Voters have expressed essential two reasons for giving this bonus.

1. It is important to keep a balance between the positions and generally try to induct the same amount of players from all positions.
2. There is limited data on catcher's defense historically, and some sort of bonus is necessary to account for this gap in knowledge.

I agree with the various posters expressing concern over #1: it's fundamentally incorrect to artificially increase the valuation of one class of players for the sake of some ideal of parity. I'd say #2 is perhaps even wronger, since it's not an ideological matter but simply bad analysis. Dealing with limited data by giving uniform extra credit is going to give you, inevitably, exactly as many incorrect evaluations as correct ones. Assumptions aren't an appropriate way to deal with a knowledge gap.

All that said, all positions require some sort of positional adjustment to reflect their relative difficulty. I'd propose, semantically, that we stop talking about a catcher "bonus" and start talking about whether the positional adjustments in the various systems we use are accurate where catchers are concerned (a question we should be asking about all positions, of course). Personally, I think that the positional adjustments for catcher in both fWAR and bWAR are woefully insufficient, not because of playing time issues or anything like that, but because they don't adequately reflect the uniqueness of the position. The positional adjustments are supposed to represent runs gained/lost through changing positions, but the ease with which a SS moves to 2B or LF doesn't carry over to the catcher position. Only catchers play catcher. The positional adjustments at fangraphs and BB-Ref don't seem to me to capture this uniqueness.

Only catchers play catcher.

I guessed before it's mostly because players with poor pop times who can't control the running game are weeded out in the minor leagues. You can't play catcher in the major leagues if every walk or single turns into a triple in two pitches.

I'd say #2 is perhaps even wronger, since it's not an ideological matter but simply bad analysis. Dealing with limited data by giving uniform extra credit is going to give you, inevitably, exactly as many incorrect evaluations as correct ones. Assumptions aren't an appropriate way to deal with a knowledge gap.

I don't give blanket extra credit to deal with the knowledge gap. I give credit based on the fielding numbers we have and the time spent playing catcher. It's not great, but it's a ballpark estimate - generally guys who are good at non-framing defense are good at framing. Guys who are bad at framing are going to get moved off the position.

I view it similarly to estimating Negro League players or War Credit. It's not exact, but it gives you the idea
---
I do agree that framing is a zero sum quantity, like all defense. There are some players who are hurt by my framing adjustments - Joe Torre, Mickey Cochrane, Gene Tenace, and Ernie Lombardi. But Torre and Cochrane are still HoMers for me and Tenace and Lombardi aren't without it. My adjustment doesn't move anyone below the line, although Roger Bresnahan gets very close.

It does move Bill Freehan and Jim Sundberg move above the line.

I also think that most negative framing value comes from non-candidates and thus won't reflect as much in the voting. If you look at the worst framers it's guys who didn't stick at catcher like Ryan Doumit and Mike Napoli or general mediocrities like Chris Ianetta. This seems to be true for most defense oriented positions. Unless you play for the late 90s Yankees, you don't tend to stick at a position you can't handle. Yadier Molina's great defense isn't set off by one exceptionally bad framer but by 4 gus who were moved off the position or out of the league.

Some of the value is also going to come out of pitchers value, not other catchers.

(*) It looks like Fangraphs is only using framing numbers since 2008. They show Posada at -43.9 framing runs in the last 1,703.2 innings of his career. He played 11,172.1 innings at catcher before that, though. If he was equally bad throughout his career (and, if anything, framing seems like something you'd get better at as you aged / gained experience), that'd cost him an additional 288 runs and drop his WAR from 40.4 (current Fangraphs - BB-Ref has him at 42.8 w/ 0 framing adjustment) to something like 12.5 - which, by the way, seems completely absurd to me.

It is.

With the current framing data, there is ample evidence that catchers, despite their relatively limited playing time, have similar value to the elite players at other positions. With only eleven seasons of framing data, Fangraphs has three catchers over 180 runs of defensive value and seven catchers over 120. In the 140 previous seasons, there are only two catchers that they credit with over 120 runs of defensive value.

It appears that the framing numbers used by Fangraphs are not realistic. See Kiko Sakata's analysis of how it has been applied to Posada.

It appears that the framing numbers used by Fangraphs are not realistic. See Kiko Sakata's analysis of how it has been applied to Posada.

Check post 100, his framing became much worse as he got older, a general trend that holds for other catchers, so Kiko's original quote as to a WAR adjustment isn't relevant in this instance.

This has been my take on the framing numbers. I don't know that I'm right or wrong, it's been my interpretation.
1) Assumption 1: Framing has three participants, the catcher, the ump, and the pitcher but it's only credited to the catcher, which seems inaccurate to me.
2) Assumption 2: I don't know what percentage of the credit all goes where, so I am not giving it all to the catcher.
3) Assumption 3: Don't worry about where the other runs are going because on a per-pitcher basis the number is some combination of a PITA to calculate and probably small enough that it doesn't matter when you think about all the pitchers who pitch to a catcher over the course of a long career.
4) Assumption 4: Don't worry about the theoretical umpire runs because it ties my head into pretzel knots.
5) Also: There's been a lot of good work done on framing, so it's probably a good idea to use BBREF, FG, and BP's runs in equal proportion.
6) My solution: Give catchers one-third of their listed framing runs for each system.


OK, start picking until it comes apart!

BTW this is similar to what I do with Max Marchi's game-handling runs. I use those at one-quarter strength because there simply hasn't been additional study of the handling phenomenon.

I basically concur with the good Doctor above wrt framing runs, except I end up just halving the framing-run average instead of just taking one-third of it (I feel that while catcher, pitcher and umpire are all factors in the equation, catcher is the biggest factor). For pre-2011 seasons, I also average in Sean Smith's (Rally on BBTF) numbers that he derived on his site, baseball-projection.com.

Here's how my thinking goes:

* Catchers are 100% vital to a baseball game. It literally cannot happen without them. That's simply not the case with "backup pitchers".

* Catchers can have as monumental an impact on the course of a game as any other player on the field.

* The Hall of Merit exists to honor history's great players.

* Catchers can be great players too.

* The statement, "Catchers burn out quickly due to their abnormal workload", is true and provable.

* The follow-up statement, "Catchers are therefore less valuable than other players", strikes me as an absurdity of the first order.

* Because the position is unique, and carries with it unique physical and mental demands, I rate catchers against other catchers, and not against players who play non-analogous positions. It would be absurd to rate shortstops, say, against first basemen historically, because we'd have hardly any shortstops comparative to first basemen. This is untenable logically.

* Taking all of the the above into consideration, it therefore seems to me reasonable that catchers should be considered just like any other players, and that there should be about as many of them as there are representatives of any other position.

That's my take anyway.

And especially:

@: "I'm not really sure that there can be any reasonable debate on this issue."

I strongly distrust absolutist statements like this. We're all here to debate, to learn, and to share ideas. An inner conviction that one's position is faultlessly correct, and therefore cannot be debated, is a pretty certain sign in my experience that the opinion in question is anchored on logically dubious grounds. It approaches religious zealotry - never sound grounds for reasoned debate. An open mind is essential to a forum like this.

* The follow-up statement, "Catchers are therefore less valuable than other players", strikes me as an absurdity of the first order.

On a career basis this statement would be "true and provable". Even on a per/season basis catchers provide less value because they only play about 120 games when healthy for a full season. On a per game basis catchers are certainly as valuable as any other position player. The Hall of Merit voters are charged with evaluating a player's career, not his contribution on a per/game basis. Catchers in general tend to play less games. This makes their careers less valuable that players at other positions.

* Catchers can be great players too.

In terms of career value the best catchers are far behind the best players at other positions. Johnny Bench is the best major league catcher of all-time. Josh Gibson may have been better. He had WAR of 75.2 and JAWS of 61.2.

There are 7 1st baseman better than that, 6 2nd basemen, 9 3rd baseman, 8 shortstops, 5 left fielders, 7 centre fielders, 7 right fielders, & 31 pitchers...so there is no other position where Bench's value would put him in the top five.

And the pay level for catchers reflects this value as well.

OK everybody, chime in now with all the flaws in BBRef and the economics of baseball designed simply to short change catchers.

OK everybody, chime in now with all the flaws in BBRef and the economics of baseball designed simply to short change catchers.

Why is BBRef-WAR this indisputable thing? It's just a model. A good one, but not a perfect one.

Just for the record, here are the 1989 Hall of Merit Election results. Electees were (alphabetically) Johnny Bench (75 career WAR), Gaylord Perry (90.0 career WAR), and Carl Yastrzemski (96.4 career WAR).

RK   LY  Player                   PTS  Bal   1  2  3  4  5  6  7  8  9 10 11 12 13 14 15
----------------------------------------------------------------------------------------
 1  n/e  Johnny Bench            1298   55  44  7  3        1                           
 2  n/e  Carl Yastrzemski        1255   55  10 40  2  3                                 
 3  n/e  Gaylord Perry           1138   52   1  6 43     1     1                        

80% of the electorate endorsed the position that Johnny Bench was a more meritorious choice for induction than either Yastrzemski or Perry, both of whom substantially outpaced him in career WAR.

Side question for kwarren (or anyone else, I guess). Here are the top 10 single-season pitching performances of all time by bWAR:



Would you agree with that list as a top-10 all-time list of single season pitching performances?

All-time 'Votes Points' thru 2020 - those still eligible in 2021 election are in CAPS

caveats: I think it was ronw or RickA who have the truly accurate figures. I have riffed off those, but it has been a few years since these were corrected. so there will be a few compiling errors, and likely a couple of players who have clawed their way into the "top 50 unelected" category who I have missed. maybe I can get this tidied up this time.

but the basic point is that for newer voters, those who have gotten many votes over the years - but no longer do - might be worth a kick of those tires. that would include:
Van Haltren, JRyan, Grimes, Doyle, BMonroe, Williamson, Traynor, NCash, CMays, and Cepeda


TOP 50, ALL-TIME, unofficial (pts this year)
DUFFY...... 28129.5 (64)
VAN HALTREN 26923.5 (x)
Beckley.... 25856
Browning... 24502.5
Redding.... 19283
MWELCH..... 18971 (26)
Childs..... 18484
Griffith... 17924
Waddell.... 17596
Jennings... 16976

TLEACH...... 16248 (56)
ChaJones....15875
Bresnahan.. 14965
Sisler..... 13892
Pike....... 13399
CRAVATH.....12839 (34)
Sewell..... 12769
JRYAN....... 12678.5 (x)
Mendez..... 12555
SThompson... 12349

WALTERS.....12273 (33)
Roush...... 12005
Bennett.... 11503
Moore...... 10904
Rixey...... 10789
Caruthers.. 10704
BJOHNSON....10227 (121)
Beckwith.... 9896
GRIMES.......9876 (x)
DOYLE....... 9816 (x)

HStovey......9576
Mackey.......8930
AOms.........8385
Start........8378.5
McGinnity....8232
McGraw.......8145
DPearce......8073
McVey........7985.5
FGrant.......7969.5
BMONROE......7947 (x)

Kiner........7746
Suttles......7690
NFox.........7587
SCHANG.......7498 (176)
Trouppe......7494
WILLIS.......7355 (136)
WFerrell.....7259
WILLIAMSON...7035 (x)
CPBell.......6968
Galvin.......6585

Others in active top 50 (X if no votes this year):

DDean 6126, BTaylor 5827, Rizzuto 5692, Bridges 5609, Elliott 5395, Joss 5117, TPerez 4407, FChance 4120, BoBonds 3881, Traynor 3848X, NCash 3810X, CMays 3803x, McCormick 3586, Cicotte 3484, Cepeda 3465X, LBrock 3305, SRice 3301X, EHoward 2947X, BClarkson 2877X, VStephens 2740X, Singleton 2723X, Tiernan 2709X, FJones 2636X, Puckett 2620, Klein 2601X, Veach 2532, Mullane 2463, Staub 2396X, GJBurns 2388X, Lombardi 2363, Dunlap 2315, Newcombe 2310, Concepcion 2189X, Bancroft 2170

Positional adjustments in WAR-type frameworks adjust all positions so that they have roughly equal value. That is all first baseman have roughly the same cumulative value over the course of a season as all second baseman, all shortstops, etc., and all catchers. The positional adjustment reflects how "difficult" each position is by mainly looking at the offensive performances of the players who play that position. Shortstops on the whole hit worse than first baseman since it is "harder" to play shortstop than first base, so shortstops get a positive positional adjustment. For the purpose of this discussion, imagine that the positional adjustments are done on a season-by-season basis.

Okay, that is the seasonal adjustment. Let's pretend that this makes all positions "whole", including catchers (putting aside for the moment that catchers typically play fewer games in a season compared to other position players). Fine and dandy.

But as everyone knows catchers typically have shorter careers than other position players. The positional adjustment doesn't know (or care) about that fact. But we sure the heck should. If catchers have shorter careers, they typically will not be able to accumulate as much "value" as other positions over the course of their respective careers. This is a stark fact that nobody can deny. It is not a knock on catchers. Some of my favorite players were catchers. It is nevertheless a reality. I personally think it goes against the tenets of the Hall of Merit to adjust a catcher's career value to "account" for shorter catcher careers.

