Bobby Abreu - PHoM 2020
19) Johnny Pesky - 3 seasons WWII credit. PHoM 2005
20) Gavy Cravath - 4 seasons minor league credit. PHoM 1927
21) Tim Hudson - PHoM 2021
22) Tommy John - PHoM 1995
23) Andy Pettitte - PHoM 2021
24) Jorge Posada - in consideration for PHoM this season
25) Wally Schang - PHoM 1987
26) Sammy Sosa - also in consideration for PHoM this season
27) Dave Concepcion
Kenny Lofton
28) Tommy Leach - PHoM 2009
29) Newt Allen
30) Frank Tanana

After looking over the new MLEs for Dobie Moore he's going to make my PHoM in 2022 in the second slot. Better glove than I had been giving credit for before.

65) Bobby Bonds
72) Buddy Bell - not much value above 3B positional average
77) Lance Berkman - not quite as good as Chuck Klein
112) Vic Willis - adjusted for standard deviations he's 46 PWAR and 21 WAA. Terrible hitter.
136) Sal Bando

A-Rod is by far the top candidate on the ballot next year. Here's my top 25 right now:

1. Alex Rodriguez, SS, PHOM 2022
2. Andy Pettitte, SP, PHOM 2020
3. Thurman Munson, C, PHOM 1985
4. Wally Schang, C, PHOM 1937
5. Joe Tinker, SS, PHOM 1925
6. Tim Hudson, SP, PHOM 2021
7. Buddy Bell, 3B, PHOM 1995
8. Tommy John, SP, PHOM 2005
9. Bobby Bonds, RF, PHOM 1987
10. Jorge Posada, C, PHOM 2020
11. Urban Shocker, SP, PHOM 1939
12. Dwight Gooden, SP, PHOM 2006
13. Bob Johnson, LF, PHOM 1960
14. Eddie Cicotte, SP, PHOM 1929
15. Sammy Sosa, RF, PHOM 2022
16. Vic Willis, SP, PHOM 1921
17. Lance Berkman, LF, PHOM 2021
18. Ron Cey, 3B, PHOM 1997
19. Willie Davis, CF, PHOM 1985
20. David Wells, SP, PHOM 2022
21. Sal Bando, 3B, PHOM 1987
22. Mark Buehrle, SP
23. Art Fletcher, SS, PHOM 1928
24. Norm Cash, 1B, PHOM 1980
25. Bobby Abreu, RF

David Ortiz is 37th right now. David Wright is 42nd.

DL, thanks for starting this thread!

Here's what I've got for a ballot as of now.

=======

Disclosures: am basing thinking on Negro Leaguers on the 2020 and 2021 ballot discussion threads with a bit of Seamheads info. Otherwise, strong preference for BBRef WAR with some influence of OPS+ and ERA+ for the rest. Am valuing hitting prowess at C, SS, 2B, CF a bit extra, especially the first of these (have moved up catchers a notch). Being best available candidate at your position helps also. Still trying to sort out peak vs. longevity, but usually favor the latter. Fine with giving Negro League credit, not presently giving credit or debit for war, injury, illness, postseason play, or minor league service. Not systematically adjusting for season length, but am giving minimal non-systematic extra emphasis for pre-1961 players. While it's not a popular approach, am currently treating 19th century pitchers pretty much equally as post-1900 if they played most or all of their career in the NL; for now am assuming AA, NA, and UA stats are suspect. Not taken with giving relievers a lot of emphasis or special treatment. Will dock 1st year candidates who bet on games, threw games, impeded players of color, were caught using PEDs post-2005 (Manny, ARod), and likely used pre-2005 if it looks like they'll get an immediate free pass by BBWAA HoF voters (IRod, Ortiz, Pettitte).

1. Wally Schang. Best C WAR available, also hit well.
2. Buddy Bell. Best WAR at 3B. Have decided to trust the metric for him.
3. Jim McCormick. Best WAR for starters not in by a mile, even when removing all his UA-earned credit. Short career, but played in NL except for one UA season.
4. Bobby Abreu. Best WAR among available RFs, better than Sosa.
5. Bob Johnson. Best WAR among available LFs.
6. Ben Taylor. Still appears to be the best NGL position player. And if he's at all equivalent to Eddie Murray, that's good enough for me. Likely will be the final NGL player I support, barring new information. Am also seeing him as the best 1B available.
7. Vic Willis. Good pitcher WAR, best after McCormick.
8. Willie Davis. Now the best CF available, and has more war at a premium OF position than Sosa. I've overlooked him badly.
9. Sammy Sosa. Better WAR than I remembered. Happy to give him some benefit of the doubt given his treatment by the BBWAA.
10. Thurman Munson. For me, he's the second-best available catcher. Put up a lot of value despite short career.
11. Tommy John. Pretty much all compiling, minimal peak. But racked up plenty of WAR this way. I'm on board.
12. Vern Stephens. I value hitting at a premium position highly, so I'm ranking him here as the best SS.
13. Mickey Welch. Next best pitcher WAR after John by a decent margin. He and McCormick are the only 19th century pitchers I currently support.
14. Urban Shocker. Pettitte and Buehrle have slightly more WAR, but Shocker's higher ERA+ pushes him ahead into this position.
15. Sal Bando. Second best WAR at an under-represented position.

16-40. Tony Lazzeri, Joe Tinker, Bobby Bonds, John Olerud, Jim Fregosi, Andy Pettitte, Mark Buehrle, Tommy Bridges, Jose Cruz, Jorge Posada, Tim Hudson, Luis Aparicio, Bert Campaneris, Robin Ventura, Chet Lemon, Johnny Evers, Jack Quinn, Brian Downing, Tony Perez, Sam Rice, Lance Berkman, Gavvy Cravath, Gene Tenace, Vada Pinson, Fred McGriff.

1B. Taylor, Olerud, Perez, McGriff, Cash, Giambi
2B. Lazzeri, Evers, Pratt, L. Doyle, Myer, Phillips
SS. Stephens, Tinker, Fregosi, Aparicio, Campaneris, Bancroft
3B. Bell, Bando, Ventura, Cey, Harrah, Elliott
LF. B. Johnson, J. Cruz, Downing, Berkman, J. Gonzalez, Veach
CF. W. Davis, Lemon, Pinson, Damon, Cedeno, Puckett
RF. Abreu, Sosa, Bonds, Cravath, S. Rice, Hooper
C. Schang, Munson, Posada, Tenace, D. Porter, Kendall
P. McCormick, Willis, John, M. Welch, Shocker, Pettitte, Buehrle, Bridges, Hudson, Quinn, Finley, Tanana, Hershiser, Cicotte, Uhle.

I've decided to give catchers a little more of a boost, which moves Schang and Munson up a bit. I have Lazzeri as the best available 2B (and that would automatically have given him a ballot position), but I feel more strongly about Davis and Shocker being on the ballot. Lazzeri will make it next time around.

All required disclosure players are on ballot or within top 40. None of the newcomers make my ballot, as I'm boycotting Alex Rodriguez (who would otherwise have been ranked #1) and Ortiz (who would have ranked 25th, between Cruz and Posada).

As of now, this would be my posting-ready ballot. Please bring it over if I haven't done so by the deadline next year for some reason.

Hey Cookie, are you a Yankee fan?
Also, can we get a some kind of permanent sticky note that reminds me how to clear my cache or whatever so I can actually post without rebooting my computer which has a 50/50 success rate?

Hey Cookie, are you a Yankee fan?

9 of his top 12 - though a few aren't as obvious as the others

:)

Noting that two of the preliminary ballots so far posted include more pre-1950 players than post-1950 players, I wanted to mention some data about representation again with respect to fairness to all eras.

From 1871 through 1950, there were around 1280 team seasons of major-league baseball and its equivalents (counting the Negro Leagues as 4 additional major-league teams per year from 1920 to 1950).

From 1951 through 2016 (when the careers of all eligible players had concluded), there were around 1600 team seasons.

So far, we have elected around 150 players who played mainly during those seasons. and we have elected around 125 players who played mainly during those seasons. That's 25% more team seasons since 1950, but 20% fewer electees.

If we were to achieve parity of representation between players from 1950-2010 with players from before 1950, the next 35 or so electees would need to come from this period: that's a full decade of elections, during which time another decade's worth of players will have become eligible. It appears, then, that to fairly represent the players of later decades, we should mainly be electing post-1950 players, and the large majority of players deserving of election should be found in these decades (and leaning much more toward the last 20 years, in fact).

If your system is not seeing a ratio of electable players along these lines, then your system may not be adjusting sufficiently for the strengthening competitive context, which makes it harder for the above average players to dominate the competition. This lesser dominance will result in lower peaks and shorter careers for top players.

bachlunch's preliminary ballot presents an interesting case. Overall, the top 40 leans toward post-1950, 25 out of 40 players. (My top 40 in 2021 leans only a little bit more toward the post-1950 cohort at 29-11, post-1950 to pre-1950.) The ballot, however, leans pre-1950, with 8 of 15 spots going to pre-1950 players. That, to me, suggests an analysis that there are a lot more "very good" players now, but they don't stand out as much as the earlier players. That's exactly the effect an improved competitive context would produce.

It has been noted that very few players have been elected from the "backlog" rather than the "frontlog" in the last eight years of balloting. I take that as an indication that we are, in general, responding appropriately to the historical development of the game and its demographic implications. But I also get a sense that there is a feeling that we've now cleared out "the frontlog," so that now the older backlog should start rising to the top again. Given the disparity between pre- and post-1950 electees, I don't think that is the case. I am not saying that we shouldn't elect anybody from before 1950, and I am not suggesting strategic voting, but rather systematic adjustment to reflect changes in competitive context. The HoM electorate would have to have erred more significantly than I think we have for there to be any unelected pre-1950 player for whom there are not 5 or so equally or more meritorious modern players in need of election. I'd say we would keep on track to fair treatment of all eras if we were to elect 0-5 pre-1950 players over the next decade and 30-35 post-1950 players.

That's the "perspective check" that I think a look at demographics can give us what our systems of analysis are enabling us to see about the merits of players from all eras.

Chris Cobb (#6), thanks for bringing this up, I agree with you. I hope voters will consider it seriously, even if not systematically accounting for it in their "systems" then at least through minor ad hoc adjustment of their rankings. Even as simple as deciding if a modern and a historic player are quite close to each other (there should be many of both that fall in this category), the ranking edge goes to the modern player. Only 4 of the top ten returnees are from the last 20 years, showing a possible inclination among the electorate to further unbalance the eras in the short term, even.

I'm already unsatisfied with the changes and tweaked my pitchers again. Part of the reason is I noticed I had been underrepresenting modern pitchers because they mostly didn't get a chance to hit. It dawned on me that I should be reconciling with my MMP methods. Here's what that looks like.

1) Alex Rodriguez
2) Tommy Bridges
3) Mark Buehrle
4) Bob Johnson
5) Phil Rizzuto
6) Urban Shocker
7) Bert Campaneris
8) Kevin Appier (up from 16)
9) Dave Bancroft
10) Tommy John (up from 22)
11) Andy Pettitte (was 23)
12) Ben Taylor
13) Roy Oswalt
14) Brian Giles
15) Bucky Walters

16-20) Norm Cash, Tim Hudson, Bus Clarkson, Frank Tanana (!!), Bobby Abreu, Johnny Pesky
21-25) Johnny Pesky, Gavy Cravath, Jorge Posada, Wally Schang, Sammy Sosa
26-30) Lazaro Salazar, Don Newcombe, Dave Concepcion, Tommy Leach, Newt Allen

The same people end up on top but they're in a different order mainly because they're all really close together to begin with. Appier, John, Pettitte and Oswalt are back on the ballot.

The big winner here is Frank Tanana, mainly because I was overcompensating for pitchers who didn't get a chance to hit when comparing them to pitchers who did hit. I really shouldn't be surprised a pitcher who was a contributor from 1973-1990 does well in a system that likes Tommy John and Andy Pettitte. Chuck Finley also moved up but not close to my ballot.

David Ortiz ends up 40th on my list which is low enough that he's unlikely to ever end up PHoM.