Feelings seem to running hot on this issue for some reason. I really hope we can all co-exist. My view is that the "founding fathers" left it up to each voter to decide how to deal with catchers.

A catcher is involved in every ball; as such his fielding value per game must be greater than anybody's except the pitcher, and perhaps 3 times the value of a shortstop who just has to stand there 2/3 of the time. Since his fielding value per game is several times that of other fielders, but he plays only 3/4 of the season, his fielding value per season must still be a (smaller) multiple of that of other fielders. Adjust for his shorter career and relatively low batting value (part of which is dictated by fewer games, part by wear and tear) and his overall career value should be comparable with that of other players. Hence there should be the same number of catchers (roughly) as any other position in the HOM.

Why is BBRef-WAR this indisputable thing? It's just a model. A good one, but not a perfect one.

It's not about BBRef. Use any of the other systems you like, or some combination of them all. With regard to catcher's career value they all tell you the same thing, as well as the way they are compensated. There is tremendous consistency between the measured value of catchers' value and their pay cheque. As much as we might wish it were so, all the analysts and the baseball executives/managers haven't got it wrong. It seems like we are trying to reinvent the wheel to make individual catchers seem more valuable than they actually are because we have been conditioned to feel the role is valuable, the starters must be equally valuable. This is not the case, as it is for all the other positions where the best starters play almost every game.

There is a huge difference between the value of the catching position, and the value of the players who play the position. Catching as a position is very important, but because the individual players' playing time is constrained the overall value of the position has to be shared more than at other positions. In this way catching has some of the same characteristics that relief pitching has, but to a far lesser extent. Relief pitching is a very important role for a team, but individual relievers are not that valuable because none of them pitches enough innings to make a meaningful contribution, no matter how well he does. As starters pitch less and less innings, the top relief pitchers will start to be closer to to starters in terms overall value. But the importance of starting pitching, both as a role, and the individual players performing that role is undergoing a considerable reduction now, and this trend will continue for a while yet, unless MLB reduces the roster size to limit the number of pitchers that can be on a team's roster.

Would you agree with that list as a top-10 all-time list of single season pitching performances?

Would depend on how you define a "top" player. It you mean relative to the league average, or to replacement level, then I would say yes, definitely.

In the 1800's there was a huge variation between the top players and average or replacement, much greater than we are familiar with. So the teams with the top players had a bigger advantage than they do now. It's sort of like basketball now. Having one of the top players in the league virtually guarantees you will be one of the top teams.

The player pool was small. There was no such thing as developing talent or farm systems. Talent was found on company teams and recruited to become a pro. Nobody thought of baseball as a profession. The top players were head and shoulders better than most of the competition so they had performances that were phenomenal. And they also pitched a huge number of innings in a season, which would clearly make the WAR totals higher. A 350 IP season an elite level would certainly rate as one of the top 10 of all time, I would think.

In those days all the pitching WAR was centered around four or five players. Now we have close to 20 players sharing the pitching WAR on many teams !


In response to #112

I agree with you 100% that the vast majority of the electorate thinks catchers need to be bumped up, just like they think that Mariano Rivera is one of the most valuable pitchers to ever pitch.

It depends on how you define "merit". If you compare Bench's "relative" superiority over other catchers, it would be much greater than Yastrzemski's and Perry's "relative" dominance at their positions. This does not equate to contributing to team wins as "value" does. It is this relative dominance that made Bench look better to most voters, and is also the reason that Mariano is so highly regarded. He is never compared to the top starting pitchers by those supporting him, but only to other relievers.

I notice that 11 people (20% of the electorate) did not drink the coolade. I would be in that group of 20%. Hopefully we can increase the percentage as people start to determine that there is no special reason why all positions should have equal representation. It makes perfect sense that the one position where playing time is constrained, that the players' would have less career value and therefore have a lower representation in the HOM and HOF than the other positions.

Johnny Bench is quite possibly the best catcher who ever played, and even he is only the 3rd most valuable player on the ballot in the year that he was eligible. This is further evidence that catchers' value in general is quite a bit lower than the top players at other positions. And this is another rationale for making him #1. No other catcher had ever been close to this before. Before Bench, the top catcher was Yogi Berra with a 59.8 WAR. So Bench was "the guy" to give the catching profession a little more "respect".

By the way, how did Josh Gibson do in the year that he was voted in ? I know a lot of evidence is word of mouth and reporters' narratives but he was a hell of a hitter, maybe at the Mike Piazza level, maybe better. And what was Gibson's reputation in terms of defensive ability ?

By the way, how did Josh Gibson do in the year that he was voted in ?

As they say in Bull Durham, you can look it up!

You can also look up his Negro-League batting records and his major-league equivalence projections.

You also might want to add Buck Ewing to your list of catchers that should be in the HoM even by your standards; it doesn't look like you've taken a look at the 19th-century players.

But hearing that 80% of HoM voters have "drunk the kool-aid" on catchers is the end of this conversation for me. There are lots of other things worth talking about, but I'm done with this one.

But hearing that 80% of HoM voters have "drunk the kool-aid" on catchers is the end of this conversation for me.

Agreed, this isn't a discussion anymore. Just some guy ranting.

"This is not the case, as it is for all the other positions where the best starters play almost every game."

Like Starting Pitchers? They only pitch every 5 days. I don't see you advocating that they are less valuable because of it. You keep ignoring the argument that catchers are involved in every defensive play. Even if a catcher plays 51% of his teams innings, he will likely be involved in a lot more defensive plays than any position player who plays every game. You have to field a catcher. The Reds with Johnny Bench were a lot better than the Reds with Bill Plummer. I'll ask you to ignore Replacement Level for a second. What do you think its easier to find by you standards; a great catcher or a great left fielder? Since you have less catchers in your personal HoM, I'd guess the answer is the LF. There is value in finding a great catcher over finding a great LF. If you have a great catcher, its super easy to find an average LF to round out your roster. If you have the LF, its relatively more difficult to find such a catcher. But you still have to field a catcher.

What do you think its easier to find by you standards; a great catcher or a great left fielder? Since you have less catchers in your personal HoM, I'd guess the answer is the LF. There is value in finding a great catcher over finding a great LF. If you have a great catcher, its super easy to find an average LF to round out your roster. If you have the LF, its relatively more difficult to find such a catcher. But you still have to field a catcher.

Yes it's easier to fine a great left fielder because there are more of them. This, in my mind, is an argument for having more left fielders than catchers in the HOM. I don't think it would be an argument for equal representation for both positions.

You keep ignoring the argument that catchers are involved in every defensive play. Even if a catcher plays 51% of his teams innings, he will likely be involved in a lot more defensive plays than any position player who plays every game.

I am not ignoring it or arguing it. It's well known and a given. I agree that catchers are involved in every pitch. I am not arguing against the importance of a catcher's defensive contribution. The defensive metrics that have been developed takes into account any extra wins a catcher contributes with his defense. What I am not in favour off is adjusting a catchers value rating on the assumption that his defensive contributions have not been counted properly. All the positions have their defensive contributions built in WAR or Win Shares or whatever you want to use. It's the lack of playing time that makes catchers less valuable, not a lack of defense contributions.

The defensive metrics that have been developed takes into account any extra wins a catcher contributes with his defense.

I disagree.

So, changing topics. I wanted to talk for a couple moments about Negro Leagues first basemen.

We currently have two of them getting votes: Ben Taylor and Luke Easter. I don't vote for either of them, but I've studied their careers as part of the MLE process, and I thought I'd talk a little about them.

Taylor, as I've said before, is kinda Eddie Murray but just a little lesser by the MLE. For me, he's below the in/out line since Murray is right on it. That said, among Negro Leagues candidates, Taylor has a very robust data set supporting him, a very robust oral history for his entire career, and MLEs commensurate with the general opinion of his career. That last bit is not intentional on my part, it turned out that way. I try not to wag the dog. The argument for Taylor is very clear: He's a longtime player with a Murrayesque arc who played during a slightly underserved era and whose MLEs we can have a fair amount of confidence in thanks to the (relative) wealth of data we have on him. I don't vote for him, but I don't think he's an obvious no-go either.

Easter is an entirely different candidate, and one that I do think presents major problems. There are, in my opinion, several major crisscrossing obstacles to analyzing him:
1) Before 1947, there is no statistical record of his exploits and little oral history about him
2) His statistical record in the Negro Leagues is not that impressive (288/380/481 in a league where we're going to dock him 20% off the top for quality of competition) and not that robust (just 419 PA)
3) His MLB statistical record is not that impressive (125 OPS+ at first base in almost 2,000 PA) but has a lot of hints at impressiveness
4) He had major injury problems from 1941 resulting from a combination of a fatal (not to him) car wreck that year and all kinds of foot and leg injuries later in his career that limited his mobility and possibly his playing time
5) It certainly looks like he wasn't given as long a rope by Cleveland as a white guy would have gotten
6) His minor league career stretched on into infinity after his MLB days were over, and his offensive output looked good.
7) He missed seasons due to wartime work.

It's not that any single one of these things is a problem. We can adjust and do what we do here for any one or even a couple of them. The problem is that they're all at issue, and taken together make evaluation of Easter pretty cloudy. As you all know, I've looked carefully and run the numbers, and Easter's MLEs are not nearly HOM worthy. He comes in under 30 WAA and under 50 WAR. I mark him as being something like Mark Grace in terms of his overall career, even if he's nearly the opposite in style and type.

So what I'm saying is that if you're looking at my current MLEs (not the old old old ones I did like ten years ago), I don't think there's a way to get from here to there and cast a vote for him. The amount of conjectural credit goes well beyond someone like Gavy Cravath, for example. In Cravath's case, we know exactly where and when and how often he played in the minors, we just don't have leaguewide stats for those minor leagues to create an MLE. But when he got to the majors to stay, he was an excellent player, so there's little imagination required to generate an MLE. He was a first string player from the moment he reached the PCL, was good enough to get callups in 1908-1909, and was awesome once he stuck in MLB. It's a pretty simple story.

Contrast that with Easter. Here's a guy for whom you have to be willing to
a) undo a car accident and resultant injuries
b) decide on a starting date to his career even though there isn't really clear evidence about when he reached a level of play we'd consider apt for MLEing
c) figure out how much of his career was futzed with by Cleveland
d) decide that a 40+ 1B with bad legs in AAA could have kept up with MLB pitching and played full time in a time of less optimized minor league stratification
e) determine what his level of play was prior to 1947 with no statistical record at all
f) determine war credit to give him even though there's no before-the-war stats
g) decide that his Negro Leagues seasons were more impressive than they actually were.

I mean nothing personal against Easter or his voters, but I like to think of a vote for a Negro Leaguer as being a balance between what we think he is and the risk that he's not what we think he is. I'm very certain that Ben Taylor is what we think he is. I'd say the ratio of knowing to risk is like 90:10. With Cravath I feel about 80% certainty. I'd like to have those leaguewide PCL stats in my hands. But 80:20 is plenty to consider a vote on. With Easter the amount of conjecture and the narrative crosswinds are so high that he's at best a 50:50 in my opinion. Probably lower than that. 40:60 maybe? That's not much to base a vote on. I'm not saying that anyone should put Taylor on their ballot instead of Easter. Really, I wanted to comment on both and to provide helpful contrast between their cases. But I will tell you that I think voting for Easter is difficult to provide solid reasoning for because all the conjecturals must go his way to even be a borderline candidate.

Respectfully submitted,

Dr. C.

The defensive metrics that have been developed takes into account any extra wins a catcher contributes with his defense

This is exactly what (almost) everyone who disagrees with you specifically disagrees with.

IF we had pitch framing data for Johnny Bench, his career WAR in all probability would have been much higher than what is ascribed to him by WAR now. Perhaps as high as the all time greats at other positions. But the defensive metrics DON't include that value for Johnny Bench, because they didn't exist. That does mean the VALUE didn't get created by Bench over the course of his career. Pitch framing numbers in the short amount of time they have existed have tended to say that catchers might provide as much as 2-3 WAR in a single season via pitch framing value. That value, if counted (at least for top framers) would easily more than compensate for a playing time reduction of 20% per season, and probably cover most of the difference left from having a shorter career.

I agree with Dr. Chaleeko that only a portion of this value should be ascribed to the catcher, and it seems to me people can have reasonable disagreements about how much of the credit for this value catchers should get. However, it is definitely something.

Now, some metrics, for very recent seasons only, do incorporate pitch framing, and I imagine that voters will tend to give far less of a vague "catcher bonus" to these players once they start hitting the ballot, because the main driver of "hidden value" from catchers will be accounted for in the metrics for them. For all the historical catchers we've ever discussed for HOM so far, it is not the case. As an aside, I think I'm looking forward to automated strike zones so this won't even be a discussion that much longer.

Johnny Bench is quite possibly the best catcher who ever played, and even he is only the 3rd most valuable player on the ballot in the year that he was eligible. This is further evidence that catchers' value in general is quite a bit lower than the top players at other positions.

Bench leads both in MVP/CY Award Shares. To put it another way, the Giants would've jumped at the chance to trade Perry for Bench, and the Reds would've laughed. The same could've been said about the Red Sox with Yaz, although it would be closer (if for no other reason than the presence of Fisk in the system).

Also, by your argument on pay equaling value, then closers must be the most valuable player on the field, as their $/WAR are leaps above any other free agent position. Which means Mariano should be a top ranked player all time. And we don't all only compare relievers to relievers. My system compares everyone to each other, and Mariano slide in between my 59th-ranked starter Stan Coveleski and #60 Jim Bunning. Seems like he is a perfectly reasonable selection to the HoM. And since I do adjust catchers, he's between #13 Bill Freehan and #14 Thurman Munson. All five are in the back end of my top 200.

Pettite is ranked 91st by WAR and his peak is worse than Tommy John's.

Among eligible pitchers on this HoM ballot, Pettite ranks:
8th in rWAR
3rd in fWAR
4th in gWAR
2nd in my composite WAA
8th in WAR per inning.

He's the only pitcher on the ballot to be in the top ten in all of these categories. He's one of five pitchers with a single digit median ranking, and the only other one that didn't pitch in the 1800s is Urban Shocker.

Among all players on my HoM eligibility list, he ranks:
16th in rWAR
4th in fWAR
14th in gWAR
5th in my composite WAA
16th in WAR per

He has the lowest sum of those rankings, and 7th-lowest median.

All time, he's 144th in my system, and the 43rd-best starting pitcher.

To put it another way, the Giants would've jumped at the chance to trade Perry for Bench, and the Reds would've laughed.

I would trade Bench's career for Perry's in a heartbeat.

Also, by your argument on pay equaling value, then closers must be the most valuable player on the field, as their $/WAR are leaps above any other free agent position.

Relief pitchers have the lowest value of all positions, and they are also the lowest paid of any position. It's perfectly consistent.

He's one of five pitchers with a single digit median ranking

A rather interesting metric.

Relief pitchers have the lowest value of all positions, and they are also the lowest paid of any position. It's perfectly consistent.

Except they are not the lowest-paid on a per win basis: https://blogs.fangraphs.com/the-evolution-of-positional-differences-in-free-agent-costs/

It's a few years old, but the trend holds true - free agent relievers are paid 1.5-2 times more per win than other free agents.

@127: "I would trade Bench's career for Perry's in a heartbeat."

Personal feelings aside, I try to fairly consider every argument advanced here - and for me, this one's a damn tough call...

Perry's got longevity on Bench for certain; he's 9th all time in games started, 6th in innings pitched. Bench's longevity is damn good too, 17th all time at catcher, but simply not on Perry's level. Big points to Perry. (I'd imagine most fans would never guess that Perry is top-10 in durability at his position, or that Bench isn't.)

I don't see Perry as the best pitcher in his league in any given year (I'm happy to be contradicted on this). When he won the Cy in 1970, it should've been Bob Gibson or Tom Seaver (imho); in 1978 it's closer, but I personally would've given the Cy to JR Richard. Bench, on the other hand, was (again imho) clearly the NL MVP in 1970. In '72 I would've given it to Joe Morgan, but Bench is right up there. So it's a big edge on peak to Bench. Career, though, has always been more important to me than peak.

Perry made it to the playoffs just once, which isn't his fault, and really sucked, which is. Bench was a very good post-season hitter and defender, with a Series MVP in '76. Big points to Bench.

Perry's career WAR is 90.0, 46th all time; Bench's is 75.2, 77th all time. The caveat there is that a lot of Bench's value is tied up in his defense - and everyone who evaluates catcher defense, in my view, is severely dialling it down - Win Shares, Linear Weights, and the newer metrics too. So, an edge to Perry - but less of one (imho) than WAR might indicate.

Perry's ERA+ is 117, not great for an 'all-time' pitcher, but a wonderful, grinding number for a 22-year man. Any pennant contender would want him. Bench's OPS+ is 126, a nice solid perennial All Star-type of number. Same as Perry, any team would take him in a heartbeat. Maybe a slight edge to Bench, but really, I call this one a wash.

It's a lot closer for me, I think, than it would be for some folks here. There have been, by my count, only three catchers in the history of baseball who could match Bench's overall level of offense and defense - Gibson, Berra and Campanella - whereas for Perry, there have been less than 20 starters who could match his longevity and grinding year-to-year excellence. Maybe only a dozen or so. (I simply don't have the time to break it down further than that).

Truth be told, despite Bench's historic stature at catcher, it's tempting to go with Perry. Over his best 15-year stretch, 1966-80, Perry averaged 285 innings pitched with a 2.82 ERA. In the context of the modern game, that's monstrous. Don't misunderstand - I'd take Bench comfortably over most pitchers - and most position players too, for that matter - but Perry's longevity and consistence are something special to behold.

Taking all in all, I think I'd end up selecting Bench, on the grounds that Bench would improve the game-calling, framing, and defense against baserunners for all of my pitchers, and I'd only see Perry every fourth day.

But it's damn close. Certainly it's not an "in a heartbeat" call in my book.

I'm speaking here, of course, as if I owned a team. If we're talking Hall of Fame/Merit status, it isn't close - it's Bench in a blowout. Bench is a reasonable candidate as the greatest of all time at his position, an Inner Circle guy, and Perry is not.

I don't see Perry as the best pitcher in his league in any given year (I'm happy to be contradicted on this). When he won the Cy in 1970, it should've been Bob Gibson or Tom Seaver (imho); in 1978 it's closer, but I personally would've given the Cy to JR Richard.

Perry's first Cy was in the 1972 AL (he finished second in the 1970 NL voting). His competition would have been Wilbur Wood (10.8 bWAR for Perry, 10.7 for Wood). He also led the AL in pitching bWAR in '74, though he finished fourth in Cy Young voting.

bWAR strongly prefers Phil Niekro over either Perry or Richard in '78, for what that's worth.

Eric: Thanks for the catch. I linger over my ballots, but otherwise I type these things so damn fast... I'm working on something else, something big, 7 days a week - but I do try to keep my hand in here, because the baseball discussion here is the best on the Net, and I learn from pretty much everyone, pretty much every day. I don't comment on everything because often I've got nothing I want to add, but I read it all.

Re: '72, Perry does have a strong argument that year. Luis Tiant was very strong per-inning wise, but had half the innings of Perry & less than half of Wilbur Wood's... I'll still take Richard as my Cy Guy in '78 though. ;)

Perry's first Cy was in the 1972 AL (he finished second in the 1970 NL voting).

Gaylord's brother, Jim, won the 1970 AL Cy Young award.

Perry's career WAR is 90.0, 46th all time; Bench's is 75.2, 77th all time. The caveat there is that a lot of Bench's value is tied up in his defense - and everyone who evaluates catcher defense, in my view, is severely dialling it down - Win Shares, Linear Weights, and the newer metrics too. So, an edge to Perry - but less of one (imho) than WAR might indicate.

Perry's career WAR is 90.0, 46th all time; Bench's is 75.2, 77th all time. The caveat there is that a lot of Bench's value is tied up in his defense - and everyone who evaluates catcher defense, in my view, is severely dialling it down - Win Shares, Linear Weights, and the newer metrics too. So, an edge to Perry - but less of one (imho) than WAR might indicate.

As I read it, you are saying that Bench has defensive value that is not reflected in the metrics of more than 15 WAR. You are also saying that "everyone who evalutate catcher defense" is severely dialing it down. Do you think this is deliberate or out of ignorance ? Do you have any stats or information to support your conclusion other than, "in my view".

I do agree with you, that if this indeed the case and your view is accurate then you are correct. Bench had more value than Perry. I just don't see how all the analysts could be so wrong on Bench's defensive value, and you have somehow got it right without even telling us what Bench's true defensive value is, or how you arrived at it.

One think about WAR is that it is fixed sum. If you add WAR to one player then you have to deduct it somewhere else, presumably in this case that would be the pitchers that Bench caught. If these pitchers had to give some of their pitching value to all the catchers that caught them, they wouldn't have a lot left.

You may be exactly correct. But, I don't see how you arrive at your conclusion, and at this point I don't buy it. That's why I prefer Perry's career.

Does anyone here have a breakdown of the decade in which each elected player's career is centered? That, for me, would be as useful as a positioinal breakdown; it would help in determining which eras, if any, are under-represented in the HoM.

"I just don't see how all the analysts could be so wrong on Bench's defensive value, and you have somehow got it right"

I'm going to try to ignore the continued provocations of a certain poster, and just talk with the rest of you. At the moment I'm sorry I bothered to respond to anything s/he wrote.

Doc and Cobb have had great posts on decade enshrinees, I have a spreadsheet I keep/I'll try to post shortly also. Side note, I've been celebrating the life's work of Neil Peart, and hope you are too progrockfan.

@137:

Bleed, can you possibly point to those posts of Doc's and Chris's? Cheers!

As to your side note (C sharp I believe), the words to Freewill changed my life. RIP Neil.

They posted in the ballot discussion threads on this topic a few times from ~2015 on, but not sure they had an actual list of electees, maybe just a total number from each area...I'll work up mine as a starting point and others can chime in with insights to assist...

Bleed, can you possibly point to those posts of Doc's and Chris's?

See comments 543, 544, and 549 of last year's discussion thread.

Cheers Kiko!

Chris Cobb had noted some players with split decades either if they were better suited to be 50/50 in 2 decades, or they were upper echelon greats and had considerable value in 2.

Apologies if my counting is off :(

1860s (3)
Dickey Pearce
Lip Pike
Joe Start
Ezra Sutton

1870s (6)
Cap Anson .5
Ross Barnes
Paul Hines .5
Cal McVey
Jim O'Rourke .5
Al Spalding
Deacon White
George Wright

1880s (17.5)
Cap Anson .5
Charlie Bennett
Pete Browning
Dan Brouthers .5
Bob Caruthers
John Clarkson
Roger Connor .5
Buck Ewing
Pud Galvin
Jack Glasscock
George Gore
Paul Hines .5
Charley Jones
Tim Keefe
King Kelly
Jim O'Rourke .5
Old Hoss Radbourn
Hardy Richardson
Harry Stovey
John Ward

1890s (17.5)
Jake Beckley
Dan Brouthers .5
Jesse Burkett
Cupid Childs
Roger Connor .5
Ed Delahanty
Bill Dahlen .5
George Davis .5
Frank Grant
Clark Griffith
Billy Hamilton
Hughie Jennings
Willie Keeler
Joe Kelley
John McGraw
Bid McPhee
Kid Nichols
Amos Rusie
Sam Thompson
Cy Young .5

1900s (17)
Roger Bresnahan
3 Finger Brown
Fred Clarke
Jimmy Collins
Sam Crawford
Bill Dahlen .5
George Davis .5
Elmer Flick
Rube Foster
Home Run Johnson
Nap Lajoie
Christy Mathewson
Joe McGinnity
Eddie Plank
Jimmy Sheckard
Honus Wagner .5
Bobby Wallace
Rube Waddell
Cy Young .5

1910s (18.5)
Pete Alexander .5
Frank Baker
Max Carey
Ty Cobb .5
Eddie Collins .5
Heinie Groh
Joe Jackson
Walter Johnson
Pete Hill
Pop Lloyd
Sherry Magee
Jose Mendez
Dick Redding
Edd Roush
Louis Santop
Tris Speaker .5
Cristobal Torriente
Honus Wagner .5
Ed Walsh
Zack Wheat
Smokey Joe Williams

1920s (21.5)
Pete Alexander .5
John Beckwith
Oscar Charleston
Ty Cobb .5
Eddie Collins .5
Stan Coveleski
Red Faber
Frankie Frisch
Goose Goslin
Harry Heilmann
Rogers Hornsby
Dick Lundy
Biz Mackey
Dobie Moore
Alejandro Oms
Eppa Rixey
Bullet Joe Rogan
Babe Ruth
Joe Sewell
George Sisler
Tris Speaker .5
Turkey Stearnes .5
Dazzy Vance
Jud Wilson

1930s: (27.5)
Earl Averill
Cool Papa Bell
Ray Brown
Mickey Cochrane
Joe Cronin
Bill Dickey
Martin Dihigo
Wes Ferrell
Willie Foster
Jimmie Foxx
Lou Gehrig
Charlie Gehringer
Josh Gibson .5
Lefty Grove
Gabby Hartnett
Billy Herman
Carl Hubbell
Ted Lyons
Joe Medwick
Mel Ott
Satchel Paige .5
Red Ruffing
Al Simmons
Turkey Stearnes .5
Mule Suttles
Bill Terry
Arky Vaughan
Paul Waner
Willie Wells

1940s: (21)
Luke Appling
Willard Brown
Lou Boudreau
Joe DiMaggio
Bobby Doerr
Bob Feller
Josh Gibson .5
Joe Gordon
Hank Greenberg
Stan Hack
Monte Irvin
Charlie Keller
Ralph Kiner
Buck Leonard
Johnny Mize
Stan Musial .5
Hal Newhouser
Satchel Paige .5
Pee Wee Reese
Enos Slaughter
Quincy Trouppe
Ted Williams .5