Here's my CASPER*** (Career Pennant Added using a Sliding Replacement Scale, regular season only) values for the top newbies:

1.76 Alex Rodriguez*
 .74 David Wright
 .71 Mark Teixeira
 .64 David Ortiz**
 .55 Jimmy Rollins
 .54 Jake Peavy
 .51 Carl Crawford
 .41 Joe Nathan
 .36 Justin Morneau
 .35 Jonathan Papelbon

*   On first glance ARod appears to be the only newbie with more than a decent chance of making my ballot
**  By way of comparison, Edgar Martinez has a 1.00 CASPER
*** I use CASPER as a guide in constructing my ballot.

Question. I was vigorously taken to task a couple years ago at this forum and explicitly told to make adjustments in favor of pre-expansion players because their seasons were shorter. I was further informed that this is REQUIRED and non-negotiable. I decided to compromise on the issue by:

"Not systematically adjusting for season length, but...giving minimal non-systematic extra emphasis for pre-1961 players."

I have no intention of going through a huge swath of players to be numerically systematic on this issue. Life's too darned short, and I want to keep this simple.

Posts 5 and 6 above would seem to fly in the face of what I was led to believe is a REQUIREMENT. If someone deliberately misled me on this issue, I would like to know this. Note that I have also had people say that I must adjust for such things as war, illness, postseason play, and minor league service. That's not true, and I don't do so. The only adjustment I do is for Negro League play as a counterbalance against prejudicial treatment, and I feel better about doing so because they actually played baseball and there are numbers one can work with. I do not feel comfortable assuming a player who took time off for military service necessarily would have the same WAR that their surrounding seasons had. I also don't believe in penalizing great players on bad teams by giving postseason credit.

Please note well that I do not subscribe to the idea that our lists must necessarily be period balanced. Five of the top six 2B on my list are pre-1950. Five of my top six 3B on my list are post-1950. Them's the breaks as far as I'm concerned, based on the approach I use which is:

"BBRef WAR with some influence of OPS+ and ERA+ for the rest. Am valuing hitting prowess at C, SS, 2B, CF a bit extra, especially the first of these (have moved up catchers a notch). Being best available candidate at your position helps also."

I also non-systematically adjust slightly in favor of players like Tony Perez (who also played some 3B) and Toby Harrah (who played a bit of SS) and Brian Downing (who also played some catcher), all tougher than their respective 1B, 3B, and LF.

I might add that my ballots seem to have been pretty mainstream, given what was posted on the results thread. Everybody has their odd players that seem to fit their system and nobody else's (as do I), but I'm not very far off.

I broadly agree with Chris Cobb. The only pre-1950 candidates I strongly support had their careers distorted/hidden by the color line (Dolf Luque, Luke Easter, Hilton Smith) and a special case, Wally Schang. Everyone else is someone I've seen playing on TV.

(Re: Schang, when we first considered him, we were under the impression he was an awful defensive catcher. I'm convinced now he was average, even slightly above average, for his cohort. When you also account for the shorter career length of catchers in his era, he's certainly no less worthy than Ted Simmons or Joe Torre.)

Too-early prelim:

1. A-Rod
2. Schang
3. H. Smith
4. Luque
5. Abreu
6. Easter
7. Pettite
8. Sosa
9. Ortiz - Yes, he had no defensive value. But we don't have much precedent for excluding a 10,000 PA hitter with OPS+ > 140.
10. Posada
11. Wright - Very comparable to, and slightly ahead of, Sal Bando.
12. Hudson
13. Taylor
14. Evers
15. Rizzuto

First ten off: Berkman, Campaneris, Willis, Munson, Duffy, Garciaparra, John, L. Smith, B. Bell, Buehrle.

Disregard.

my clarification would be that there has been understandable guidance not to abandon a "hopeless cause" only for that reason.

meaning, as long as you think that player is among the best who have yet to be elected, don't dump him just because the crowd doesn't agree.

11 of my 15 turn out to be modern guys, but I still believe in Bob Elliott, for example. so he gets my "hopeless" vote.

it's a tricky balancing act, frankly - though I greatly appreciate the wisdom of Chris Cobb and others here.

With regard to Post #6 & #7

I don't follow the logic presented here that seems to be saying that when there are more teams & more players then there should be more Hall of Merit inductions. Just because the number professional players has increased from 500 to 750 doesn't mean that the quality of the top 10 players in the world has changed at all. The quality of play of the 10th best player of 500 is going to be the same as the quality of 10th best player of 750. By adding a bunch of AAA players to the major league pool doesn't do a thing for the quality of the top ten. I see no reason why there should be more Hall of Merit inductions in an era, simply because there are more teams. All of the additional players are not germane to Hall of Merit considerations. They are simply place holders so that all teams can fill their rosters

I understand that this can become very complex and nuanced when issues such as:

- time lining player skills and abilities
- opening up the major leagues to all races
- opening up the major leagues to players from all nations
- fewer of our top athletes are going into baseball and more are going into football, basketball, hockey, and soccer
- fewer players are playing professionally in more than one sport

are taken into consideration.

Reading these posts I feel that Big Brother is watching and judging my ballot by some very arbitrary, and quite likely, not appropriate standards to determine if my ballot is statistically spread over the eras in the "appropriate" way. As has been pointed out repeatedly, treating all eras "fairly" does not mean "equal representation from each era" Just make sure that each voter is taking all the eras into equal consideration is what is expected. If it is indeed true that the distribution of talent is not spread out as much in modern times as it was in the past, then obviously having the 5th best player as opposed to the 15th best player is not as big an advantage now as it used to be, therefore we should not expect players in the modern era to sail into the Hall of Merit if there are players from the past who were more dominant who are not yet inducted, as appears to be the case.

I don't think it is important at all that we all agree on the same philosophy. Each person should be free to decide how to handle these issues using their own feelings and best judgement. rather than feeling they have to fit their ballot to some pre-determined and arbitrary quota system.

So this discussion brings me to make a request.

When a fellow poster told me it is required that I take season length into consideration on my ballot, I assumed that was true. Could someone please verify this is the case and specifically cite where it says this in the requirements? I’d like very much to know if that’s indeed true, or whether I was misled for whatever reason. “Should” doesn’t count, as “should” has been advanced in arguments regarding war and post-season credit.

Thanks.

This was posted at the top of the 2021 ballot, I think this speaks to what all of us are needing to accomplish:

"You do need to treat all eras of baseball history fairly. You do need to stick to what happened on the field (or your best estimate of what would have happened if wars and strikes and such hadn’t gotten in the way). You may be challenged and ask to defend your position, if someone notices internal inconsistencies, flaws in your logic, etc.. This is all a part of the learning process."

I THINK Chris Cobb's point is about treating all eras of baseball history fairly.
If we've inducted more candidates from the 1920/30s than the 1970s/80s, we probably should review our calibrations to see if we are underestimating 70s/80s guys in some systemic way. With this, how to handle truncated seasons, wars, the negro leagues players, minor league play, etc. need to be part of our consideration when forming a ballot. How we accomplish this isn't an etched standard, just that we are doing our best to be fair to all eras.

bachslunch,

My sense is that you probably have more latitude than you have felt that you have had.

Let me put that response into historical context, as I understand it. Fairness to all eras cuts both ways. Back in the early days of the HoM project, win shares and Bill James' use of them to rank the top 100 players at each position in the New Historical Baseball Abstract were of huge significance to the HoM: our work was responding to James as much as to the Hall of Fame. One of the things that the electorate saw as a major shortcoming in the NBJHBA was that the rankings gave extremely short shrift to 19th-century ballplayers. One way in which James' rating system fell short was that he did not make any adjustment for the length of season, so that there was no way for a player who played 80-game seasons to accumulate as much value as a player who played 154-game seasons (162-game seasons were not yet on the horizon).

We wanted to do a better job with the early players. It became accepted at that time then that to be fair to all eras was crucial to adjust for season length, and "a pennant is a pennant" became one of our guiding principles: each season of play should be considered just as valuable as any other. The simplest way to achieve this goal was to pro-rate seasons to a common length, which was usually done by expanding the short ones. (Over time, there are fewer of those, so there's less multiplication involved.)

What was also found over the years is that a simple expansion of short seasons gives short-season players too much advantage, because it's easier to put up extreme numbers (and the players we are concerned with are mainly on the extreme positive side) in a short season: the longer the season, the more ups and downs and regression to the mean, just as across a career. Thus, this method of assessing value, which was introduced to be fair to early players (and that is in some fashion necessary to be fair to them), was also overcompensating to some degree. Therefore, it's better not to take the results of adjusting season length purely at face value.

So, coming out of the HoM's history, taking season length into account is an important part of the evaluation picture, but our evaluations of players aren't obligated to assume that (to take an extreme example) Ross Barnes' 1872 season, which he earned 4.3 bWAR in a 45-game season (which prorates to 15.5 WAR over 162 games), was better than Barry Bonds' 2001 season, in which he earned 11.9 bWAR across 162 games. At the same time, we are obligated to acknowledge that, in context, Barnes' performance contributed an even larger portion of his team's value for the season than Bonds' did.

To boil this history down into my understanding of the rules (which is only my understanding): My sense is that, in the evolution of the Hall of Merit, the Constitutional obligation to be fair to all eras became understood as including taking season length into consideration and not ranking early players below modern players simply because they couldn't compile as much value due to shorter seasons. However, (1) there is no single, prescribed method for taking season length into consideration, and (2) taking season length into consideration doesn't mean setting aside awareness of differing competitive contexts and their impacts on value and merit.

I have also had people say that I must adjust for such things as war, illness, postseason play, and minor league service. That's not true, and I don't do so. The only adjustment I do is for Negro League play as a counterbalance against prejudicial treatment, and I feel better about doing so because they actually played baseball and there are numbers one can work with. I do not feel comfortable assuming a player who took time off for military service necessarily would have the same WAR that their surrounding seasons had. I also don't believe in penalizing great players on bad teams by giving postseason credit.

I would never suggest adjusting for illness or injury.

I add war credit because my system basically demands it or I'll leave out some of the best players. If you're voting mainly on peak I can see how war credit wouldn't be as necessary. However, if you look mostly at career value you're giving these players a disadvantage. They didn't stop playing baseball to hang out at the beach. What would their value have been if they had been playing baseball? Nobody can ever be sure but the LEAST likely answer is that they would have contributed nothing at all. Even projecting their wartime service at the level of their 3 least productive years seems more fair than just giving them a zero.

I give minor league credit, especially for players before 1930 when Branch Rickey established the farm system, because the rules at the time didn't require a team to sell a star player to the big leagues. Baltimore loved having Lefty Grove pitch for them and draw big crowds and had little incentive to let him play in the big leagues without significant compensation. The path from the 1910 Pacific Coast League to the major leagues was not as simple as calling the player up from your AAA affiliate. The other era where minor leaguers were unfairly held back is with respect to black players during the "quota years" in the 1950s. I feel comfortable giving credit to minor league seasons because they did play baseball and we can come up with reasonable MLEs and because I don't give credit to an MLE season that projects as below average in the majors.

I give playoff credit but it typically isn't much more than a tiebreaker. Modern players are getting a LOT more playoff baseball time than previous generations and not giving playoff credit is going to systematically disadvantage the modern stars. Teams are not playing for "most regular season wins" and wild card spots have diminished pennant races. If teams have clinched a playoff spot they are giving their stars extended rest periods in September to keep them fresh for the postseason. This especially matters for pitchers because every high-effort pitch they make in the playoffs is likely shortening their career.

With respect to modern players I think we probably should be inducting more of them now than players who played 100 years ago but that's mainly because the older players are pretty well picked over and the newer players keep appearing every season. On my ballot the older players who bubble up to the top are the ones where there is a disagreement with the larger electorate on their merit, mainly due to MLE credit. There is a reason we have perpetual eligibility.

One possible exception to that rule is I would argue against including any more pre-modern pitchers. Pitchers like McCormick, Welch, Bond and Mullane put up innings totals that look gaudy compared to a starter in 2019 but they also stood 50 feet away, threw underhand, had 8 balls to a walk and had to pitch the ball where the batter called for it. Hell, just look at Mickey Welch. When they dropped the number of balls for a walk from 8 to 6 in 1884 his walk totals went from 66 to 146 and he led the league in walks allowed for the next 3 seasons. When you consider that baseball was dominated by balls in play and a home run happened about once every 5 games I am pretty sure what we are actually measuring is the defense to a great extent. Error totals were huge and double plays were quite low. I am also going to guess that catcher defense mattered quite a bit in that era since catchers typically produced almost no offense, often batting worse than the pitchers. McCormick, Welch, Bond and Mullane were not players we overlooked as the best of that era, they're the next tier down. There's even a good argument we should have left Pud Galvin out.