1950s: (18)
Hank Aaron .5
Richie Ashburn
Ernie Banks
Yogi Berra
Roy Campanella
Larry Doby
Whitey Ford
Nellie Fox
Bob Lemon
Mickey Mantle .5
Eddie Mathews .5
Willie Mays .5
Minnie Minoso
Stan Musial .5
Billy Pierce
Robin Roberts
Jackie Robinson
Duke Snider
Warren Spahn
Ted Williams .5
Early Wynn

1960s: (21.5)
Hank Aaron .5
Dick Allen
Ken Boyer
Jim Bunning
Roberto Clemente
Don Drysdale
Bill Freehan
Bob Gibson
Al Kaline
Harmon Killebrew
Sandy Koufax
Mickey Mantle .5
Juan Marichal
Eddie Mathews .5
Willie Mays .5
Willie McCovey
Brooks Robinson
Frank Robinson
Ron Santo
Joe Torre
Hoyt Wilhelm
Billy Williams
Jimmy Wynn
Carl Yaz .5

1970s: (24)
Johnny Bench
Bert Blyleven .5
Rod Carew
Steve Carlton
Darrell Evans
Rollie Fingers
Carlton Fisk
Bobby Grich
Reggie Jackson
Fergie Jenkins
Joe Morgan
Graig Nettles
Phil Niekro
Jim Palmer
Gaylord Perry
Rick Reuschel
Pete Rose
Nolan Ryan .5
Tom Seaver
Mike Schmidt .5
Ted Simmons
Reggie Smith
Willie Stargell
Don Sutton
Luis Tiant
Carl Yaz .5

1980s (23.5)
Wade Boggs
Bert Blyleven .5
George Brett
Gary Carter
Roger Clemens .5
Andre Dawson
Dennis Eckersley
Dwight Evans
Goose Gossage
Tony Gwynn .5
Rickey Henderson .5
Keith Hernandez
Paul Molitor
Eddie Murray
Willie Randolph
Tim Raines
Cal Ripken .5
Nolan Ryan .5
Bret Saberhagen
Ryne Sandberg
Mike Schmidt .5
Ozzie Smith
Dave Stieb
Alan Trammell
Lou Whitaker
Dave Winfield
Robin Yount

1990s (23)
Roberto Alomar
Jeff Bagwell
Craig Biggio
Barry Bonds .5
Kevin Brown
Will Clark
Roger Clemens .5
David Cone
Tom Glavine
Ken Griffey
Tony Gwynn .5
Rickey Henderson .5
Randy Johnson .5
Barry Larkin
Greg Maddux
Edgar Martinez
Pedro Martinez .5
Mark McGwire
Mike Mussina .5
Rafael Palmeiro
Mike Piazza
Cal Ripken .5
Ivan Rodriguez
Curt Schilling .5
Gary Sheffield .5
John Smoltz
Frank Thomas
Larry Walker

2000s (15)
Barry Bonds .5
Jim Edmonds
Vladimir Guerrero
Roy Halladay
Todd Helton
Derek Jeter
Randy Johnson .5
Andruw Jones
Chipper Jones
Pedro Martinez .5
Mike Mussina .5
Manny Ramirez
Mariano Rivera
Alex Rodriguez
Scott Rolen
Curt Schilling .5
Gary Sheffield .5
Jim Thome


If anyone else has run this type of analysis, share your thoughts so I can tweak my accounting here :)

Re: the distribution of HOMers through time, it occurred to me that my database can (partially) answer this. Here's a table of total pWins earned by HOMers by season for all of the seasons for which I've calculated Player won-lost records. I don't have data before 1918 or for Negro Leaguers. For seasons between 1918 and 1931, I extrapolated missing games to tie to actual games - that might cause some quirkiness (it definitely does something in 1919, for example). I also normalized all seasons to 162 games. I show totals and also pWins per team - hopefully the latter will pick up some of the problem from lacking Negro Leaguers.

Anyway, the results seem to make sense and, frankly, produce a surprisingly (to me) smooth series outside of 1943-45 and since about 1998 or so, for what I assume are obvious reasons in each case. We're maybe a little soft in the early days of integration - although there may be some Negro Leaguers filling in some of the gap there - and in the late 1970s and 1980s. But overall, the results look pretty well distributed. I could do something similar by position if people are interested.

Thank you so much for posting this analysis, Bleed.

Quick reactions:

It looks like baseball prior to the live-ball era is less well represented than post-1920. The electorate may well feel that's a correct balance. I'd love to hear opinions on that. Obviously the balance is influenced by the number of active franchises & the number of players on major league rosters.

It also strikes me that the 1950s are perhaps under-represented - and that the HoM should think long & hard before electing more players from the 1930s.

Overall, though, I'm struck by what an excellent job of proportionally representing the whole of baseball history has been done by the electorate - certainly better than the HoF. It matches my subjective impression of the site before seeing these numbers. And to be honest, that was the #1 factor that first encouraged me to post here - the fact that the electorate not only encourages, but mandates a view across the whole of baseball history.

@143: "I could do something similar by position if people are interested."

Yes please Kiko!

My lists match Bleed's very closely, although I place a few players (12) differently: I'll show my differences below.

Anyone wanting to use Bleed's lists -- I found only two errors: (1) The count for the 1940s should be 20, not 21; (2) we haven't elected ARod yet (he's listed for the 2000s). That's a very clean list!

To address the issue of representation, I'll give my per-decade "quotas" here. These are based on the number of top-level professional teams in each decade. I use rough MLEs for Negro-League teams, and I leave the UA and FL out of consideration. For the 1860s, I use the existence of professional players as a proxy for teams. I don't mark a contraction for the 1890s, but I do mark contraction for the 1940s and the 1950s, due to the impact of WW2 and the collapse of the Negro Leagues before ML became fully accessible to Black and Latin American players.

Decade (Quota) # Elected -- My Differences from Bleed the Freak's list

1860s (2) 2 -- Pike and Sutton to 1870s

1870s (7) 8 -- Pike and Sutton from 1860s; Hines to 1880s only

1880s (14.5) 18 -- All of Hines here

1890s (14.5) 17.5 -- Same as Bleed the Freak

1900s (17) 18.5 -- All of Wagner here; Hill here instead of 1910s

1910s (19) 16 -- .5 Wagner and Hill to 1900s; Roush to 1920s

1920s (20.5) 22.5 -- Roush here from 1910s; Willie Foster here from 1930s; Oms to 1930s

1930s (20.5) 28 -- Foster to 1920s; Oms here from 1920s; .5 Ray Brown to 1940s; Greenberg here from 1940s

1940s (19) 19.5 -- .5 Ray Brown here from 1930s; Greenberg to 1930s

1950s (18) 18 -- Same as Bleed the Freak

1960s (20.5) 21.5 -- Same as Bleed the Freak

1970s (24.5) 23 -- Fisk to 1980s

1980s (25.5) 24.5 -- Fisk here from 1970s

1990s (27.5) 22.5 -- .5 Ivan Rodriguez to 2000s

2000s (30) 14.5 -- .5 Ivan Rodriguez here from 1990s

Comparing our inductees to demographic quotas based on the number of teams, I see pre-WW2 baseball as very well represented: over-represented, in fact for some decades. There's a bit of a gap in the Deadball era, which Dr. Chaleeko's more exact player counts pick out better than rougher by-decade counts, but it would be filled by adding just a player or maybe two with careers in the 1910-20 area. I see 1940-70 as about right, although for these decades to have parity with the pre-WW2 game, we could afford to elect another half dozen players. 1970-90 still needs some attention, but just a few players would need to be added to be in proportion to the demographics. I agree entirely with Dr. Chaleeko that most attention needs to go to the period of 1990 to the present: we should continue to elect the bulk of the players from this period, inducting the obvious HoMers as they come in and selecting the best from the high backlog.

That doesn't mean we shouldn't elect any earlier players. My estimates are that, at the current size of the game, we should be electing about 30 players per decade (pending future expansions or contractions). Our schedule has us electing 35 players per decade, so there's room for about 1 player every other year from the deep backlog to go in without short-changing recent players too much. However, since those early eras are already "above quota," we might do better to concentrate on the borderline players from 1940-90 instead of the borderline players from 1870-1940.

The trickiest question for us is how to deal with changing evaluations of earlier players as more data becomes available. Fielding and baserunning values could shift the evaluation of earlier players in ways that reshuffle the lower-tier HoM rankings quite a bit. Already the shift in the electorate from Win Shares as the primary metric in early years to bWAR and fWAR makes some selections (e.g. Edd Roush) look like mistakes, and some of the NeL picks have likewise come into question as the statistical record has been filled in. This sort of shift will surely continue.

My own approach, at this point, is to continue to support early players whom my system sees as clearly above a quota-derived in-out line for their period. That is, they have to be better than a clear standard for their era, not just better than other elected players from the era or (apparently) better than post-1990 players. For me, there are just a few of these, as I have been and continue to be a high-consensus voter. The situation may look different to others.

Overall, I am not worried about fairness to all periods becoming a serious problem: the backlog is lined up so that if it moves forward, mainly recent players will be elected, with an occasional earlier player. That seems about right to me, but it's still something worth keeping track of and discussing.

Thanks to Chris Cobb, Bleed the Freak, Kiko Sakata and anyone else who has looked into our era representation. It appears that we have done a pretty good job on that front.

Both oWAR and dWAR include the positional adjustment. They're kind of measuring different things (and dWAR is mis-named). The premise of oWAR is "How valuable would this guy be if he was an average fielder?" - the idea is this would be one's starting point if one wanted to substitute in one's own fielding numbers.

The premise of dWAR is to be able to compare players at different fielding positions - "How much better (if any) is a bad-fielding shortstop than a good-fielding first baseman?" dWAR equals the positional adjustment plus fielding runs - both of which are calculated relative to average, so this should really be called dWAA - which would have the added benefit of making it more obvious that WAR <> oWAR + dWAR.

So in summary if I understand you correctly, the positional adjustment is counted twice when you add offensive WAR and defensive WAR together.

So in summary if I understand you correctly, the positional adjustment is counted twice when you add offensive WAR and defensive WAR together.

Correct.

A question for the electorate:

In your view(s), what's the single best summary stat to show the measure by which a hitter veers above or below the mean of a league - and removes as many illusions of context as possible? Preferably this stat will be comparable across eras.

I'm thinking OPS+...?

151. progrockfan Posted: January 22, 2020 at 05:29 PM (#5918448)
A question for the electorate:

In your view(s), what's the single best summary stat to show the measure by which a hitter veers above or below the mean of a league - and removes as many illusions of context as possible? Preferably this stat will be comparable across eras.

I'm thinking OPS+...?

wRC+ is a step improvement from OPS+ at Fangraphs as a good blunt starting tool, then factor in baserunning and defensive ability, then consider adjusting players that performed better in contextual "clutch" situations with Baseball-Reference or Fangraphs clutch value or pWins at Kiko's site.
Standard deviations impact wRC+ though, so using at face value will hurt/help different leagues and eras.

Since there is no single stat that is immune from "biases" I would not recommend any one single stat. I would guess that most HOM voters utilize one or more of the "value" stats that have been developed which take into account playing time, era, offense, defense, baserunning, etc. Not to mention that since we now have a full year between votes there is plenty of time to look at many players in great detail (many different stats and approaches).

@143: "I could do something similar by position if people are interested."

Yes please Kiko!

Okay, had some other stuff to do so I apologize that this took a couple days. I just appended it to my original link from #143 - repeated here for convenience.

If you scroll past the first table - HOMers by Year - the next small table calculates the total mix by position (based on pWins earned at the position). I show total numbers for all players, HOMers, and HOFers. The latter include Derek Jeter and Larry Walker (the HOM also includes this year's electees). Catchers have the fewest pWins of any position - as I discussed on the last page, I'm not doing anything with framing or game-calling and catchers don't really handle that many balls-in-play, for example. Anyway, the percentage for HOM catchers (4.8% of all pWins) is a little lower than for total catchers (5.3%). Pitchers are the most under-represented - which I think we've discussed before: ballpark, pitchers generate 1/3 of pWins but are 1/4 of HOMers. First basemen and corner outfielders are the most over-represented in the HOM - although that may be, in part, because HOMers tend to have long careers and strong enough bats that they end up there late in their career (e.g., Ernie Banks, Andre Dawson, etc.).

Anyway, those are aggregates. Below that is by year. The numbers here are the % in the year divided by the aggregate HOM %. So, a number less than 1 means fewer HOMers than in other years; a number greater than 1 means more HOMers. I lumped DH w/ 1B because I wasn't sure how else to deal with the fact that the position didn't exist before 1973.

The year-to-year numbers here can be volatile. And, in fact, there are seasons - and stretches of seasons - where we're just flat-out missing a position (not all of which are problematic).

We appear to have no HOMers who played catcher in MLB from 1918 - 1921 (hi, Wally Schang!) (although we might have Negro League catchers that cover these years). We appear to have no HOMers who played first base in 1943 or catcher in 1944-45 - neither of which seem like a problem to me; I think we all know why we might be missing those.

We have no HOMers who played shortstop at all in 1964-65, 1969, and 1973, and no HOMers who played shortstop regularly from 1962 (when Ernie Banks moved to 1B) to 1974 (when Robin Yount debuted). My personal Hall-of-Merit fills this gap with Bert Campaneris, Toby Harrah, and Dave Concepcion, all of whom I would suggest others might want to consider. Offsetting this, our HOM third basemen seem to be heavily concentrated between 1958 and 1986: worth thinking about whether we want to elect any more third basemen from this era (which, to be fair, Toby Harrah was a regular third baseman for part of this time period).

I'll leave the rest to people to draw their own interpretations.

If people would like, I can smooth some of this out by doing 5-year averages or something, but I was curious about the individual years. Oh, and the table goes through 2014 but it certainly stops being informative by at least 2007 or so, if not before that (the % of HOM games played by SS in 2014 was ridiculously high - because the only regular we've elected so far from that year was Derek Jeter).

Thanks for the response, Bleed.

@152: "Standard deviations impact wRC+ though, so using at face value will hurt/help different leagues and eras."

That's precisely what I'm seeking to avoid.

Also:

* Doesn't wRC+ rely on wRAA - and if so, won't it treat players differently depending on number of league games played?

* Does data to calculate wRC+ exist uniformly thgouout baseball history, including pre-1900 ball and the Negro Leagues?

"We have no HOMers who played shortstop at all in 1964-65, 1969, and 1973, and no HOMers who played shortstop regularly from 1962 (when Ernie Banks moved to 1B) to 1974 (when Robin Yount debuted). My personal Hall-of-Merit fills this gap with Bert Campaneris, Toby Harrah, and Dave Concepcion, all of whom I would suggest others might want to consider."

All good names to consider. I've also kicked the tires on Jim Fregosi and Luis Aparicio and they're in my 16-40 range at present (as is Campaneris). And Harrah is on my radar screen at 3B.

wRC+ is a rate stat, so it doesn't treat players differently based on games played. The components to calculate it should exist for most or all of MLB history. You can check for yourself here https://library.fangraphs.com/offense/wrc/

And standard deviations impact every stat that isn't specifically expressed in terms of standard deviations. So keep that in mind.

I made some tweaks to my calculation of wWAR (which is the main metric I use as a guidepost for HOM rankings), to smooth what used to be drastic swings at hard WAR cutoff points. While most players saw a small increase in their score, there were a few players who made big moves up or down, and it will affect my ballot for 2021. The biggest move of importance to me is that Buerhle is not going to make my ballot, and in fact he got moved from my 11th spot down to #32, so it doesn't look great for him to make a future ballot either. This makes me a bit sad as a Buerhle fan, but overall I think the new version of the metric is fairer. The other major move is that I can't deny Sosa a ballot spot anymore, regardless of how much un-clutch, etc. is in there. He jumps from a shaky back of ballot spot to #5. Bobby Bonds also moved on-ballot. Other players shuffled their order a bit as well, and a few guys made major jumps to a just-off-ballot spot who used to be further down. In parentheses is how much their rank changed with the new calculation.

1. Kenny Lofton (0)
2. Sal Bando (+1)
3. Buddy Bell (+2)
4. Bobby Abreu (+2)
5. Sammy Sosa (+9)
6. Tommy John (-4)
7. Kevin Appier (0)
8. Johan Santana (0)
9. Andy Pettitte (0)
10. Lance Berkman (-6)
11. Bob Johnson (-1)
12. Eddie Cicotte (+3)
13. Bobby Bonds (+4)
14. Roy Oswalt (-2)
15. Joe Tinker (-2)

John Olerud, Dwight Gooden, Sam McDowell, Wilbur Wood, and Jason Giambi moved up considerably and are now among those next in line for ballot spots, along with Vic Willis, Jeff Kent, Robin Ventura, and Nomar Garciaparra, who were already in this range before. Ron Cey is among those who took a bit of a tumble, and Tim Hudson took a large drop, from ~30 to #58.

Other (not top 50) players with big moves:
Cliff Lee (+) into top 50
Norm Cash (+)
Vida Blue (+)
Gene Tenace (+)
Jack Chesbro (+)
Dave Parker (+)

Mark Langston (-) out of top 50
Jorge Posada (-) out of top 50
Art Fletcher (-) out of top 50
Orel Hershiser (-)
Wally Berger (-)
Wally Schang (-)

FYI that the NLDB just added the 1932 Negro Southern League, and along with it a number of games among Indy and EWL teams. Once I finish updating the affect players (there are many), I'll post the latest and greatest on any of them who have gotten votes recently.

This update means that we now have continuous summer-league data for all the NNL/ECL/EWL/NSL/NAL/NNL2 seasons, 1920 to 1948. This in addition to independent play prior to it, winter leagues, and some Mexican League play. It's an amazing achievement, and Gary Ashwill, Kevin Johnson (aka: KJOK around here), Dan Hirsch, and Dan Lynch deserve special commendation for what they've done. And what they are continuing to do because they are planning more. But this is a pretty big moment for them.

Anyway, just wanted to let folks know about it and that revised MLEs will follow. I'd guess they will arrive either before Valentines Day or after Presidents' Day Weekend (b/c I'll be traveling in between them).

@159: "the NLDB just added the 1932 Negro Southern League, and along with it a number of games among Indy and EWL teams...Gary Ashwill, Kevin Johnson (aka: KJOK around here), Dan Hirsch, and Dan Lynch deserve special commendation for what they've done."

I've been waiting for this announcement for months. CONGRATULATIONS to the team at Seamheads!

On the same topic, I thought it might be useful to review the Hall of Merit’s roster of Negro League inductees prior to the 2020 election, and compare it with the Hall of Fame’s inductees.

Please bear in mind that this list includes all HoF and HoM players who appeared in the Negro Leagues for any duration, and lists players at their principal position in the Negro Leagues, irrespective of later position changes in the Majors.

CATCHERS: HoF 4, HoM 5
In the HoM only: Quincy Trouppe

FIRST BASEMEN: HoF 3, HoM 2
In the HoF only: Ben Taylor

SECOND BASEMEN: HoF and HoM, both 1

THIRD BASEMEN: HoF 3, HoM 2
In the HoF only: Judy Johnson (who I agree was a HoF mistake)

SHORSTOPS: HoF 6, HoM 8
In the HoM only: Home Run Johnson, Dick Lundy, Dobie Moore
In the HoF only: Sol White (probably more as the author of History of Colored Base Ball)

LEFT FIELDERS: HoF 1, HoM 0
In the HoF only: Cumberland Posey (principally as a manager)

CENTER FIELDERS: HoF 9, HoM 10
In the HoM only: Alejandro Oms

RIGHT FIELDERS: HoM and HoM, both 0
(The Hammer played short in the NgL.)

PITCHERS: HoF 11, HoM 8
In the HoF only: Andy Cooper, Leon Day, Hilton Smith

Total Negro Leaguers in both Halls: HoF 38, HoM 36

In sum, it strikes me that the HoM:

• Has generally done a damn fine job with Negro League candidates – better than the Hall of Fame in my view.

• Has honored far less Negro League first-, second- and third-basemen than shortstops – possibly because Negro League managers tended to put their best infielders at short. (This is a potential argument for the induction of Ben Taylor.)

• Has yet to honor a left- or right-fielder - and again, the worthiest candidates may well have been concentrated in center by Negro League managers.

• Has far too few Negro League pitchers. However severe the shortage of pitchers in the HoM is considered to be by the electorate, it seems the Negro Leagues are even more under-represented. (From my own point of view, I think Hilton Smith deserves a closer look.)

Updated the Negro Leaguer's MLE career lines workbook at my site. You can download and enjoy here.

In addition to adding the 1926 NNL and 1932 updates, I updated virtually every pitcher. The reason was that I had failed to notice 8-12 seasons in which the STDEV of the league's RAA/IP was exorbitantly high (more than 2.0). In these seasons, this occurs due to the presence of one or two players who give up a s******d of runs in 1 or fewer innings. I removed those players, allowing those leagues to return to a more normal level of variation. This helps most pitchers out because the higher the STDEV, the harder it is create separation from the league. It doesn't materially change our impressions of pitchers, the rankings stay similar, and the players don't gain more than a few WAR (usually less), but it does help a few guys out who need it.

Among players who got votes in the last two HOM elections:
Ben Taylor's MLE did not change.
Don Newcombe's MLE did not change.
Bus Clarkson's MLE did not change.
Luke Easter's MLE did not change.
Ellie Howard's MLE did not change.
Silvio Garcia's MLE did not change.
Carlos Moran's MLE did not change.

Now in closer a few guys.

Hilton Smith's MLE now looks like this thanks to data from the 1932 Monroe Monarchs of the NSL:

HILTON SMITH
MLE ESTIMATE INTO NL
           PITCHING              HITTING  
YEAR AGE   IP RAA   WAA  pWAR    PA hWAR   WAR
==============================================
1932  25  180   5   0.5   2.3    60  0.4   2.5
1933  26  210 - 3  -0.3   1.8    70  0.4   2.1
1934  27  200 -15  -1.4   0.7    67  0.4   0.9
1935  28  220   4   0.4   2.7    73  0.4   1.6
1936  29  270  12   1.2   4.0    90  0.6   2.6
1937  30  260  17   1.8   4.5    87  0.6   3.1
1938  31  260  13   1.4   4.0    87  0.6   4.7
1939  32  260   7   0.7   3.4    87  0.6   4.0
1940  33  270  43   4.9   7.6    90  0.6   8.2
1941  34  210  18   2.0   4.1    70  0.5   4.6
1942  35  180 -21  -2.2  -0.4    60  0.4   0.0
1943  36  160  25   3.0   4.5    53  0.3   4.8
1944  37   20   2   0.3   0.5     7  0.0   0.5
1945  38  180  11   1.1   2.9    60  0.3   3.3
1946  39  180   2   0.2   2.0    60  0.4   2.4
1947  40  180 -15  -1.5   0.4    60  0.4   0.7
1948  41  180   9   1.0   2.8    60  0.4   3.0
----------------------------------------------
TOTAL    3420 115  13.1  47.8  1141  7.3  55.1

Smith is hampered by the seasons 1932-1936:
1927-1928: Starts playing town leagues (per Riley)
1929-1930: Plays at a two year college and begins pitching
1931: Plays semipro ball
1932: 5.67 IP for Monroe of the NSL
1933: Playing in black minors, no stats
1934: Ditto
1935: 2 IP for Bismark
1936: 2 IP for KC
The MLE includes full-season credit for those seasons, but it doesn't love him during that time. I don't foresee us getting any more data from these seasons for him unless there's more to the 1935 and 1936 seasons out there. But I don't think they are priorities of Gary/KJOK at this moment. At least I haven't heard that from them during our various exchanges. Anyway, Smith's probably a little underserved by the MLE, but the fact is that his career started very late. So did Buck Leonard's and Jud Wilson's, and I have not given them any credit seasons prior to their rookie years (at ages 25 and 26 respectively).

Here's Heavy Johnson.

HEAVY JOHNSON
MLE ESTIMATE INTO NL
             
YEAR AGE   PA  BAT  RUN  FLD   WAA   WAR  WAR162
================================================
1916  21  160    9   0     0   1.3   2.0   2.1
1917  22  200   11   0     0   1.6   2.5   2.6
1918  23  300   16   0     0   2.3   3.6   4.6
1919  24  270   14   0     0   2.0   3.1   3.6
1920  25  490   42   0     0   5.1   7.1   7.4
1921  26  500   44   0   - 1   3.7   5.5   5.8
1922  27  530   47   0   - 1   3.9   5.6   5.9
1923  28  610   41   0   - 1   3.3   5.3   5.6
1924  29  560   44   0   - 1   3.8   5.8   6.1
1925  30  560  - 1   0   - 1  -0.8   1.1   1.1
1926  31  510   17   0   - 1   1.1   2.9   3.1
1927  32  530   26   0   - 1   2.0   3.9   4.1
1928  33  490   13   0   - 1   0.7   2.4   2.6
1929  34  440   10   0   - 1   0.5   1.9   1.9
1930  35  300   10   0   - 1   0.6   1.5   1.6
1931  36   25    1   0     0   0.1   0.2   0.2
1932  37  200    9   0     0   0.6   1.3   1.4
------------------------------------------------
TOTAL    6675  351   0   -11  31.7  55.7  59.8

The last couple updates have either been nonproductive or have decreased Johnson's MLE. He's still an outstanding hitter, and if you squint he's sorta like John Beckwith in that he's a little low on PAs, pretty high on Rbat, and isn't a good glove. Actually Beckwith is an AWFUL glove. And they both started off as catchers to boot. The difference between them is pretty simple: Beckwith was an awful infielder while Johnson was an iffy outfielder. The positional value is why Beckwith comes out a little better overall. The implication here is that Johnson's worth just about 40 WAR in his best seven years combined with about 60 WAR for his career. In Chaleekosville, that's awfully similar to Vlad G (40/60), Bobby Abreu (38/60), and Reggie Smith (38/61) among RFs and Jose Cruz (39/61) and Indian Bob Johnson (39/61) among LFs. I would note that all of these guys are on or adjacent to my in/out line. I would also note that I have less confidence in a subset of player seasons than I do in the career figures in the MLE. That said, there's good reason for caution with Johnson: His career data is very sketchy.
1916-1919: Wreckers, no data.
1920: Wreckers and NNL: 10 PA
1921: Wreckers, no data.
1922-1928: Full seasons of data.
1929: No data
1930: 13 PA
1931: 21 PA
1932: 5 PA

Obviously, missing about half a guy's career doesn't make me confident. I'm not entirely sure what happened to Heavy in his late thirties. The record and Riley's account are skimpy, but he didn't play much at all in any of the big leagues.