When a fellow poster told me it is required that I take season length into consideration on my ballot

I would say this isn't a "rule" but a group norm that a "pennant is a pennant" meaning a 2020 60 game season is worth as much as a 2019 162 game season.

The only official rules are here and some of them are out of date (we don't always start and end elections on Mondays anymore).

https://www.baseballthinkfactory.org/hall_of_merit/discussion/our_constitution

The only pre-1950 candidates I strongly support had their careers distorted/hidden by the color line (Dolf Luque, Luke Easter, Hilton Smith)

On Hilton Smith, thanks to the work of contributors to Seamheads we now have a much more complete picture of his body of work. The updated MLEs from Dr. C are actually pretty underwhelming. Smith made my PHoM several years ago but he's dropped to 131st on my current rankings on the basis of that MLE. 13 PWAA is not impressive.

HILTON SMITH                  
Year Age Lg G IP R RA9 lgRA9 RAA WAA pWAR Rrep RAR PA Rbat Rpos RAA bWAR WAR
1932 25 NL 36 180 87 4.35 4.60 5 0.5 2.3 17 22 60 -1 6 6 0.4 2.7
1933 26 NL 34 210 95 4.09 3.97 -3 -0.3 1.8 17 15 70 -1 7 6 0.4 2.2
1934 27 NL 36 200 119 5.34 4.68 -15 -1.4 0.7 19 5 67 -1 8 7 0.4 1.1
1935 28 NL 35 220 111 4.54 4.71 4 0.4 2.7 22 26 73 -1 7 7 0.4 3.1
1936 29 NL 42 270 130 4.32 4.71 12 1.2 4.0 26 38 90 -1 10 9 0.6 4.6
1937 30 NL 37 260 113 3.92 4.51 17 1.8 4.5 24 42 87 -1 10 9 0.6 5.1
1938 31 NL 40 260 114 3.96 4.42 13 1.4 4.0 24 37 87 -1 10 9 0.6 4.7
1939 32 NL 38 260 121 4.20 4.44 7 0.7 3.4 24 31 87 -1 10 9 0.6 4.0
1940 33 NL 37 270 88 2.94 4.39 43 4.9 7.6 25 68 90 -1 10 9 0.6 8.2
1941 34 NL 34 210 81 3.45 4.23 18 2.0 4.1 18 37 70 -1 8 7 0.5 4.6
1942 35 NL 28 180 99 4.94 3.90 -21 -2.2 -0.4 15 -6 60 -1 6 5 0.4 0.0
1943 36 NL 31 160 45 2.52 3.94 25 3.0 4.5 13 38 53 -1 4 4 0.3 4.8
1944 37 NL 4 20 7 3.18 4.25 2 0.3 0.5 2 4 7 0 1 1 0.0 0.5
1945 38 NL 29 180 79 3.93 4.46 11 1.1 2.9 17 27 60 -1 5 5 0.3 3.3
1946 39 NL 33 180 77 3.86 3.96 2 0.2 2.0 15 17 60 -1 6 5 0.4 2.4
1947 40 NL 33 180 107 5.34 4.57 -15 -1.5 0.4 17 2 60 -1 6 6 0.4 0.7
1948 41 NL 33 180 79 3.96 4.43 9 1.0 2.8 17 26 60 -1 6 6 0.4 3.2
   560 3420 1552 4.08  115 13.1 47.8 312 427 1141 -12.3 120.2 107.9 7.3 55.1

The other NGL pitcher who saw a drop in his stats compared to his reputation was Rube Foster, he did not sustain his success very long.

@20: Thank you for including the link. Given what's there, I feel I'm complying with the guidelines put forth.

If it is really that important to the Hall of Merit leadership, that we have equal representation of all eras, then we should simply have seperate votes for each era with a certain pre-determined number of players elected from each era. Of course that would raise issues for all players who careers spanned two different eras. I suppose each player would have to be assigned to the era where he had the most WAR, but voters would be instructed to consider his whole career.

The sad truth is that over time the impact of individual players on their teams success has gradually and continually declined. We are now in the middle of a trend that diminishes the value of great starting pitchers, and this trend is showing no sign of letting up. The reason for this is that the ability of relief pitchers to pitch to a lower ERA than the top starters in short stints and with the platoon advantage, after the starter has faced a line-up twice. As more and more teams are taking advantage of this knowledge, the role of starting pitching has lost a lot of it's lustre and importance. Hopefully the cream of the crop, will still get to pitch later into games and accumulate 200+ innings per season, but the empirical evidence seems to suggest that this will not be the case, as even the best pitchers are being somewhat limited in their usage. Blake Snell to confirm.

Could anybody have imagined a few years ago that post season baseball would resort to a string of relief pitchers for both teams in every game. Well, that is where we are.

This trend could easily be abated by severely limiting the number of pitchers on a major league roster, and by limiting teams ability to shuffle pitchers up and down from the minors to keep the pipeline sending pitchers back and forth on almost a daily basis.

If major league rosters were limited to 10 pitchers, at one time, and 15 pitchers for the course of the season we would be back to playing baseball where good starting pitching would be critical to success. But this makes to much sense and will never happen.

If it is really that important to the Hall of Merit leadership

This is not really a top-down organization as much as it's a big committee meeting.

The sad truth is that over time the impact of individual players on their teams success has gradually and continually declined.

For starting pitchers, yes, this is true. Partially because the job has perpetually gotten more and more difficult as baseball has evolved (something which could theoretically be adjusted for and therefor not necessarily be a strike against modern SP compared to historical ones) and partially because teams have realized there is a more efficient and effective way to use a roster of pitchers (which indisputably is a point scored for the point you are trying to make). For position players? This isn't true at all - they play more games in modern times than they ever have in historical times. However, STDEV is on a perpetual downward trend, to be interrupted only by expansion, really. To hold it against modern players that they are forced to play in an era with a massively larger and more skilled talent pool doesn't really seem like it's in the spirit of "fairness to eras." However, as others have noted, how to deal with this is largely a subjective matter.

Hey Cookie, are you a Yankee fan?

Yes, but my system doesn't have any Yankees bonus. Perhaps the HoM has an anti-Yankees bias ;)

:)

although "David Ortiz is 37th right now. David Wright is 42nd" is not helping your cause!

@19: Might have posted about this before, but if not, I'll do so now.

I don't give war credit because it's impossible to know what might have happened had the player been active instead of going off to the service. For Phil Rizzuto, for example, any of these scenarios might have occurred from 1943-45 if he had played instead:

-he puts up three amazing seasons, the best of his career, then continues on as he did in real life, maybe even better because of the extra experience. That's the best case scenario.
-he has three healthy seasons that are about par with what he has put up in 1941-42 and 1946-47. He continues on as he did in real life. This is the default you've chosen.
-he gets a Ray Chapman beaning early in 1943. If that happens, it's over. He plays two seasons and part of a third and dies.
-he plays typically in 1943, but gets his knee torn up in a takeout slide turning a double play in 1944. He's able to play, but nowhere near his expected capacity and never really recovers well. His later seasons aren't as good as what he actually did put up, and his career is shorter and maybe reduced to utility play.
-he reaches the series in 1943 after any sort of year, but he's caught betting on MLB games and banned from baseball for life starting in 1944.

Problem is, any of these scenarios (or something else) could happen, and we don't know which one. I'm not keen to pretend I'm clairvoyant and have the answer.

Don't get me wrong, I have nothing but respect and admiration for someone who goes off to combat like he did. But how to extrapolate his career given that he didn't play baseball at all? That's a can of worms I just don't want to stick my hands in.

With Negro League candidates, it's different. They actually did play somewhere, putting up stats that have been broken down and scrutinized, at least to some extent. For Hurley McNair, we know he played what amounts to a full season in 1915, then from 1920-28. We know that he barely played in 1910-14, 1916-17, 1932, 1934, and 1937, and did not play at all in 1918-19, 1929-31, 1933, 1935-36, before 1910, or after 1937. We also know that there was no option whatsoever for McNair to play in the AL or NL at the time because of his skin color. There's something to actually work with that's real life based and there's far less guesswork involved.

Minor league credit is a tough one. The player did play somewhere, at least. Exactly why they weren't in the majors? Good question, and there's potentially more than one answer. I'm also not sure there's been any systematic attempt to normalize minor league stats to major league equivalence as has happened with NeL numbers. Again, it's pretty nebulous, falling somewhere between war credit and NeL credit. Where it falls? Got me there. Maybe I'll figure some way to extrapolate this, but I don't know where to begin.

If folks feel confident filling in the blanks, more power to them -- but I'm just not comfortable with that level of speculation. Again, I'm not expecting others to think the way I do, but that's where my comfort zone sits for better or worse.

For my part, I'm willing to fill in war and minor league credit very conservatively. If a "natural career" progression looks like a guy may have lost three 5 WAR seasons to war service, I'm willing to give credit for roughly 5 or 6 WAR for that, maybe a bit more. If a guy missed ages 19-21 for the war, it's going to be pretty tough to convince me that's worth much more than basically a tiebreaker with another player. I take speculative credit very conservatively. Still, it also seems somewhat unfair to me to give no credit, though I think it is constitutional to do that also. The player missed opportunities due to circumstances outside their control, a full no-credit-at-all penalty just seems a bit too harsh to me. On this each issue, I do feel that it is "to each his own." I feel much more inclined to take a stance on other issues of fairness.

I respectfully would argue that anybody who is not giving war credit is not "being fair to all eras".

I am paraphrasing but JoeD, this project's founder, said that one of his biggest regrets regarding the HOM is that he did not mandate that all voters be required to give war credit. I 100% agree with JoeD on this.

"Don't get me wrong, I have nothing but respect and admiration for someone who goes off to combat like he did. But how to extrapolate his career given that he didn't play baseball at all? That's a can of worms I just don't want to stick my hands in.

With Negro League candidates, it's different."

well, both groups were MLB-quality players who didn't get to play in MLB in certain years through no fault of their own. it's not THAT different.

also, they did play baseball every year of the war - just not at a level even as organized as the Negro Leagues.

"We also know that there was no option whatsoever for McNair to play in the AL or NL at the time because of his skin color."

well, "We also know that there was no option whatsoever for Rizzuto to play in the AL or NL at the time because of his.... military service."

I'd be about equally surprised if someone made a pitch for giving war credit - but not credit for time in the Negro Leagues for players who also clearly were of MLB quality in certain years.

both groups got screwed in real life - granted, in very different ways - but I don't see any glory in insisting on perpetuating one screwing while alleviating another.

Lots to talk about here!
1) we are not expected to have equal or equivalent representation from all eras but only to be fair to all eras. To me that means -ish. What I mean is that anyone who treats eras fairly should have a roughly equivalent number of honorees for wide spans of time. But it’s not a rigid rule. So -ish.

1a) Cobb, I been banging this drum for years, and I’m glad to have some company! I appreciate the instructional perspective you brought to the matter as well. I agree fully that we overcompensated in the nineteenth century. I also agree that team-seasons is the best lens they which to view the question of fairness to eras. I would want to know if you included Negro Leagues seasons in your count. I ask because there are two schools of thought that could be applied. The first says, we should expand the pool of honorees for the Negro Leagues era because, obviously, there were more Leagues and teams. Another school says, we are electing an integrated list of players. Rather than add to the number of players from the Negro Leagues era, we should elect a number that would represent a sixteen team integrated MLB. If one follows the former school, then the number of honorees from before 1950 might look a little better. The second school pushes the Stan Hacks out of the HOM and would look more swollen.