Now, Hurley McNair.

HURLEY MCNAIR
MLE ESTIMATE INTO NL
             
YEAR AGE   PA  BAT  RUN  FLD   WAA   WAR  WAR162
================================================
1910  21  460   23   0     1   2.1   3.7   3.9
1911  22  530   24   0     1   2.1   3.8   4.0
1912  23  600   28   0     1   2.3   4.2   4.4
1913  24  590   31   0     0   3.0   5.1   5.3
1914  25  590   28   0     0   2.8   5.0   5.2
1915  26  590   28   0     1   2.7   5.0   5.2
1916  27  580   24   0     0   2.5   4.8   5.1
1917  28  540   20   0     1   1.9   4.0   4.2
1918  29  480   13   0     3   1.3   3.2   3.3
1919  30  530   17   0     3   1.7   3.7   3.9
1920  31  630   20   0     3   1.9   4.1   4.4
1921  32  640   30   0     3   2.8   4.8   5.1
1922  33  570   35   0     3   3.1   4.8   5.1
1923  34  560   11   0     3   0.8   2.6   2.7
1924  35  550   35   0     1   3.1   4.9   5.2
1925  36  490   11   0     1   0.7   2.2   2.3
1926  37  410    6   0     1   0.3   1.7   1.8
1927  38  370    3   0     1   0.0   1.2   1.3
1928  39  300    0   0     2  -0.1   0.8   0.9
------------------------------------------------
TOTAL   10010  385   2    31  34.8  69.6  73.2

The MLE implies a seven-year peak of about 35 WAR against 73 career WAR. This is where we really see the tendency of the MLE process to level things out a bit across a career. A 35/73 peak/career combination is basically Jake Beckley with an extra five career WAR. Or Eddie Murray with four fewer peak WAR and six extra career WAR. It's Kina like Ozzie Smith (37/71) or Orator Jim O'Rourke (33/67). He's evidently a career candidate. But I don't think working with individual seasons is necessarily a great use of an MLE. There's practically no way a 35/73 guy would happen in real-life baseball. In fact, nothing comes all that close to it.

Lastly, I'm not going to run all the numbers on Sam Bankhead. I just wanted to point out that he hasn't changed much, and that he remains a little problematic because while he's sporting a high estimated career total
a) He's only got 30 Rbat, and batting is thing we can measure the best and on a season-by-season basis
b) He's got an unusually high 42 Rbaser, which we can only really measure at the career level and then dole out in proportion to PAs, but which has a lot of comping and educated-guess work in it
c) His fielding is a high 78 Rfield, which is also very inexact, and which is based on only 203 games at SS across eight noncontinuous seasons
d) He gets about half his RAA from positional runs (150), which are a little inexact in that they are not calculated as if he played all of his games at the primary position of a given season.

I'd worry about over extrapolating from the less rigorous running and batting information if I were considering a vote for him.

If you'd like to see any other player's articulated season-by-season MLEs, please let me know.

Dr C - it would be awesome if I could have access to the yearly MLEs instead of just the overall summaries. They have been invaluable for MMP voting.

That would be awesome. The problem is that I've got a hundred or two MLEs each in their own Excel file. It's easier if you just tell me what guy(s) you want to see.

Dr. C - are there updates for Conrado Marrero, Lazaro Salazar, Marvin Williams, George Scales, or Carl Glass (still very low completeness from last update)?

To answer my own question, Marrero is virtually unchanged and still looks like the best unelected pitcher to me. Carl Glass took an ENORMOUS hit and now is going to need an even better revision in the other direction on new data that comes in. That's the danger of looking at guys with lots of missing data, he will clearly drop out of my rankings (which go to about ~300) until significant positive data comes in. George Scales with a minimal upward tick. Marvin Williams with a minimal downward tick. Lazaro Salazr with a minimal upward tick.

I guess the moral is I second Dr. C's caution about players with copious missing data. The newest updates look like I will move McNair farther ahead of Johnson as well; I had them very very close together.

It's easier if you just tell me what guy(s) you want to see.

The best ones. That's a flippant answer but it is really what I'm interested in. Taking all of your excel files and spitting out the top 20 position players and 15 pitchers sounds like a job for a Python script.

When it comes to Python, you might as well call me Monte.

But I can, indeed, do something more small scale like what you’ve suggested without much ado.

If at any time you want some other player’s MLE just ask me in a ballot discussion thread. I’m not involved in MMP and don’t read those threads so I would never see the request.

If you could start with 1924 that would be great. Check the 1923 MMP discussion and see what I have been doing.

Another option would be to import the excel files into an Access database.

So, bottom line for NGL players not in the HoM — it’s not unreasonable to continue thinking Ben Taylor is the best available candidate, and perhaps the last good option left? Because that’s my take, at least.

bachslunch - I would disagree with that take. I don't see how to get Ben Taylor ahead of Allen, Williams, Bankhead, or McNair no matter how you slice it. I also put Johnson and Scales ahead, but I guess you could at least make a case for Taylor ahead of them. Carlos Moran is superior if you take the MLE as a 3B. That is only among position players. Salazar is also essentially a tie just as a position player, and then he's a pitcher as well and would move ahead of Taylor very easily if you consider him as a two-way player based on the MLEs. And then there are pitchers Marrero, Welmaker, McDonald, and Holland who are at worst difficult to separate from Taylor based on the MLEs.

I don't see Ben Taylor as standing out from the pack. Or really any Negro League players, to be honest. I think with Redding's election we've inducted anyone above borderline level. And I'm more hesitant to support a NGL player whose estimates are only borderline due to the uncertainty.

I think with Redding's election we've inducted anyone above borderline level.

I'm okay with that statement, though we are electing players on the borderline at the moment.

DL,

I'm not QUITE giving you want you want. Instead I've created a workbook that shares the yearly MLEs for about 50 players. There are tabs for each position, and each player receives a "card" that expresses his MLE WAR info in a very BBREF-like presentation. It's now live here.

There are extensive notes on both the batting and pitching MLEs that will guide understanding of where the weakness and strengths are in the numbers themselves.

I hope this is helpful and meets your needs in some fashion.

Regarding 172 and 173: I concur with both of your comments. The only player that might stand out is Newt Allen, but I won't believe in his defensive numbers until we have A LOT more data. He's got ungodly good Rfield, but there's not enough of a sample for me to say with confidence that he's the Brooks Robinson/Keith Hernandez/Roberto Clemente/Andruw Jones/Ozzie Smith of second basemen. Maybe someday. For now, I'd say keep the idea in your back pocket, but don't act on it.

BBREF just added retrosheet PBP data back to 1918. The prewar era is very complete now, and back toward the early 1920s feels pretty darned complete as well. I'd reckon 1918 and 1919 to be sparse for some teams, so it's not perfect, but it's pretty great to have this info now. It allows us to go deeper back into some very interesting careers. For now, this is an update of a chart I ran in the 2020 discussion thread that shows the runs I'm currently estimating the player to gain/lose once BBREF updates its WAR calculations to include this huge backlog of PBP data that's accumulating. The figures are in runs. Also, for DRA fans, there's estimates of the rOF the throwers will earn and how that compares to their DRA arm estimates (which are not very useful). I have updated all those players. I have not yet added any additional players, though I have plans to do so. Whether I get to them.... As always, if my math looks funny, it's because it probably is.

          EST  BBREF EST  EST  EST  EST  DRA  NET    NET
NAME      BSR  BSR   DPR  LFR  CFR  RFR  OFR  NODRA  DRA
=========================================================
BANCROFT    7   -8     5                      +20
BARTELL    -25  -5     8                      -12   
BERGER      7   -1     5   -0    7    1   -9  +13    +30
BISHOP     -8   -4    11                      + 7
BOTTOMLEY  -9    1    20                      +10
CAMILLI     4   -2    18                      +24 
CHAPMAN    18    5    -4    6    9   -7   18  + 9    - 1
CLIFT      29    0    12   +0              0  +41    +41      
COMBS      20   -1    26   -1  -10    0  -35  +47    +71 
W COOPER   -4    0    -8                      -12
CRAMER     10  -10    14    0   -6   -1   -5  +34    +32 
CUYLER     80   21    17    2   -1   17    9  +76    +94
D DIMAGGIO 20    0    -0   +0   29   +1   20  +20    +30  
ELLIOTT     1   -5   -10   -0   -4   -7   -0  - 4    -15
R FERRELL  -4   -2   -10                      -12
FREY       26   -2    18   -0   -0    1   +0  +44    +45
FURILLO    -3    0    -2   +0    9    7   14  - 5    - 3
GALAN      22   -1    33    5    5   +0   -4  +56    +70 
HAFEY      20   -3    11   14   -2    4   -0  +28    +44
HENRICH    15    6    24   -1    1   16    6  +45    +55  
BA HERMAN -13   -3     4   -1        -1  -21  - 6    +13  
HODGES     -0    0    -0                      - 0 
HOLMES      4   -1     6   -0    8   -7    3  +11    + 9 
T JACKSON   6   -3   -16                      - 7
B JOHNSON   2   -3   -10   25    4   -0   40  - 5    -10
JUDGE      -2    1     3              0    0  - 1    - 1
KELL        5    1    -8                      - 4 
H KELLY   - 2    2   - 1    1    0   -0   -0  - 1    - 1
KLEIN      13   -3    27   -3   +0   14   23  +43    +31 
LAZZERI   -23    5    -9                      -37  
LINDSTROM  17   -7     2   -1    5   -4    2  +26    +24
LOLLAR     -1    0    -1                      - 2
LOMBARDI  -23    5   -52                      -80 
MANUSH     25    6    35  -16   -1   -3  -34  +54    +68 
MARANVILLE  6    5    15                      +16
MARION      3   -1     5                      + 9    
MARTIN     25    5    11    1    4    3    4  +31    +35 
MYER        9   -5    24                      +34
NICHOLSON   1   -1    19    0    0  -11  -10  +21    +20
PESKY      17    9    16                      +24
S RICE     43   12    23   -1    4    6   12  +54    +51 
RIZZUTO    12    6     1                      + 7
ROLFE      22    5    29                      +46 
SCHALK    - 8    6    -8                      -22
SCHANG    -14    4    -1                      -19 
SCHNDIENST 12    3    -1    2   -0         1  + 8    + 9
STANKY      2   -1     4                      + 7
STEPHENS    5    1     1    1             +0  + 5    + 6
TRAYNOR    45   -2     2                      +49 
D WALKER    6   -2    27    2    3   15   -1  +31    +50 
L WANER    12   -4    16   -0   -5   -1   -1  +32    +27 
K WILLIAMS  5   -6    10  - 9    0   -0   16  +21    - 4  
H WILSON    1    1    -2   -1   -6    2  -10  - 2    - 7  
---------------------------------------------------------
APPLING    27    1     6                      +32
ASHBURN     0    0     0    1    4        +0    0    + 5
AVERILL    15   -3    42    2  -36   -2  -11  +60    +35
BERRA       0    0     0    0  - 0        -1    0    - 1 
BOUDREAU    1   -3    -7                      - 3 
CAREY     139   89    18   -0  +19   +3   22  +68    +68 
COBB       25   23    23   -0  -16   -1    7  +25    +15 
COCHRANE   14    2    33                      +45
E COLLINS  98   39    34                      +93
CRONIN     -7   -4    -2                      - 5
DICKEY    -19    1     3                      -15 
J DIMAGGIO 33    7     9    3   35    1    3  +35    +71 
DOBY        0    0              -3    2   -7    0    + 6
DOERR      -1  -12    -4                      + 7 
FOXX        8   -6    -5   -0    0   +0   -0  + 9    + 9
FRISCH     70   31    11                      +50 
GEHRIG      0  -13    50   +0        -0    0  +63    +63 
GEHRINGER  55   11    27                      +71
GREENBERG  -5    7    -0   +0              2  -12    -14 
GORDON     13    2    -3                      + 8
GOSLIN      7   11    22   12    2    2    6  +26    +36
HACK        2   -9    17                      +28
HARTNETT  -25    7   -14                      -46  
HEILMANN  - 7   -1   -13    0    0  -10  -13  -19    -16 
BI HERMAN   9   -4    -8                      + 5
HORNSBY    -2   -9   -12             -1   -0  - 5    - 6 
KELLER     -7    2    23   -3         1  -17  +14    +29   
KINER      -3    0    -0  - 9   -3        -9  - 3    - 6
MAYS        0    0     0         5         8    0    - 3
MEDWICK     3    0   - 7   15    1   -1    6  - 4    + 5
MINOSO      0    0     0   -0   -0   -0   -5    0    + 5
MIZE       14    2    15             -1    0  +27    +26
MUSIAL     15   -3    11    1   -1    6   -3  +29    +38
OTT        29   -4    57    1   +0   16   11  +90    +96 
REESE       6   -2    10                      +18
J ROBINSON  6    1     6                      +11
ROUSH     -20   -9     0   -0   -2    0    2  -11    -15  
RUTH       17  -12    19  -12         7  -23  +48    +66 
J SEWELL   13   -2    13                      +28
A SIMMONS  23    2   -11   15    7   -0  -11  +12    +45
SISLER     57   13    30                      +77
SLAUGHTER  15   -3    16    2   -1    7   -2  +34    +44
SNIDER     -1    0     1   +0   -3   -1    2    0    - 4
SPEAKER    77    2    40        57    0   56 +115   +116  
TERRY      13   -4    15                      +32
VAUGHAN    15   -9    37   -0             +0  +61    +61
P WANER    27   -8    22    1   -0   -8   -7  +57    +57
WHEAT      36   -6    24  -18    0    0  -25  +54    +61
T WILLIAMS  5    1     7    6        -1    1  +11    +17

Note: Only includes value accrued through 1948 (running/dp) or 1952 (OF throwing).