1b) kwarren, your point about the top ten is the top ten seems accurate at first glance, but I don’t think it necessarily is. Let’s say we are talking about home runs. Pretend that in 2012 NL in a 16 team league, Bob finishes 5th and is just five homers away from the leader. Let’s also say that Fred, in 1950 in an 8 team league, finished fifth in homers and was also five away from the leader. Which is more impressive? I would argue that Bob’s achievement is more impressive. With 80 more competitors in the league to stave off, he ran a tougher gauntlet than Fred. Bob competed with 140 regulars; Fred competed with about 70. Why would we view these as equally impressive? By the same token, being one of the top ten players in 2016 is more impressive than it is in 1950. Combining that with the observation that the STDEV of performance gets higher the further back in time we look, i don’t think the top-ten-is-top-ten perspective is accurate. (It’s possible i misunderstood your reasoning, in which case my apologies.) That said, I agree that the total number of players in the league is not a good denominator. As you say, we don’t care about the LOOGIES and the September callups. We care about players with substantial playing time who have the opportunity to compete for a place in the top-n of players.

2) The core of support for remaining players from 100 years ago or more is pitching. At this point I have stopped ranking pre-a
1893 pitchers with post-1892 pitchers for all the reasons mentioned upthread. If you really don’t believe it, copy and paste MLB league pitching totals and error totals. Assign an appropriate value to the components, and look at what percentage of the RA by pitchers are actually their responsibility. Compare to 1908, 1938, 1958, 1968, 2018, whatever. I knew that olde tyme pitchers were more like initiators than what we think of as pitchers, but I didn’t know just how true that was. The number of runs allowed due to errors was MASSIVE in the early game and still prodigious in the 1880s. You might also find that pitchers today are responsible for more of the run prevention and RA than ever because error rates are minuscule and K rates so high. Pre-1893 pitchers just cannot be reasonably compared to their successors. I have gone to a practice of separating pre/post 1893 pitchers, figuring roughly how many slots the pre guys should occupy, and not dipping below that line. Electing more of them would be like buying the most fuel-efficient car built in 1920. Yeah, it got better MPG than its competitors, but that doesn’t mean it gets good MPG.

3) Credit scenarios. I’m cool with war credit for batters. I don’t give it to pitchers. The reason why is that traumatic, career-altering or career-ending arm injuries are far more common among pitchers than hitters. Far more common. We can all name a few hitters whose traumatic injuries ended or effectively ended their careers. Tony C. Thurman Munson if you want to be a little morbid about it. Guys like that. For all the hitters we name across baseball history, we can probably find the same number of pitchers from among a quarter or less of big league history. TJS is commonplace now, and until recent times it was a crapshoot whether a guy would come back. Dennis Leonard falling off the mound in agony is far more likely than Ray Chapman.

4) Bachs, sorry you’ve been feeling micromanaged. Don’t take it personally, we all get challenged. Just ask Rob Wood how many times he’s had to remind me that our constitution doesn’t call for positional balance! ;)

5) Minors credit: As some may know I’m generally cynical about baseball ownership and a perfect bleeding heart for the cause of trapped minor leaguers. My own research on the matter can be found in this series of articles: https://homemlb.wordpress.com/category/special-consideration/. Nutshell, there are relatively few who appear to merit it based on a blunt-force MLE approach. (A more nuanced approach might reveal more.) But guys like Cravath, Luque, and Cey may deserve it and could have their cases made by it. I’d recommend anyone who goes thru these articles to remember that the MLE shown there are not the real McCoy, more like a quick pass at them to see if a fellow’s case for special consideration might hold water.

My school of thought on war credit (which, as I have never voted in an HoM election, may not mean much) is basically the same as what Bill James said in (I think) the historical abstract: Don't give players credit for what they might have been (say, Herb Score without the injury), but do give players credit for what they WERE. Joe Dimaggio, Ted Williams, and Hank Greenberg were among the very best baseball players in the world in 1943; they just weren't wearing MLB uniforms at the time.

I am curious - what approaches does everyone take to downgrading (or not) the performance of the players who were still in MLB during the war? My general expectation would be that people who give war credit would downgrade the war seasons, and the people who ignore war credit would also ignore the simultaneous drop in league strength. But if anyone varies from that, how so and why?

Eric J,

thanks for your comments!

yes, I am a Bob Johnson acolyte and continue to vote for him - while also acknowledging that his best OPS+ came in a WW II war year. I do downgrade it, but don't dismiss it.

I do give war credit, but I don't downgrade the war years for players who didn't go to war. Part of it is the 'pennant is a pennant' idea - I don't particulally like the idea of counting the '44 and '45 seasons as less important, or expansion years like '61 either for that matter. I account for the change in run environment, but what a player does in that environment is fair game for me.

My general expectation would be that people who give war credit would downgrade the war seasons

I let the standard deviation adjustment do most of the work.

@32 #4: Thanks. Eh — getting your thoughts challenged isn’t fun, but it’s part of the deal here. I realize few people share my way of looking at things and some may think it’s utterly indefensible and ignorant, but so be it. It’s not like I’m trolling, which would be different.

I felt it was reasonable to explain myself, look at what’s said in response, incorporate or not as I see fit, and move on. If my ballot gets disqualified, so be it. I’ll just leave. In the grand scheme of things, it’s small beer compared to other life matters.

Dr. C--

You asked, "I also agree that team-seasons is the best lens they which to view the question of fairness to eras. I would want to know if you included Negro Leagues seasons in your count. I ask because there are two schools of thought that could be applied. The first says, we should expand the pool of honorees for the Negro Leagues era because, obviously, there were more Leagues and teams. Another school says, we are electing an integrated list of players. Rather than add to the number of players from the Negro Leagues era, we should elect a number that would represent a sixteen team integrated MLB."

I adhere to the first position. Here's an expanded account. The second view is untenable because it ends up setting a much higher standard for election in a quota-based system for the period of 1910-1950 than existed for the prior period of 1860-1910. There's a much larger pool of active, recorded talent in baseball once the Black Baseball starts to organize in the teens than there is in the prior half century, when African-American players were unable to participate in a game sufficiently organized to enable them to establish a record or, in most cases, to make a career. Therefore, I consider the advent of the Negro Leagues as a de facto expansion. The question then becomes how to account numerically for that expansion appropriately. I don't think that adding a number of team slots equal to the number of teams in the top Negro Leagues is the appropriate adjustment: because the Negro Leagues were drawing on a smaller population base than the segregated major leagues, that population could not support the same number of teams at the same level of quality as the white majors. In the simplified accounting I used in the post above, I simply added four teams to the major-league count for every season from 1920 to 1950. In my own more complicated decade-by-decade accounting, I start adding a smaller number of teams in 1900 and then contract to three teams during the 1940s. If the HoM electorate's voting is to be taken as a guide, the ratio of major-league to Negro-League electees from 1920 to 1950 is about 5 to 2, which suggests that six teams should be added for each decade 1920-50. In that case, however, the number of teams should continue to be set at 22 through the 1969 expansion -- otherwise, players in the interim will be squeezed. For this reason, I think setting the de facto expansion at 4 teams as preferable because it makes the HoM standard more consistent across the transition from segregation to integration.

To put the matter more bluntly: if you don't expand in 1920, you have to squeeze the 19th-century, or you will have held the 1920-50 players to a much higher standard. For very good reasons, not squeezing the 19th century was one of the fundamental goals of the HoM project. Therefore, expanding the number of electee slots as the Negro Leagues organize in recognition in the growth in the number of outstanding players, becomes the only sound option for fairness to the first half of the twentieth century. I am pretty sure that Joe D used just this sort of model when he revised the election schedule as we began to see in practice what was entailed to create a fully, appropriately integrated Hall of Merit for the period of segregation in which Black Baseball was organized (I expect enough digging would turn up the conversation threads about this: some of it at least is probably in the Number of Electees per year thread.)

Another crucial factor in favor of the expansion model is that it is compatible with the creation of MLEs for players who were excluded from the majors. It's challenging enough--as you well know!--to project a single Negro-League ball-player into a major-league competitive context. How much harder would it be if the process entailed projecting both Negro-Leaguers and major-league players into a hypothetical 16-team integrated league? The MLEs depend upon and project an expansion model, in which the level of competition for the MLE is set to the level of competition that historically existed in the major leagues.

bachslunch -- I don't think there's anything about your process that would create any basis for disqualifying your ballot, and I would encourage you not to worry about that prospect being a subtext of this conversation. If it's any reassurance, I would point out that the consensus scores show that you are closer to the current electorate's consensus rankings of candidates than I am or than DL of MN, who is doing so much work to sustain the HoM project!

I raise questions and suggest directions in order to shift the consensus in directions that it seems to me would lead to better results for the project, not to change the rules about what individual voters can and can't do. Sticking to rigid rules as a group would reduce our effectiveness. There are a few lines that define the mission and the boundaries of the project, but within that it's really about the voters considering each other's ideas and then using their own judgment.

I'd be about equally surprised if someone made a pitch for giving war credit - but not credit for time in the Negro Leagues for players who also clearly were of MLB quality in certain years.

To be consistent if you don't give white players war credit you should be aware that some of the MLEs for Negro Leaguers in the threads are assuming war credit. Others are assuming equivalence to a war-depleted major leagues so the same discounts should apply.

This is getting way too complicated for me, TBH. I want this to be fun, not an exercise in heavy duty higher math.

I’m a convert to WAR after being cautious of the approach for a while. (I also tout the stat in online HoF discussion forums that are sabermetrically hostile, and I’ve been known to make a nuisance of myself on the subject — which I haven’t minded in past.) I’m looking for a convenient catch-all stat that works, without a lot of monkeying around. I’ve found that controlling for position is needed, but WAR is what I’m looking for. If I can’t rely on this, participating in this exercise is pointless for me.

Moreover, if NeL players have war credit baked in while BBRef WAR doesn’t, then how do I reconcile this? As far as I can tell, I have two options:

-drop Ben Taylor from my ballot, as now I don’t know what to do with him.

-bull on ahead with Taylor and not be able to explain my ballot if called on it. Terrific.

Maybe dropping him is the best idea. I’ve seen enough chatter suggesting there aren’t any more good NeL candidates anyway. I’ll resubmit the ballot without Taylor. At least I won’t have an outlier to undermine the ballot. I’m also going to restructure the ballot to be more WAR strict. Bad news for guys like Vern Stephens, Tommy Bridges, and Gavvy Cravath, but I want to leave as few inconsistencies as possible.

Ben Taylor didn't play during WWII. He was active during WWI (didn't miss time) which isn't quite as tricky.

The player missed opportunities due to circumstances outside their control, a full no-credit-at-all penalty just seems a bit too harsh to me. On this each issue, I do feel that it is "to each his own.

It's not a penalty. We are supposed to evaluate what a player did on the field. He didn't play. Pretty simple. There are a lot of people giving to credit to relief pitchers for all the innings they don't pitch, and giving catchers credit because they don't play as much as other position players. They didn't play as much. It's not accurate to give them credit for what they did not do, and it's not unfair or a penalty to evaluate their career on the basis of what they did while they played. Seems that not many are giving credit to players who lost time because of injury. I really not sure why this type of absence is not excused as we do for players in the military, and for players who play positions that are so physically demanding where they don't get the opportunity to make equal contributions as other positions.

"What if" is a fun game to play and can be very interesting. But it's not appropriate to use it in comparing the "value" of player's career, particularly when you are comparing to players who did not miss any time. No contribution on the field means no value, and in the case of relief pitchers and catchers limited contribution means limited value. This is how it works in other industries. If you leave your job to do something noble, you don't ever get credit for what you "might/could/would" have done had you stayed at work either monetarily or in terms of rewards or recognition.

It's perfectly reasonable to opine that if Ted Williams hadn't had military service he might have a better career than Babe Ruth or Barry Bonds. But to go ahead and revise his career value, based on this assumption, and recognize him as if he actually did it, strikes me as a little absurd.

I am curious - what approaches does everyone take to downgrading (or not) the performance of the players who were still in MLB during the war? My general expectation would be that people who give war credit would downgrade the war seasons, and the people who ignore war credit would also ignore the simultaneous drop in league strength. But if anyone varies from that, how so and why?

My approach to war credit is to give it where appropriate and I also don't really downgrade active players during WW2. For war credit I see if a more conservative credit is enough to get them over the line, those cases are more likely to have traction with me than someone that needs high performing war credit seasons in order to get to the borderline. I do agree that we really don't know about potential injuries that may have occurred if they had been actively playing, but that to me is countered by us knowing that a lot of the guys we are evaluating did in fact avoid getting maimed or killed in combat (or were able to pick it back up after the layoff when not all could), so I think in those circumstances I can give them the benefit of the doubt in terms of hypothetical injury avoidance.

As for the actual war seasons, I go by something Steve Treder wrote quite some years ago about how the effect of the balata ball used during the war seasons actually depressed offensive levels. His approach, which is one I tend to follow in my assessments, is that the league quality drops are counterbalanced by the effect of the balata ball on the offense. So I end up calling it a wash overall for those seasons and don't adjust.

The sad truth is that over time the impact of individual players on their teams success has gradually and continually declined.

For starting pitchers, yes, this is true. Partially because the job has perpetually gotten more and more difficult as baseball has evolved (something which could theoretically be adjusted for and therefor not necessarily be a strike against modern SP compared to historical ones)

What makes you think that the job has "perpetually gotten more and more difficult". And now you want to treat them like catchers who play less than other position players. We will adjust their total value stats, so the we continue to elect as many as we used to. It would be absolutely terrible to a lot of the voters to induct fewer pitchers and more players at the other positions to reflect how the game is actually play. No that's unfair to catchers to now pitchers. I know we would like for all position to be equal, but they clearly are not. Let's base our voting on reality.

Look, things are getting a bit heated around here. There are certain inviolate principles on which this project rests. "Being fair to all eras" is number one. The Prime Directive.

There was a great deal of discussion pertaining to varying season lengths during the "founding" of this project. It was collectively decided that a 60-game season is just as important as a 100-game season. This was memorialized in the phrase "a pennant is a pennant". Obviously, league quality issues also should be taken into account when evaluating a player, but we will not allow any voter to dismiss a player's performance/contribution solely based upon a shorter season (a "standard deviation" discount was discussed and deemed appropriate though I am not sure how many voters actually employed this tack).

All that is by way of background for the following. I am certain that if this project began its evaluations of players right after WWII, we would have collectively decided that war year "credit" would be required in the spirit of being "fair to all eras". And I am positive that this requirement would have been written into the Constitution.

But, alas, the project began evaluating 19th century players and the war year issue was not pushed to any conclusion. However, even though no such requirement can be found in the Constitution, all voters agree to follow the spirit of the Constitution and its guiding principles such as being "fair to all eras" and "a pennant is a pennant".

As for the actual war seasons, I go by something Steve Treder wrote quite some years ago about how the effect of the balata ball used during the war seasons actually depressed offensive levels. His approach, which is one I tend to follow in my assessments, is that the league quality drops are counterbalanced by the effect of the balata ball on the offense. So I end up calling it a wash overall for those seasons and don't adjust.

This works if you look at raw offensive stats, but not if you look at league-adjusted measures like WAR or OPS+. (Also, the pitchers who stuck around during the war would get double benefits, from weaker competition and lower scoring context.)

It's not a penalty. We are supposed to evaluate what a player did on the field. He didn't play. Pretty simple. There are a lot of people giving to credit to relief pitchers for all the innings they don't pitch, and giving catchers credit because they don't play as much as other position players. They didn't play as much. It's not accurate to give them credit for what they did not do, and it's not unfair or a penalty to evaluate their career on the basis of what they did while they played. Seems that not many are giving credit to players who lost time because of injury.

Military service is treated differently from injury because the situations are different. A player who misses time due to injury is not capable of playing baseball during the missed time. Many of the players who served time in the military were not only capable of playing baseball during that time, they were actually playing baseball during that time. Giving credit for those seasons is not conceptually different from giving credit to players who are kept out of the majors for other reasons beyond their control, but demonstrate an MLB level of ability. Which is a common practice among many voters here.

#32 - Dr. Chaleeko

1b) kwarren, your point about the top ten is the top ten seems accurate at first glance, but I don’t think it necessarily is. Let’s say we are talking about home runs. Pretend that in 2012 NL in a 16 team league, Bob finishes 5th and is just five homers away from the leader. Let’s also say that Fred, in 1950 in an 8 team league, finished fifth in homers and was also five away from the leader. Which is more impressive? I would argue that Bob’s achievement is more impressive. With 80 more competitors in the league to stave off, he ran a tougher gauntlet than Fred. Bob competed with 140 regulars; Fred competed with about 70. Why would we view these as equally impressive? By the same token, being one of the top ten players in 2016 is more impressive than it is in 1950.

Your comparison of whether Bob or Fred achievement is more impressive comes down to the relative strengths of the player pools that they are competing in. You are treating the 80 additional players in Bob's league as if they are germane to the accomplishment. In reality I don't think they are. I am assuming that the 80 additional players in Bob's league are non-factors in the home run race. They would have been minor leaguers in Fred's time.

I used to play in a lot of world class bridge tournaments. Let's assume you had a competition with 100 pairs which included the top 20 pairs in the world. Depending on the length of the competition you would expect all 20 pairs to finish in the top 30 to 40 pairs. If we had another competition with the same 100 pairs, but adding 400 'AAA' pairs, over time the top 20 pairs in the world would again end up in the top to 30 to 40 pairs. In other words coming 10th in the first competition is not much different than coming 10th in the 2nd competition even with 5x's the number of competitors. There are at least 400 pairs who are not relevant and have no hope of seriously competing in the long haul.

Going back to your example Fred came 5th in the small competition and Bob came 5th in the large competition but the relative achievement is about the same. I don't buy that Bob "ran a tougher gauntlet", unless the additional competitors were of the same strength as the original ones. And that would not be the case given that they were all minor league call-ups. You seem to be implying that MLB can snap it's fingers and add players to fill their new teams that have the same % of Hall of Merit players as the original teams had. This doesn't seem to be a reasonable assumption, especially since many voters like to downgrade accomplishments that players have in expansion years. ie Roger Maris & Norm Cash

Going back to your example Fred came 5th in the small competition and Bob came 5th in the large competition but the relative achievement is about the same. I don't buy that Bob "ran a tougher gauntlet", unless the additional competitors were of the same strength as the original ones. And that would not be the case given that they were all minor league call-ups. You seem to be implying that MLB can snap it's fingers and add players to fill their new teams that have the same % of Hall of Merit players as the original teams had.

Before integration, this was essentially literally true.

What makes you think that the job has "perpetually gotten more and more difficult". And now you want to treat them like catchers who play less than other position players.

There's a little bit of putting words in my mouth here. But it's not hard in the least to prove that the job of starting pitching has continually gotten more difficult through history. Only one rule change in history made pitching easier - foul bunts being strike 3. On the other side of the coin: the mound was moved back, fewer balls to a walk, the ball was livened, the mound was lowered, the DH was introduced (in one league, so far), the strike zone was shrunk, first vertically then horizontally (de facto, at least). Structurally, outside of rule changes, the game has simply evolved to being far more difficult on starting pitchers. Hitters got access to video, essentially limitless practice time with pitching machines, video/tech swing machanics analysis, launch angle optimization, general improvements in conditioning, and so on. Pitchers got more accurate scouting reports on hitter weaknesses.

I don't mean to be snide here, but it's self-apparent. In the 1920s pitchers could survive on 85 MPH fastballs and a single offspeed pitch. The danger of opposing lineups is drastically different than it was 80 years ago, or maybe even 40. 94 MPH is average now. Pitchers have to go full strength from the outset of games. Batters are selected on their ability to draw walks, and trained to work counts to an extent that is much different than the conventional wisdom placed on how to approach at-bats historically.

Take those arguments or leave them, they're persuasive to me. But ultimately, K rates and HR rates are all-time high; pitchers are far more single-handedly responsible for run suppression than they ever have been in history. That point is entirely undebatable, and it becomes a very drastic difference the farther back you go. Walk rates have trends through history that are more nuanced and more interesting to trace, which I'll leave alone.

The forced increase in pitcher effort and stuff means the game has gotten notably more difficult for hitters as well. And all of this feeds into the STDEV observations which are latent to this whole discussion.

giving catchers credit because they don't play as much as other position players. They didn't play as much.

This, of course, is only true depending on how you measure playing time. A DH who plays 162 games gets credit for playing all of the innings but we all know they're only involved in 4 or 5 half innings and don't contribute at all on defense. The pitcher and catcher are the most important players on defense, this is indisputable. They represent 90% or more of the defensive effort expended during the game. A left fielder who fields a ball 3 times had a fairly busy day. The catcher will field the ball 150-200 times in a game. As strikeouts, walks and home runs have trended up the pitcher has had MORE importance in run prevention and the fielders have even less to do. Pitchers (and catchers) are taking on more and more responsibility for the defense.

Revised ballot.

Disclosures: I'm keeping things simple and what I believe to be consistent. Am assuming that all the deserving Negro Leaguers are already enshrined, an opinion seemingly held by several in the electorate. Players are ranked by position first using BBRef WAR amounts for the AL, NL, and PL as the sole determining factor. After that, I collect the top non-pitcher candidates at each position and order them as I think best, scattering anywhere from one to three pitchers into each group. The result is a 15 player ballot and 27 ranked off-ballot players, encompassing 10 pitchers and 4 players from other positions. While these are not popular approaches:

-I do not credit or debit for war, injury, illness, postseason play, or minor league service.
-I treat pitchers from all periods equally, but only consider the NL, AL, and PL legitimate. When considering 19th and early 20th century pitchers, I remove NA, AA, UA, and FL totals, with final numbers being approximate.
-I do not give relievers special treatment.

Will boycott 1st year candidates who bet on games, threw games, impeded players of color, were caught using PEDs post-2005 (Manny, ARod), and likely used pre-2005 if it looks like they'll get an immediate free pass by BBWAA HoF voters (IRod, Ortiz, Pettitte).

1. Wally Schang. Best C WAR available.
2. Buddy Bell. Best WAR at 3B. Have decided to trust the metric for him.
3. Jim McCormick. Best WAR for starters not in by a mile, even when removing all his UA-earned credit. Short career, but played in NL except for one UA season.
4. Bobby Abreu. Best WAR among available RFs.
5. Bob Johnson. Best WAR among available LFs.
6. Willie Davis. Now the best CF available by WAR.
7. John Olerud. Best WAR for available 1B.
8. Tony Phillips. Best WAR for available 2B.
9. Luis Aparicio. Best WAR for available SS.
10. Vic Willis. Good pitcher WAR, best after McCormick.
11. Sammy Sosa. Second best RF WAR. Happy to give him some benefit of the doubt given his treatment by the BBWAA.
12. Sal Bando. Second best WAR at 3B.
13. Joe Tinker. Second best WAR at SS.
14. Tommy John. Third in pitcher WAR.
15. Gene Tenace. Second best C WAR.

16-42. Jose Cruz, Tony Perez, Mickey Welch, Buddy Myer, Johnny Damon, Andy Pettitte, Bobby Bonds, Chet Lemon, Thurman Munson, Mark Buerhle, Lance Berkman, Bert Campaneris, Urban Shocker, Robin Ventura, Johnny Evers, Fred McGriff, Eddie Cicotte, Sam Rice, Norm Cash, Dave Bancroft, Chuck Finley, Ron Cey, Jorge Posada, Tony Lazzeri, Tim Hudson, Vada Pinson, Juan Gonzalez.

1B. Olerud, Perez, McGriff, Cash, Teixeira, Giambi
2B. Phillips, Myer, Evers, Lazzeri, Pratt, L. Doyle
SS. Aparicio, Tinker, Campaneris, Bancroft, Fregosi, Tejada
3B. Bell, Bando, Ventura, Cey, Harrah, Elliott
LF. B. Johnson, J. Cruz, Berkman, J. Gonzalez, Downing, Veach
CF. W. Davis, Damon, Lemon, Pinson, Cedeno, Puckett
RF. Abreu, Sosa, Bonds, S. Rice, Hooper, J. Clark
C. Schang, Tenace, Munson, Posada, Kendall, D. Porter
P. McCormick, Willis, M. Welch, John, Pettitte, Buehrle, Shocker, Cicotte, Finley, Hudson, Tanana, Whitney, Hershiser, Uhle, J. Powell.

All required disclosure players are on ballot or within top 40 except Ben Taylor; am accepting the idea that there are no viable NeL candidates left. None of the newcomers make my ballot, as I'm boycotting Alex Rodriguez (who would otherwise have been ranked #1) and Ortiz (who would have been a few notches off ballot).