TL;DR? Here's the guys you probably need to know about:
Speaker: +116
Ott: +98
Cuyler: +94
E Collins: +93
Sisler: +77
Combs: +71
DiMaggio: +71
Gehringer: +71
Galan: +70
Manush: +68
Carey: +68
Ruth: +66
Gehrig: +63
Wheat: +61
Vaughan: +61
+++++++++++
B Johnson: -10
Bartell: -12
Cooper: -12
Ferrell: -12
Greenberg: -14
Elliott: -15
Dickey: -15
Roush: -15
Heilmann: -16
Schang: -19
Schalk: -22
Lazzeri: -37
Lombardi: -80

And I thought I was being friendly to Kiki Cuyler with my adjustments, but those numbers would safely put him just behind Lance Berkman for second on my ballot this year, and that's without crediting him for 1927. I hope I'm not the only one voting for him again.

OK, so let me run through the Cuyler calculations for everyone so that if he emerges as being possibly on-ballot for you, you'll know exactly how I'm deriving these numbers. It's a good thing I did because I found an error where a cell was pointed to the wrong numerator. It caused Cuyler to lose about 14 runs of base running, so he's probably not atop anyone's list anymore. But I'll go through this exercise anyway so that you know that I'm not that big a dope.

First off, here's broad-based look at how his my estimate of his value is changed by additional data.

RUNNING: BBREF's estimate of rBaser is 21 runs. I am estimating 61 runs.
DP AVOIDING: BBREF does not estimate this before 1948, but based on current information, I estimate him with 17 rDP.
THROWING: DRA estimates him with 8.5 ARM runs. BBREF does not estimate rOF prior to 1952. I estimate him at 18 runs.
TOTAL vs BBREF: +40 in rBaser and +17 in rDP.
TOTAL vs DRA ARM: 9.5 runs.

Now, let me get into the nitty gritty just a bit so that you can see exactly where these estimates come from. I do not endorse them as absolutely precise, but you must judge the degree of precision for yourself.
RUNNING
Running is split into five parts: Stolen-base runs (rSB), Pickoff runs (rPO), Outs-on-base runs (rOOB), Bases-taken runs (rBT), and Extra-bases-taken runs (rXBT). I'll explain the differences between the last two as we go along. Here is how I calculate Cuyler in each of these facets of running. I'll use his 1930 season as an example because it appears to be a crazy good season.

rSB
Using Extrapolated Runs' weights, I multiply every stolen base of Cuyler's by 0.18 runs. I multiply every caught stealing by -0.32 runs. I add these together. Then I use the league's SB% to figure its SB and CS in Cuyler's attempts. I multiply its SB by 0.18 and its CS by -0.32 and add them. Finally, I subtract the league from Cuyler to get his rSB.
Cuyler attempted 46 steals.
He stole 37 bases: 6.7 runs
He was caught 9 times: -2.9 runs
Total of: 3.6 runs
The league had a 59% SBpct, therefore, in Cuyler's attempts
it stole 27 bases: 4.9 runs
it was caught 19 times: -6.1 runs
Total: -1.2 runs
Cuyler's 3.6 runs - the league's -1.2 runs = 4.8 rSB

rPO
I weight every pickoff like a caught stealing at -0.32 runs. I levy that against Cuyler's POs, then I figure the league's pickoffs in Cuyler's stolen-base opportunities (sbOPP) and apply the same penalty to it. Then I subtract the league from Cuyler.
Cuyler was picked off 4 times: -1.3 runs
Cuyler had 262 sbOPP.
The league was picked off 166 times in 32693 sbOPP, or 0.5% of the time.
In Cuyler's 262 sbOPP that would be 1.3 POs, which calculates to -.4 runs
Cuyler's -1.3 runs minus the league's -0.4 runs = -0.9 runs for Cuyler.

rOOB
I similarly weight every out on base (OOB) like a caught stealing at -0.32 runs. For the league, we need to calculate Cuyler's opportunities (and the league's), which we do by summing the number of times Cuyler was on first base when a single was hit, the number of times he was on first when a double was hit, and the number of times he was on second base when a single was hit. These are the three situations in which a player most often has to decide whether to attempt to take an extra base, though they are surely not the only times that he does so. This is close enough for me. YMMV.
Cuyler made 9 OOB: -2.9 runs
Cuyler had 32 opportunities while on first base when a single was hit; he had 11 opportunities while on first when a double was hit; and he had 42 opportunities while at second when a single was hit. That's a total of 85 opportunities to make an OOB.
The league made 1297 OOB.
The league had 5176 opportunities while on first when a single was hit; 1337 opportunities while at first when a double was hit; and 3021 opportunities at second when a single was hit. That's a total of 9534 opportunities.
The league made an OOB 13.6% of the time.
In Cuyler's 85 opportunities that's 11.6 OOBs or -3.7 runs.
Cuyler's -2.9 runs minus the league's -3.7 runs is 0.8 runs in Cuyler's favor.

rBT
This is runs accrued from bases taken on fly balls, passed balls/wild pitches, balks, and defensive indifference. I treat assign BTs the same weight as a steal, 0.18 runs. We figure the league's rate by dividing its BT into its sbOPPs.
Cuyler took 30 bases in 1930 (which is an unusually high figure, in fact, it's the highest I recall seeing other than the 32 he took in 1926): 5.4 runs.
The league took 2064 bases in 32693 sbOPPs, 6.3% of the time.
In Cuyler's 262 sbOPPs that's 16.5 BT: 3.0 runs
Cuyler's 5.4 runs minus the league's 3.0 runs = 2.4 runs in Cuyler's favor.

rXBT
This is runs accrued from bases taken on batted balls. It's assumed that the runner will take any base he is forced to, which means first to second on a single, second to third on a single, and first to third on a double. Many moons ago, I did some calculating using averages of base-out states given a runner's starting base, and I came up with these values for the XBT itself:
first to third on a single: 0.19 runs
first to home on a double: 0.38 runs
second to home on a single: 0.43 runs.
So we apply that to the instances in which Cuyler did so.
First to third on a single: Cuyler went first to third on a single 21 times in his 32 opportunities: 4.0 runs
First to home on a double: Cuyler scored on a double 10 times in 11 opportunities: 3.8 runs
Second to home on a single: Cuyler scored 35 times in 42 opportunities: 15.1 runs
Total of 22.9 XBT runs for Cuyler.
The league went first to third 2146 times in 5176 opportunities, or 41.5% of the time
In Cuyler's opportunities, the league went first to third 13.3 times: 2.5 runs
The league went first to home on a double 525 times in 1337 opportunities, or 39.3% of the time.
In Cuyler's opportunities, the league scored 4.3 times on a double from first: 1.6 runs
The league scored from second on a single 2223 times in 3021 opportunities, or 73.6% of the time.
In Cuyler's opportunities, the league scored 30.9 times: 13.3 runs
Total of 17.4 runs for the league.
Cuyler's 22.9 runs minus the league's 17.4 runs = 5.5 rXBT

rBaser
So Cuyler's rBaser is the sum of those five components. In this case it's
4.8 rSB + -0.9 rPO + 0.8 rOOB + 2.4 rBT + 5.5 rXBT = 12.6 rBaser, which is ridiculously good.

So if we do this for every season of his career we get this

1921  -0
1922   0
1923   1
1924   9
1925  11
1926   9
1927   5
1928  10
1929  10
1930  13
1931  -2
1932   2
1933  -0
1934   1
1935  -1
1936   0
1937  -1 
1938  -0
========
TOT   66

We are currently in the process of reconstructing hundreds of 1930 games for which play-by-play descriptions are not available (and we are working backwards). We recently completed 1931 reconstructions but they have not yet been released. My guess is that 1930 and 1931 complete seasons will be released in June. KikoS undoubtedly has all the data on complete/incomplete seasons, possibly on his website or the Retrosheet website.

So I don't see how "the prewar era is very complete" unless I am not understanding what you mean.

Rob, this is an invaluable page on retrosheet data: https://baseball.tomthress.com/Articles/Seasons.php

Rob,

I don’t have the level of expertise you do on this, so I will defer to your judgment. As I went through the data, it appeared to me that totals for sbOPP were similarly robust from the very late 1920s onward. Most full-time players were accumulating about 200-300, depending on their own performance and the league’s run context. That was up from what I’d seen previously in general and looked comparable to the more settled seasons of 1946 and 1947 that had few if any changes.

Hey, everyone, the NLDB did some updates this week, which meant some updates to the MLEs. I’ve updated the career lines and the year lies (for select players), and you can view them at the usual place.

Shameless self-promotion: I've written a book Baseball Player Won-Lost Records: The 2019 Retrospective which is available as a free PDF at my website or for purchase at Amazon. See here for details. https://baseball.tomthress.com/Books/Books.php. The focus here is mostly 2019, so it's not necessarily hugely relevant here, but I threw some history in there, too - including a look at Vern Stephens.

Re: the conversation between Rob, Bleed, and Dr. C. above, if you're interested in seeing exactly how many games Retrosheet is missing (for seasons for which it has released partial play-by-play) for a specific player, you can go to that player's page on my website and there should be an option "Player records with missing games extrapolated". If you click that, the number of games shown in the first column of the first table will be the player's actual games played. So compare the games played with and without games extrapolated and that's how many games Retrosheet is missing. For example, here's Kiki Cuyler without extrapolating missing games (1,767 career games) vs. Kiki Cuyler with missing games extrapolated (1,878 career games - so, we're missing 111 career games for Cuyler, all prior to 1932).

183. Kiko Sakata Posted: March 05, 2020 at 10:52 AM (#5928297)
Shameless self-promotion: I've written a book Baseball Player Won-Lost Records: The 2019 Retrospective which is available as a free PDF at my website or for purchase at Amazon. See here for details. https://baseball.tomthress.com/Books/Books.php. The focus here is mostly 2019, so it's not necessarily hugely relevant here, but I threw some history in there, too - including a look at Vern Stephens.

Re: the conversation between Rob, Bleed, and Dr. C. above, if you're interested in seeing exactly how many games Retrosheet is missing (for seasons for which it has released partial play-by-play) for a specific player, you can go to that player's page on my website and there should be an option "Player records with missing games extrapolated". If you click that, the number of games shown in the first column of the first table will be the player's actual games played. So compare the games played with and without games extrapolated and that's how many games Retrosheet is missing. For example, here's Kiki Cuyler without extrapolating missing games (1,767 career games) vs. Kiki Cuyler with missing games extrapolated (1,878 career games - so, we're missing 111 career games for Cuyler, all prior to 1932).

Thanks Kiko, pdf downloaded, Cuyler and Stephens are in my pHOM and it's largely to be thanked by your research, Vern wouldn't stand a chance otherwise, and Cuyler is fringey from a heavy 20s/30s era.

At a high level, here are the number of games missing play-by-play data by year (I got these from Kiko's website):

1918 383
1919 490
1920 297
1921 261
1922 259
1923 239
1924 221
1925 227
1926 217
1927 194
1928 220
1929 289
1930 216
1931 184

I don't like to make predictions, but I think the current pace is completing the reconstructions of two seasons every six months (semi-annual Retrosheet release).

I don't like to make predictions, but I think the current pace is completing the reconstructions of two seasons every six months (semi-annual Retrosheet release).

I just have to say that I have an unabashed man-crush on Retrosheet and always will.

In fact, I have three big man-crushes:
Retrosheet
BBREF
Negro Leagues Database.

My life is better for all of them.

The 2020 Baseball Reference WAR update is live.

https://www.sports-reference.com/blog/2020/03/2020-war-update/

Biggest change is to Ernie Lombardi, but there are others impacted to varying degrees.

Really good estimates by Dr.Chaleeko, changes tend to track with what he worked up previously.