If no further adjusted ballot is forthcoming, consider this version final -- and please post it to the appropriate thread if I'm not able to do so.

kwarren, when you say, "let's base our voting on reality," you state a value that I think most of the electorate would heartily agree with. The challenge of the HoM project is that the process for accurately discerning reality is not simple at all, and people are going to see reality very differently from one another. Appealing to reality and asserting that you see it clearly and others don't isn't going to go very far to persuade other people to change the way they see things.

If you expect other voters to accept or share your view of what reality is, you're likely to find their responses frustrating, and they are likely to be turned off by your arguments, which seem to seek to persuade by denigrating alternative views as not accepting reality. And that's too bad, because you're definitely raising issues and advancing a perspective that we should be thinking about. That's valuable to the project.

Consider that it's taken decades of pretty dedicated work by researchers and statisticians to get close to constructing a statistical model of value in baseball that is widely accepted, and there are still many different versions of WAR, many recognized shortcomings of WAR (especially around fielding & the value of catcher defense), and many serious dissents from the WAR methodology. And that's mostly dry mathematics. It's taken decades of work to create the Retrosheet play-by-play database, and that work is still ongoing, as is the work to recover the statistical record of Negro-League baseball. And WAR-based approaches have by no means cracked the question of how to adjust for changes in competition to make direct, numerical cross-period comparisons tenable with WAR alone, although there have been multiple attempts. Every judgment that we make is based on limited, imperfect knowledge, and the constructions of reality that we arrive at are going to be different. That's why we vote, and why having more participants with a variety of perspectives strengthens the project, just as listening to other perspectives and adjusting what we do based on what we learn also strengthens the project.

But when things get heated, as Rob Wood has noted that they are, people shout more and listen less, and that's neither fun nor beneficial to improving our collective judgment.

bachslunch, it's really interesting that Tony Phillips is a player that has really come forward in your re-thinking. Thanks for sharing a second preliminary ballot!

@55: Phillips is the best-ranked available 2B by BBRef WAR. If I'm going to be consistent and able to defend my ballot, he's in the top nine overall. I had originally discounted him because he played less 2B than the other top six, but not sure I can justify such thinking without getting fancier than I'd like. Gene Tenace is in the same boat, but so be it.

As for the actual war seasons, I go by something Steve Treder wrote quite some years ago about how the effect of the balata ball used during the war seasons actually depressed offensive levels. His approach, which is one I tend to follow in my assessments, is that the league quality drops are counterbalanced by the effect of the balata ball on the offense. So I end up calling it a wash overall for those seasons and don't adjust.

This works if you look at raw offensive stats, but not if you look at league-adjusted measures like WAR or OPS+. (Also, the pitchers who stuck around during the war would get double benefits, from weaker competition and lower scoring context.)

You are right about it being more for raw offensive stats and that it applies more to hitters than pitchers. I don't like using OPS+ for cross-era comparisons, I prefer using it for evaluation and comparisons within specific eras. And I also forgot to mention that another reason I have for not adjusting the WAR values for the WW2 years is that rWAR already bakes in some adjustment in the numbers. Here are the total number of base wins assigned for those years:

Year WAR NL WAR AL Total WAR % WAR NL % WAR AL
| 1948 | 292 | 216 | | 508 0.575 0.425
| 1947 | 292 | 216 | | 508 0.575 0.425
| 1946 | 279 | 228 | | 507 0.550 0.450
| 1945 | 235 | 213 | | 448 0.525 0.475
| 1944 | 236 | 214 | | 450 0.524 0.476
| 1943 | 236 | 214 | | 450 0.524 0.476
| 1942 | 233 | 211 | | 444 0.525 0.475
| 1941 | 254 | 254 | | 508 0.500 0.500
| 1940 | 252 | 252 | | 504 0.500 0.500

Another reason you see catcher at-bats dropping has to be the increasing length of a game. Compare the effort to sit in a squat for an extra hour every night to the effort it takes the DH to spit sunflower seeds in the dugout.

There are three primary actors in a baseball game: pitcher, catcher and batter. Arguing that "catchers and pitchers don't play enough" reduces the exercise to basically a "Hall of Batters." Yawn.

Going back to your example Fred came 5th in the small competition and Bob came 5th in the large competition but the relative achievement is about the same. I don't buy that Bob "ran a tougher gauntlet", unless the additional competitors were of the same strength as the original ones.

Well, let's look at why expansion has happened, and whether it has even kept pace with talent pool expansion or not. This would be a very interesting area of study. 1901 was the first year of the AL.

1901 MLB talent pool: White American players. US population: 70M white (~26.4M white ages 20-44, ~13M white males 20-44)
1901 MLB: 16 teams. 33K potential candidates for each roster spot.

1920 MLB talent pool: US population: 95M white (~36.5M white ages 20-44, ~18M white males 20-44)
1920 MLB: 16 teams. 45K potential candidates for each roster spot.

1940 MLB talent pool: US population: 118M white (~46M white ages 20-44, ~23M white males 20-44)
1940 MLB: 16 teams. 58K potential candidates for each roster spot.

1960 MLB talent pool: US population: 179M (~29M males 20-44). Latin America (Mexico + Cuba + Dominican Republic): ~37M (~7M males 20-44)
1960 MLB: 16 teams. 90K potential candidates for each roster spot.

1980 MLB talent pool: US + Mexico + Cuba + DR + Venezuela + Colombia ~350M (~65M males 20-44)
1980 MLB: 26 teams: 100K potential candidates for each roster spot.

2000 MLB talent pool: Above + Japan, South Korea ~637M (~118M males 20-44)
2000 MLB: 30 teams: 157K potential candidates for each roster spot.

2020 MLB talent pool: Above ~738M (~120M males 20-44 [US population has aged a lot without replacing that age group])
2020 MLB: 30 teams. 160K potential candidates for each roster spot.

This is a very quick and dirty spot check. But based on some obvious numbers, I'd say that Bob's accomplishment is more impressive, because he was playing in a league that drew from a much larger talent pool. The top X players in Bob's league were extremely likely to be of higher quality than the top X players in Fred's league. You're (mostly) right that we don't much care about the 60 players that get added to the back of a league in an expansion year. But that is not how expansion has actually worked in MLB's history. Expansion has not even come remotely close to keeping up with the explosion of the underlying talent pool that the league is drawing from. Add the ever-increased specialization in training for sports over time...but that is another discussion.

bachslunch (#53): I don't believe you should be ranking players in the way you are. That sounds blunt, so apologies. But do you believe Tony Phillips had more on-field merit than all the players ranked below him? It seems to me you have Tony Phillips ranked there because you are picking one guy from every position, and forcing other players who don't happen to be #1 overall at their position below him, even if you believe their overall value was higher. Does anyone else object to this ranking?

Bachslunch 53, straight Baseball Reference WAR has Jeff Kent ahead of Tony Phillips if you are looking for a 2b candidate.

Bleed in 61: Um, as of Jan. 7, 2021, Jeff Kent is no longer eligible for election to the Hall of Merit.

kcgard2 in 62: I don't think there's anybody below Tony Phillips on the ballot that some portion of the electorate wouldn't rank below Tony Phillips. The only players on bachslunch's prelim below Phillips that I have above Phillips are Bando and Sosa. Bando is not far ahead, and there are lot of arguments for ranking Sosa lower than I do. With the differences in value between the players being so small at this point, I don't think it's unreasonable to see getting the best eligible player at each position as a way of ensuring that you are putting the best players on your ballot, especially if you are doubtful about the way the position adjustments are weighted by WAR.

@61: Jeff Kent was just elected. Tony Phillips has the best available WAR among non-HoM 2B per BBRef as of now.

@60: That's exactly how I did this exercise. My understanding is that I need to be fair across positions, and the only way I can see to easily do that is to pick the best available at each position by WAR and arrange them. Best for each position not elected yet is:

1B: Olerud
2B: Phillips
SS: Aparicio
3B: Bell
LF: Johnson
CF: Davis
RF: Abreu
C: Schang

In order of worth, I'm listing them as:

Schang
Bell
Abreu
Johnson
Davis
Olerud
Phillips
Aparicio

I've also added the best pitcher to the mix, who in the approach I use is McCormick. I put him between Bell and Abreu.

The next set of eight position players is:

1B: Perez
2B: Myer
SS: Tinker
3B: Bando
LF: Cruz
CF: Damon
RF: Sosa
C: Tenace

They are ranked

Sosa
Bando
Tinker
Tenace
Cruz
Perez
Myer
Damon

For pitchers, I'm intermixing Willis, John, and Welch.

Lather, rinse, repeat. It's simple, clear, consistent, and defensible, no muss, no fuss. At this point, that's all I'm looking for. Like it or not, Phillips is the best rated 2B at BBRef by WAR. I have no desire to have to defend why I'm not being fair to all positions. I also don't want to get nuanced and have to defend something that's potentially nebulous.

Lol Chris Cobb, mind stuck in 2020 :;

This is a very quick and dirty spot check. But based on some obvious numbers, I'd say that Bob's accomplishment is more impressive, because he was playing in a league that drew from a much larger talent pool. The top X players in Bob's league were extremely likely to be of higher quality than the top X players in Fred's league.

Thanks for the quick number run on player pools over the years. The numbers do need an adjustment to account for the other professional sports growing throughout the 20th century onto today. NBA, NFL, NHL, MLS, etc they all take their players that in bygone eras would have gone to MLB.

Estaban (#57) - Thanks; those are really interesting numbers. Are the number of base wins for each league and year tabulated somewhere, or did you have to calculate them yourself? Has bbref provided an explanation/rationale for the league quality adjustments that it's making?

Hey, Bachs, just dropping a dime here because I like Tony Phillips a lot. I haven’t voted for him recently if ever because he’s been a little behind the frontlog, but I see him as electable.

Generally, let me toss out a link about Phillips and why I think his specific attributes may have accrued more wins to his team than any WAR framework might say: https://homemlb.wordpress.com/2015/04/08/how-much-extra-value-did-tony-phillips-versatility-create/

There aren’t many players this line of thinking actually applies to because most multipositional guys’ position changed year over year not game over game. Ben Zobrist is one player who might require this kind of analysis.

kc, while I don't agree with bach's method (or some of his position rankings), but I don't see anything unconstitutional about it either. Many of us (including me) have moved guys up or down their list because "he was the best catcher for 25 years" or "there's already x CFs from that era). Many of us want voters to be fair to all positions and I guess bach's has taken that view to an extreme. I personally think that this is not the same as having an exactly equal representation from each position. That said, words cannot express how I much prefer bach's ballot to someone with 15 OFs, 1Bs, and pitchers on it (I'm thinking of an MMP voter not an HOM voter).

@67: " The numbers do need an adjustment to account for the other professional sports growing throughout the 20th century onto today. NBA, NFL, NHL, MLS, etc they all take their players that in bygone eras would have gone to MLB."

There are other factors at work here other than pure demographics, which is that in addition to there being a lot more people available for each spot in major-league baseball in modern baseball in the early game, the resources available for making people into outstanding professional athletes are greatly increased.

As I am thinking about it now, I increasingly see expansion of the major leagues as being a measure of the resources available for creating and sustaining major league baseball teams.

The basic underlying assumption of this model is that the most straightforward way to understand what is needed to create a competitive major-league baseball team is resources, which should be defined broadly to include players, coaches, a scouting and player recruitment system, time and equipment and expertise for player development, and overall knowledge of the game. In our society, all these resources can/must be acquired with money, which is provided by the fanbase and advertisers who want access to that fanbase. The major leagues expand when it becomes possible and profitable to create and field additional major-league teams. The number of teams, then, is our best practical proxy for the number of major-league quality players that it is possible to develop via the baseball system that is centered on the United States and of which Major League Baseball is the apex.

If the ability to throw a baseball 100 miles per hour, with accuracy, from a pitcher's mound were purely an inborn talent, and the NFL were peeling off have off those physical geniuses to become quarterbacks, the fact of MLB's competition with the NFL for talent would be reflected in a reduction in power pitching. What we see, in fact, is the opposite trend: more and more pitchers are throwing harder and harder all the time. Improvements in pitching technique and physical training regimens are easily outstripping any dilution of the talent pool created by competition with other sports, abetted by the fact that the overall pool of people potentially available as professional baseball players is still expanding much faster than the available positions.

I think this goes along with kwarren's point that the contribution of any individual players to team success are, overall, diminishing. As major-league baseball becomes better and better at producing major-league players, the unique factors (physical and mental) brought to the game by individuals that enable them to stand out tend to diminish in significance.

That process includes recruiting, developing, and maintaining professional baseball players.

   66. Brent Posted: January 11, 2021 at 09:38 PM (#5999134)
Estaban (#57) - Thanks; those are really interesting numbers. Are the number of base wins for each league and year tabulated somewhere, or did you have to calculate them yourself? Has bbref provided an explanation/rationale for the league quality adjustments that it's making?

Yes, I got them from the following page at Baseball Reference, scroll down to the bottom and you'll find the table (it's named Replacement level for MLB and the WAR assigned to a league by year and league).

https://www.baseball-reference.com/about/war_explained_position.shtml

It's interesting to look at, particularly if someone is applying some adjustment to the numbers in their evaluations (there may be a case of over adjusting if one is not aware of this) or if one disagrees with the adjustment already built in by the win allocation (be it too harsh or nor harsh enough or in some cases being too sudden in shifting the league strength).

The page also has the positional adjustment values by year (which is a topic that's been discussed before).

Re post 69 -- I thought I had deleted the final line. If you thought that didn't make sense, that's because it doesn't: it's left over from an earlier version.

Esteban - Thank you! I've actually looked at that page several times but apparently I never made it to the bottom haha

That said, words cannot express how I much prefer bach's ballot to someone with 15 OFs, 1Bs, and pitchers on it

I also prefer it to that! Unconstitutional is obviously a strong word, I don't think it's unconstitutional probably. I guess to express myself better, it's a small step to make RP the 10th position, and then do rankings of positions 1-10, then the next 1-10...that might be a better way to highlight why I wonder about it. Or add DH as position 11. It makes clearer the probability that the ranking would not actually be order of merit. It would be constrained - hard to put into words what I mean, but perhaps something similar to strategic voting.

Someone recently wrote on a ballot or prelim or some thread somewhere on BTF that they considered pre-integration baseball to be AAA level compared to modern, fully integrated baseball. I’m doing a little research of my own that suggests that assessment might well be spot on to the penny. I haven’t gotten complete results yet but when I’ve got something more definite it’ll end up on my site. Just wanted to give props to whomever it was. I don’t know if you did your own research or intuited it, but good on ya.

As I do most years, I added a new component to my rankings this year, in my quest to make my primary ranking metric as fair as possible and as representative of my values as I can. This year, that addition was a small fixed value added for each year above specific WAR thresholds, so that burning extremely bright for a year or three doesn't outweigh having consistent above average to all-star level seasons for 8-10 years ( I don't like that outcome in my wWAR metric so this helps balance that out). Next year I will probably start to incorporate per IP/PA rate of accumulating wWAR, but without a ton of weight on that, just a little.

I like both peak and career value, I'm not especially picky about how you accumulate your value, but wWAR prefers peak over career (thus my bit of balancing). Overall my system still has a preference for peak over career just not quite as strong as it was. I've also more systematically handled my catcher value that I previously was kind of eyeballing as far as "uncaptured" value out there. Prelim:

1) Alex Rodríguez, SS (489 wWAR): extremely easy #1, averaged 116 b/fWAR, 76 WAA, could go on and on, dude was crazy productive, an all-timer
2) Sal Bando, 3B (192 wWAR): you can see how far ahead of the pack ARod is, I have Bando comfortably HOM worthy
3) Buddy Bell, 3B (183 wWAR): Bell is slightly more borderline to me, but this is still a quite electable range, had a very solid offensive prime to pair with all-time defense, the surrounding average is still valuable with his defense
4) Bobby Abreu, OF (180 wWAR): Well rounded player with career value that fits solidly in HOM corner OF ranks
5) Sammy Sosa, OF (200 wWAR): Serious peak seasons, but unclutch is not accounted in my wWAR number, if it were handled systematically would put Sosa in the range of 175 instead of 200, thus this ranking. Still fits in HOM corner OF
6) Tommy John, SP (170 wWAR): my pet candidate, please don't hold career length against him, that's a positive trait not negative :)
7) Kevin Appier, SP (165 wWAR): adjusted ERA 17% better than league average, underrated guy, look at him fellow voters
8) Andy Pettitte, SP (168 wWAR): wWAR edges out Appier, but in way more innings and adjusted ERA only 14% better than average to Appier's 17%. They could flip-flop in order but I like Appier better (better peak)
9) Bobby Bonds, OF (181 wWAR): Behind the pitchers because of pitcher representation, unlike corner OFers just above (virtual ties), Bonds comes from a more represented era (era fairness)
10) Bob Johnson, OF (168 wWAR): is this guy ever gonna get elected, or stuck around here forever? Not systematically handling minor league credit currently, could move him up a spot or few, but era over-represented
11) Lance Berkman, OF (167 wWAR): Berkman had a very nice peak inside a short-ish career
12) Eddie Cicotte, SP (167 wWAR): adjusted ERA 18% better than average (and FIP 12%)
13) Roy Oswalt, SP (157 wWAR): More similar to Santana than people realize, IMO... and we just elected Santana. Adjusted ERA 21% better than average, FIP 22%, both fantastic numbers
14) John Olerud, 1B (165 wWAR): Has been in the 15-25 range of my rankings for the last 4 years and finally gets to low-ballot spot after years of weaker classes.
15) David Wright, 3B (159 wWAR): I'm surprised, but he debuts in my final ballot spot, impressive peak in a career cut very unfortunately short. Edged out perennial Olerud ballot companion Robin Ventura by the narrowest of margins

16) Robin Ventura
17) Joe Tinker (hurt very slightly by my system tweak)
18) Sam McDowell
19) Brian Giles
20) Dwight Gooden

69) David Ortiz
78) Mark Teixeira
123) Jimmy Rollins
178) Carl Crawford
232) Jake Peavy
295) Prince Fielder

Ranks by position:

C: Munson, Tenace, Posada, Sundberg, Kendall (big mover), Porter
1B: Olerud, Perez, McGriff, Cash, Chance, Camilli
2B: Newt Allen, Marvin Williams, George Scales, Knoblauch, Lazzeri, Pesky
3B: Bando, Bell, Wright, Ventura, Cey, Harrah
SS: Rodríguez, Tinker, Garciaparra, Fregosi, Bancroft, Rizzuto
CF: Lemon, Cedeño, Lynn, Willie Davis, Pinson, Puckett
OF: Abreu, Sosa, Bonds, Bob Johnson, Berkman, Giles, McNair, Dale Murphy, Heavy Johnson, Luis González, Jim Rice, Colavito
SP: John, Appier, Pettitte, Cicotte, Oswalt, McDowell, Gooden, Finley, Wilbur Wood, Koosman, Buehrle, Kaat
DH: Giambi, Ortiz, Downing

Anyone want any more player threads? I considered throwing all the pitchers into a 2021 pitchers thread.

I think throwing the pitchers together makes sense. Nathan and Papelbon might ignite discussion of how we value relievers, and Peavy deserves a spot for at least someone to mention him.

And since Prince Fielder is not gonna get a thread, I'll throw a mention here. He snuck onto my rankings in the second-to-last spot.

I'll add Fielder to Teixeira's thread.

I've seen multiple members mention z-score work. Does anyone have such data available in spreadsheet format for me to tinker with for next year's ballot?

Cookie, I think it depends on what you are looking to do z-scores on. I use them for MLEs, and i imagine you have a very different use for them. Nonetheless, in case it helps, and in case anyone else is interested, here’s how to go about it.

If you are using z-scores in reference to BBREF’s WAR components, copy/paste from that huge page on their WAR explainer page that contains the daily updated WAR figures. All of their sub components of WAA/WAR are there for your use. Two formulas are particularly useful for this task. One is AVERAGEIF(), and it’s explainer on Excel is pretty clear. That’s how I figure the means that I’ll be using to compute the z-scores.

The other is the following statement and syntax. I’m going to assume you use the stats from every player in the league (STDEV.P) instead of a sampling of players (STDEV.S).

{=STDEV.P(IF(REFERENCE ARRAY=CONDITION, CALCULATION ARRAY, “”))}

It’s an array formula, so you need to press SHIFT-CTRL-ENTER when entering it. If CONDITION is a unique identifier you give a specific season in given league (e.g. 1980NL), it will calculate STDEV.P on all rows containing that unique identifier in the column you specify it to look in, and it will return a blank if nothing in the array contains that unique identifier.

Hope that is helpful.

this was WAY WAY too hard to dig up, but Edgar got in on his first try in the 2010 HOM voting

*Copied from the David Ortiz thread.

I was fascinated to look back on the 2010 top ten. #4 Cone is in, as are #9 Tiant, #10 Reuschel, and #12 Redding - a helpful re-balancing towards pitchers.

But of the remaining top ten, only Phil Rizzuto is still a viable candidate. The cases of #6 Gavvy Cravath, #7 Hugh Duffy, and #8 Bucky Walters have collapsed. Fred McGriff (debuted at #13), Dave Concepcion (#15), and Kirby Puckett (#16, including three elect-me votes) are also on life support.

Wally Schang, Ben Taylor, and Urban Shocker have risen through the backlog to pass them. So have a gaggle of 1970s stars: Bell, Bonds, Bando, Munson, and John.

Baseball Almanac recently posted a biography of Wally Schang. Seems very heavy on individual game details, but I thought it might be of interest here.

FYI - Retrosheet has started to add pre-integration Negro baseball games to its site. Our first release includes 215 box scores and just over 100 play-by-play accounts. Which isn't a lot, but hopefully we'll eventually add a lot more.

So looking at the next four years, we have 14 elect me spots, and only six that look like they have a good shot at being elected into the HoM: A-Rod and Big Papi next year, Beltran in 2023, Beltre and Utley in 2024, and CC in 2025. So that's at least 8 backlog guys getting elected. It's time for everyone to start stumping for their guys, especially if they're just outside the top ten returnees.

2024 also has Joe Mauer and Ichiro (his 2 games in 2019 fall under the token appearances rule). I'd put them in the "good shot at being elected to the HoM" category.

Yep, not sure how I missed them

am waiting for a BBWAA member to write a "hot take" column proposing that they switch to Hall of Merit rules for induction. would make their lives SO much easier.

:)

They might also consider adopting our policy of not inviting new inductees to give a public speech. That way, both electors and electees stick to what the players did on the field . . .

I’d like to request your help in assessing the various versions of WAR and other uberstats that are now available.

I recently decided to return to the HoM project after a six-year absence. During the historical elections, I was primarily a WinShares voter. When baseball-reference WAR came online, I switched to using it but always had concerns about some of their numbers. Back in 2012-13, I recall that they had a couple of large revisions to the WAR data.

I’d now like to review the various data sources and learn more about what’s available and the advantages and disadvantages of various sources of historical baseball data. I’ll start by telling you what I know and asking a few questions about various metrics.

Win Shares:

Published by Bill James in 2002, it was very influential during our historical elections held from 2003 to 2007. It might be the most comprehensively documented of the systems—I remember when reading the methodology thinking that if I took the time, I could probably replicate his numbers. But the system also had some well-known flaws:
- The replacement level was too low. My recollection is that it was around .200, compared to the current consensus of around .300.
- James forced a team’s win shares to match 3 × team wins. This implied that players on teams that were “lucky” in exceeding their expected (Pythagorean) wins or expected runs would have a higher win shares than players on teams that were less lucky. Other systems and most analysts have rejected this approach.
- Pitchers are undervalued relative to position players.
- The variation in fielding win shares is quite constrained compared to other measures of fielding.
- I don’t believe James utilized play-by-play data that was just starting to come out then.

Dan R’s WARP:

Developed by HoM participant Dan Rosenheck in 2007 as part of the HoM project, this system was also very well documented. The thread has 763 posts and Dan revised the system several times, so I’m not sure I’m aware of all of the modifications, but here are some of the features:
- Replacement levels are set empirically for each position based on the average of the three worst starters at each position in each league-season (taking a nine-year moving average and with an adjustment based on a Nate Silver study of freely available talent).
- Wins are standardized using a projected standard deviation for each league-season. Dan emphasizes that this standardization is not a timeline or league quality adjustment.
- His defense measure is based on an average of fielding Win Shares and the BP FRAA statistic; post-1987 it blends in play-by-play, and post-2000 it is mostly based on UZR ball-in-play data.
- Although one is not required to use it, Dan used a “salary estimator” to balance peak and career value.
- I don’t believe the data have been updated. I’m not even sure if the datasets are still publicly available.

My main concerns with Dan’s methods were:
- The estimation of replacement values seems is based on small samples and seems fragile. Also, because it is done position by position, it doesn’t adequately deal with players who are able to keep major league jobs because of their versatility in being able to handle multiple positions.
- I either don’t understand or don’t fully buy the standardization using projected standard deviations. I think this part of the methodology makes a lot of sense for handling seasons of different length. But given that Dan acknowledges that it’s not an adjustment for league quality, it’s not clear what it is supposed to be doing in other years.

Baseball-reference WAR:

Originally developed by Sean Smith, though later extensively modified by Sean Forman, rWAR seems to be the most widely used version of WAR. The methods are also pretty transparent, though there’s a bit of hand waving at the edges. Some of the system’s main features are the following:
- Pitching is based on runs allowed adjusted for team defense, with a different replacement level and leverage adjustment for relief pitchers.
- Defense is based on defensive runs saved (DRS) since 2003. For earlier years, it uses total zone rating (TZR), which is really several different systems. For 1952 to 2002 it’s based on play-by-play data, while for earlier years it’s just based on traditional statistics.
- Positional adjustments are based on differences in defensive performance for players switching positions. They are recalibrated every decade, which can lead to some fairly big shifts in the positional adjustment factors.
- There are some quite big quality adjustments that affect the historical replacement levels assigned to the two leagues. I haven’t seen an explanation for how those adjustments were derived for the era before interleague play.

In terms of problems:
- Maybe the most controversial feature of rWAR has been the adjustment of pitcher replacement level to account for team defense. If a team has a bad defensive efficiency ratio (DER) while one of its pitchers has a low BABIP, it’s likely that the team’s defense was bad for the other pitchers and relatively good for the pitcher with the low BABIP. Tangotiger wrote a blog post illustrating that problem for Aaron Nola in 2018.
- I’m also skeptical of some of the historical positional adjustments and league quality adjustments. I’ll discuss these in a subsequent post.

Fangraphs WAR:

Implemented in 2008 by Dave Cameron and David Appelman. The big differences compared with rWAR are:
- Pitching WAR is based on fielding independent pitching (FIP). So, by design, the pitching numbers leave out the effects of balls in play and of sequencing that are counted in pitching measures based on runs allowed. The website does, however, publish an alternative version based on runs allowed called RA9-WAR.
- Fielding is based on UZR since 2002 (TZR for earlier years).

While fWAR has good basic documentation, it left me with a lot of questions about the historical period. So, for example, I don’t know if positional adjustments are different for historical periods or if there are historical differences between the leagues in replacement levels. Do any of you know how fWAR handles these issues? When you see notable differences between fWAR and rWAR for a position player in the historical data, how do you interpret the differences?

Baseball-Gauge WAR:

I’ve just recently become aware of this one. I believe the main difference compared to rWAR is that fielding is based on Michael Humphreys’ defensive regression analysis (DRA). I’ve started reading his book, Wizardry, and Humphreys makes a case that his method is better than other methods for historical analysis of fielding.

My understanding is that gWAR is essentially rWAR but with fielding measured by DRA instead of the baseball-reference metrics. It uses a multi-year average of DRA to estimate runs saved above/below average—it isn’t clear to me why a multi-year average is needed, other than maybe the single year data may simply be too noisy. It appears that gWAR uses the same position adjustments, etc., as rWAR, even though in principle I think one would want to recalibrate those adjustments with a different fielding measure.

Are there any other features of gWAR that I should be aware of? Any comments on its usefulness?

Baseball Prospectus WARP:

I know they did a major overhaul of WARP a few years ago, but I don’t really know anything about it. (I haven’t really spent time on the BP website in years.) So I’ll ask you, what should I know about the latest version of WARP? What are its advantages/disadvantages relative to the other versions of WAR?

I would stump for Tommy John, but he's been covered quite thoroughly, and people seem really dug in on him. So I'll stump for...Kevin Appier.

Against
Tommy Bridges: better bWAR, fWAR, bWAA, fWAA, JAWS (by a lot), WHIP (by a lot); worse Win Shares (The 30s-40s Tigers were MASSIVELY better than the 90s Royals).
Urban Shocker: better fWAR (by a lot), bWAA, fWAA (by a lot); roughly equal bWAR, JAWS, and WHIP; worse Win Shares (again Shocker played for much better teams).
Bucky Walters: better bWAR, fWAR, bWAA, fWAA, JAWS, adjusted ERA, and FIP (all by a lot), WHIP; worse Win Shares (pattern, plus Walters had an innings advantage but not that big).
Dizzy Dean: better bWAR, fWAR (a lot), bWAA, fWAA (a lot), JAWS!, and Win Shares! (more innings); worse ERA and WHIP (more innings). It seems like Dean had a higher peak, but look, it's really not true (by WAR). Surprising?
Babe Adams: better bWAR, bWAA, JAWS, and adjusted ERA; roughly equal fWAR, fWAA; worse Win Shares (again Adams on MUCH better teams, and innings advantage).
George Uhle: Appier a much better pitcher, the question is whether you buy that a pinch hitter should get WAR credit for comparing against replacement level of pitcher hitting, basically. Uhle gets 10 WAR from it.

Against all these guys Appier played against tougher competition (integrated leagues and international scouting), as well as being from under-represented era. Appier also had a strike interruption, none of these players had any strike or war interruptions (I'm pretty sure).

Against
Frank Tanana: better bWAA, fWAA, JAWS, FIP, and adjusted ERA (all by a lot), roughly equal WHIP, bWAR; worse fWAR, Win Shares (huge innings disadvantage). This comparison is rough because of the drastic IP difference. Appier drastically better per IP, still better even if only comparing best seasons.
Chuck Finley: better bWAA, JAWS, adjusted ERA, FIP, and WHIP; roughly equal fWAA; worse bWAR, fWAR, Win SHares, somewhat notable innings disadvantage.
Tim Hudson: better fWAR, fWAA, JAWS (a lot), FIP; roughly equal bWAR, bWAA, adjusted ERA, WHIP; worse Win Shares (team quality and innings, same as always).
Mark Buehrle: better JAWS (a lot) and adjusted ERA; roughly equal fWAR, bWAA, fWAA, WHIP; worse bWAR and Win Shares (same story).
Andy Pettitte: I was trying to keep to similar IP, but for the modern guys I will expand. Better adjusted ERA, JAWS (by a lot), and WHIP; roughly equal bWAA, fWAA; worse bWAR, fWAR, Win Shares - Pettitte has big IP advantage (and team quality of course, by miles).

For every pitcher I've listed, Appier has the better peak/prime, by WAR. And it would be an even steeper victory if not for the strike against most of these guys.

That's my quick and dirty case for Appier.

I think that's a very fair and accurate assessment of the WAR frameworks, Brent. I don't know anything about BP WARP, either.

Kcgard92, #92,

Tommy Bridges has WWII and PCL.
Urban Shocker has WWII.
Babe Adams has WL/AA.

All that said, Appier is a good candidate, one I support/in the running for 11th-15th on this year's ballot.

While fWAR has good basic documentation, it left me with a lot of questions about the historical period. So, for example, I don’t know if positional adjustments are different for historical periods or if there are historical differences between the leagues in replacement levels. Do any of you know how fWAR handles these issues? When you see notable differences between fWAR and rWAR for a position player in the historical data, how do you interpret the differences?

As far as the positional adjustment goes, I believe Fangraphs and BBRef use the same numbers, but BBRef makes a slight adjustment to ensure it sums to zero. The difference is pretty negligible. They do also have a league adjustment, but it's only visible for hitters. You can see it here - for Ray Durham it's 27.3 runs over his career. BBRef does a similar calculation, but incorporates it into Rrep I believe.

A few more minor differences
- Park factors, with BBRef tending to be a bit more extreme
- BBRef uses a 59/41 split for position players vs pitchers, while Fangraphs uses a 57/43 split.
- Fangraphs baserunning is based on UBR, which is derivative of UZR. It's unclear what they use before 2002. BBRef uses PBP data for Rbaser and Rdp, so those are good back to 1931 for now.

Brent 91, I would suggest familiarizing with Kiko Sakatas Win-Loss records, of which he can explain here. Utilization of Retrosheet and a metric tied to team wins gives us some alternative insights.

For BP WARP, they have useful catcher defense metrics, but I don't use any other aspects from them.

For Baseball Gauge, Humphreys modified his DRA totals after his book published and they can be found at the website. One note is that the arm ratings aren't reliable. Second note that the Fielding spread is much wider than for TZ.

Fangraphs WAR includes modern framing estimates for catchers (2008 or so to present).

Brent, I'll second kcgard2's general response to your description of the frameworks. It matches what I know.

Here are a few notes.

1) For bWAR's cross-league quality adjustments, their notes say that they are based on players who moved between leagues and on interleague play.

2) fWAR definitely changes their positional adjustments over time. For example, in 1903 Jimmy Collins gets about 5 runs/162 games at third base, while in 2001 Robin Ventura gets about 2.3 runs/162 games. I have never found a place where the historical positional adjustments are presented or the methodology for those adjustments is explained. Their current adjustments are similar but not identical to those of bWAR. The current adjustments are listed for both systems (I can post them or point to them if you want): the historical ones can be derived with a fair degree of accuracy from the player records, which can be a tedious process.

3) Based on comparisons between fWAR and bWAR, I am pretty certain that fWAR either does not include a league quality adjustment at all or includes a much smaller one than bWAR. Fred Dunlap 1884 is the most obvious case, but 1920s NL players and 1950s AL players are pretty steadily higher by fWAR, I think. Their documentation suggests that, if they do adjust for league quality, they don't do it via adjustments to replacement level in the way that bWAR does.

4) When I see differences between fWAR and bWAR, my assumption is that league quality adjustments and differences in positional adjustments are likely to account for the majority of the difference between their evaluations for much of baseball history, although there can be significant differences in any area, except for fielding when both systems are using TZ. For recent players the differences between DRS and UZR can be significant, and the inclusion of pitch-framing for catchers in fWAR but not bWAR can make for vast differences in that particular case. For position players, if you want to spot where the differences come from for any given player, comparing the player value blocks in each system works pretty well. That's not helpful, of course, when crunching the numbers for a large number of players.

5) I don't know anything about gWAR or BP's current WARP beyond what you've described. I hope others can add more!

Jaack, Bleed the Freak, and Chris,

Thank you very much. Your explanations have been most helpful.

To add a few more details where my sense differs from Jaack's

1) The positional adjustment differences between bWAR and fWAR can be significant over a career. Here are the contemporary adjustments:

bWAR: C+9, SS+7, 2B+3, CF+2.5, 3B+2, RF/LF-7, 1B-9.5, DH-15
fWAR: C+12.5, SS+7.5, 2B/CF/3B+2.5, RF/LF-7.5, 1B-12.5, DH-17.5

2) The "League" adjustment Jaack has pointed out in fWAR is not a league quality adjustment: the numbers certainly don't vary in a way that would suggest that is what is going on. The description says that it is to "zero out wins above average." I think Jaack is correct that this league adjustment is similar to something that bWAR does within positional adjustments to get the numbers to add up to zero.

2) The "League" adjustment Jaack has pointed out in fWAR is not a league quality adjustment: the numbers certainly don't vary in a way that would suggest that is what is going on. The description says that it is to "zero out wins above average." I think Jaack is correct that this league adjustment is similar to something that bWAR does within positional adjustments to get the numbers to add up to zero.

It seems to function as a sort of league adjustment - for example Willie Mays is at 42.1 runs for his career while Mickey Mantle is at 9.5, which lines up with the NL vs AL difference for their careers. It's especially evident for 2020, where all the East players are in the negatives, which seems to only happen otherwise in the 60s AL.

That being said, it looks completely borked for the 19th century.