Great tip Esteban, thanks for sharing, and be safe all, excited to incorporate these updates :)

Looking over the changes it looks like there's another adjustment that was made and not mentioned in the notes. Players active in the late 1880s and early 1890s had some revisions in their WAR numbers, which may be due to adjustments to the seasons surrounding the three league year of 1890 and the end of the AA.

Yes, very exciting update. Esteban, I was noticing some subtle differences too, and not just around the time you mentioned. Worth noting that this update does not include rOF (outfield throwing) for 1931-1952.

I'll be posting some analysis of my estimates on my site (The Hall of Miller and Eric) on Monday morning. The key question will be whether you'd want to use my numbers or BBREF's previously published numbers for rBaser and rDP if you were electing your own Hall or working on the HOM. Has pertinence to pre 1931 estimates.

Hi, everyone,

In light of BBREF's WAR update, I'm updating my chart of rBaser and rDP and rOF estimates found in post 176. A few players have been removed (mercifully) because we no longer need estimates of any of their seasons, and they have no rOF value. In other words infielders and catchers. So what we've got is estimates of seasons prior to 1931 rBaser and rDP and for rOF prior to 1952. BBREF has not yet updated rOF. I'll note here also that BBREF updated some of their pre-1931 baserunning estimates. Just a couple runs here or there, but I noticed it. OK, here's the chart.

ESTIMATE OF PDP-BASED RUNS NOT CURRENTLY ACCOUNTED FOR IN BBREF WAR
rBaser and rDP estimates through 1930
rOF estimates through 1952
          EST  BBREF EST  EST  EST  EST  DRA  NET    NET
NAME      BSR  BSR   DPR  LFR  CFR  RFR  OFR  NODRA  DRA
=========================================================
BANCROFT    7   -3     5                      +15
BARTELL    -6    1     3                      - 4   
BERGER      2   -1     0   -0    7    1   -9  + 3    +20
BISHOP     -4   -8     5                      + 9
BOTTOMLEY  -5    3    14                      + 6
CHAPMAN     1    1     1    6    9   -7   18  + 1    - 9
CLIFT                      -0              0    0    - 0
COMBS      16    4    17   -1  -10    0  -35  +29    +53 
CRAMER      1    0     0    0   -6   -1   -5  + 1    - 2 
CUYLER     66   25    15    2   -1   17    9  +56    +83
D DIMAGGIO                  0   29    1   20    0    +10
ELLIOTT                    -0   -3   -7   -0    0    -10
R FERRELL  -3   -1    -1                      - 4
FREY                       -0   -0    1    0    0    + 1
FURILLO                    +0    9    7   14    0    + 2
GALAN                       5    5    0   -4    0    + 6
HAFEY      11    3     7   14   -2    4   -0  +15    +31
HENRICH                    -1    1   16    6    0    +10  
BA HERMAN -10   -6     3   -1        -1  -21  - 1    +22  
HOLMES                     -0    8   -7    3    0    - 2 
T JACKSON   5   -1    -3                      + 1
B JOHNSON                  25    4   -0   40    0    -11
JUDGE      -3   -1     3              0    0  + 1    + 1
H KELLY    -1    0    -6    1    0   -0   -0  - 7    - 6
KLEIN       0   -3     5   -3   +0   14   23   +8     -4 
LAZZERI    -6    1    -4                      -11  
LINDSTROM  16  -10     2   -1    5   -4    2  +28    +26
MANUSH     18    8    18  -16   -1   -3  -34  +28    +42 
MARANVILLE 13    6    17                      +24
MARTIN      0    0     0    1    4    3    4  + 4    + 4 
MYER        8   -4     6                      +18
NICHOLSON                  +0       -11  -10    0    - 1 
S RICE     38   14    20   -1    4    6   12  +44    +41 
SCHALK    - 8    4    -8                      -20
SCHANG    -10    3    -1                      -14 
SCHOENDIENST                2   -0         1    0    + 1
STEPHENS                    1              0    0    + 1
TRAYNOR    37    6     1                      +32 
D WALKER                    2    3   15   -1    0    +21 
L WANER    10   -2     5   -0   -5   -1   -1  +17    +12 
K WILLIAMS  5   -7    10  - 9    0   -0   16  +22    - 3  
H WILSON    3    2    -3   -1   -6    2  -10  - 2    - 7  
---------------------------------------------------------
APPLING     2    1     0                      + 1
ASHBURN                     1    4        +0    0    + 5
AVERILL     6   -1     9    2  -36   -2  -11  +16     -9
BERRA                       0  - 0        -1         - 1 
CAREY     139   94    18   -0  +19   +3   22  +64    +64 
COBB       25   56    23   -0  -16   -1    7  +25    -32 
COCHRANE   10    5    18                      +33
E COLLINS  97   39    34                      +93
CRONIN     -1   -1     1                      + 1 
DICKEY     -5    3     1                      - 7
J DIMAGGIO                  3   35    1    3    0    +36 
DOBY                            -3    2   -7    0    + 6
FOXX        6   -3     2   -0    0   +0   -0  +11    +11
FRISCH     66   28    13                      +51 
GEHRIG      7   -6    23   +0        -0    0  +36    +36 
GEHRINGER  16    7     6                      +16
GREENBERG                  +0              2    0    - 2 
GOSLIN      3    8    12   12    2    2    6  + 7    +17
HARTNETT  -13   -1    -4                      -16  
HEILMANN   -8   -2   -13    0    0  -10  -13  -19    -16 
HORNSBY     0   -8   -12             -1   -0  - 4    - 5 
KELLER                     -3         1  -17    0    +15  
KINER                      -9   -3        -9         - 3
MAYS                             5         8    0    - 3
MEDWICK                    15    1   -1    6         + 9
MINOSO                     -0   -0   -0   -5    0    + 5
MIZE                                 -1    0  +27    +26
MUSIAL                      1   -1    6   -3    0    + 9
OTT         1   -6    10    1   +0   16   11  +17    +23 
ROUSH     -19  -13     2   -0   -2    0    2  -12    -16  
RUTH       22  -11    17  -12         7  -23  +50    +68 
J SEWELL    3   -6    10                      +19
A SIMMONS   8    1    -2   15    7   -0  -11   +6    +39
SISLER     57   12    30                      +76
SLAUGHTER                   2   -1    7   -2    0    +10 
SNIDER                     +0   -3   -1    2    0    - 6
SPEAKER    77    2    40        57    0   56 +115   +116  
TERRY       2   -3     6                      +11
VAUGHN                     -0             +0    0      0 
P WANER    26   -3    12    1   -0   -8   -7  +41    +41
WHEAT      36   -5    24  -18    0    0  -25  +65    +72
T WILLIAMS                  6        -1    1    0    + 4

I've done some deep dives into two of the players on my list, Dolph Camilli and Chief Bender. I have put together a pretty substantial report on both, but I keep it to the highlights.

Dolph Camilli
- I tried to come up with a list of similar players to Camilli (strong offensive primes, short careers, 1B/LF/RF). My working list was Albert Belle, Bob Elliott, George Foster, Jason Giambi, Brian Giles, Matt Holliday, and Will Clark. Clark is the only HoMer, and none of the others seem likely to get elect in the near future. I do like Camilli the best of these eight, but it's marginal preference for Camilli's balanced profile.

-Camilli does appear to be a good situation hitter. We don't have full RE data for him, but he does do well with what we do have. I don't typically incorporate RE data for pre-1950 players, but since he is so close to ballot for me and his data is pretty complete, I'm okay with incorporating it.

-Camilli has a case for minor league credit. He spent 4.5 seasons in the PCL before breaking into the majors and he was pretty godo for that entire stretch. That being said, he does seem to be a late bloomer, as he didn't become an above average player until his third seasons as a starter. I decided that giving him two seasons of credit for showing up is the best I can do for him - he was probably a viable major league player a couple years prior to his debut, but it's unlikely he would have been any better than average.

-Camilli missed half of 1943 and all of 1944, but not due to WWII. He simply did not report to the Giants when he was traded there because he didn't like them. I cannot justify giving him credit for that, even hif he was likely still an above average player at that point.

Overall, I don’t think it’s likely that I am significantly underrating Camilli. If I give him a very minor bit of minor league credit and incorporate the RE data that we have, it moves Camilli up from 34th to 26th. Barring some sort of hidden data that really helps, I think that’s about as high up in my system I can rank Camilli. If I ever got around to systematically building a PHoM, he is probably right around the borderline.

Considering his era and position are both pretty well accounted for and the fact that he is the type of player that typically does very well in my system, I can’t endorse him all that much.

-----------------------
Chief Bender
-Before this deep dive, Bender ranked 107th among eligible players, which was just close enough to intrigue me.

-Bender was a strong postseason performer, and it also seems he was leveraged into important pennant race games. Connie Mack believed him to be the most clutch pitcher he ever had and a cursory look shows he did pitch well down the stretch. I gave him a very modest bump for this.

-Bender's career was short on the back end for a combination of reasons. His ineffectiveness in the Federal League seemed to lessen his reputation around the league, although he was effective in 1917 with the Phillies. He served in wartime industries in 1918 and then went to the minors to pitch/manage. I think it's likely he wasn't afforded an opportunity to manage in the majors on account of his race - he was regarded as being highly intellegent by many in the game, but despite his interest in managing, he was never given the opportunity in the majors. He pitched decently well in the minors through 1921, but I can't imagine he would have been better than league average in the majors, and he likely wouldn't have been pitching a full load. He was never a workhorse and his managing duties would likely have limited him to no more than 100-150 innings a season.

I decided to give him ~300 innings of league average pitching for the 1918-1921 period. Had he been white and WWI never occured, I think he likely would have pitched in the majors for that portion of his career. He pitched 800 innings in the minors in that period, and none of those in 1918, so that's a conservative total, but it is more in line with what he had been doing in his last two years in the majors, and accounts for him likely being a player-manager and being less inclined to pitch heavier loads. He was very effective in 1919 and 1920, but it was in the Eastern League, so not exactly top level competition.

Overall, I don't think Bender was HoM level talent. He simply was great in his prime, but not brilliant, and lacks longevity, both within individual seasons and for his whole career. His career was probably cut short due to a combination of WWI, racism, and his own decline and desire to manage. He does move up significantly in my rankings, but only up to 72nd, which, needless to say, is well off ballot. Barring a major information dump that is unlikely to happen, he will not be a candidate for my vote.

I've got a few short comments, and would be getting in very late on some discussions, so I'll just list some of them here:

1) - Dr. Chaleeko, I like your work, but your charts in comments 176 and 191 have NOT EVEN ONE column heading that I understand. I could probably guess some of them, but not all. Could you do me the favor of just telling me what the column headings mean?

2) About Ben Taylor - There's been a lot of talk about Taylor getting some sort of "best man still standing" support because his era has so few 1B in the Hall. Well, so does 3B during the same period. The best 3B between Jimmy Collins and Pie Traynor is - well, Home Run Baker, but he's so atypical for a 3B at the time that I guess you could make a case for Heinie Groh. There's a couple of reasons for this. a) Both 3B and 1B were, during the Dead Ball Era, treated primarily as defensive positions, largely because of all the bunting. b) Hitting is 3 times as valuable as fielding (baseball is, essentially, half hitting, 1/3 pitching and 1/6 fielding. So, you've got a lot of hot gloves with not-so-hot bats. In this context, the candidacies of Ben Taylor (and Ed Konetchy) might actually be overrated, and the lack of players from those spots in the DBA should be seen as the simple result of the conditions of the time, and a warning flag when you see it.

3) My two cents' worth is that NO system should be used that is LESS sophisticated in its stat categories than the New Historical Abstract, published in 2001. Career accumulation, 3- year non-consecutive peak, 5-year consecutive prime, rate per 162 games. I believe in the timeline, but not everyone does. To be honest, I have no idea why the proponents of any other system do not do this. I mean, how hard can it be? You could take BB-Ref WAR and list its accumulation (adjusted by harmonic mean, using a number close to 8.33 instead of Bill's 25), the three best years, ignoring consecutive, the 5 best consecutive years, and the WAR per 162 games. This would be trivial to do. The only hard part - and the only part that keeps me from spending a week tor so doing this - is the lack of a complete database. BB-Ref has one. I do not. But if they gave access to their database, I'd have those numbers and rankings done in a week. About 3 days to write the program, and a couple more for the thing to run in background. Why don't they do this? Why don't those of you who have individual systems - WAA, WPA, etc. do this? I probably sound old-fashioned because I keep going back to the New Historical, but no one has improved on its system, in 20 years. What can I do?

4) I came up with a new method at the last minute last year. I haven't given up on it, but I now have it under pretty good control. My main problem is lack of Win Shares stats for players in the last 20 years. I really like this method for two big reasons - 1) it's not based on a system that I wrote, and 2) It doesn't involve looking at one player compared to another, as players catch your fancy. It produces a list of the most probable candidates, using a method that is not mine (New Historical, based on Win Shares). I spent years worrying whether my constant placement of Lou Brock and Dizzy Dean were just me being a Cardinals fan. Then the method put both of them in the Consideration Set, and I quit worrying. Brock, and especially Dean, come out as among the absolute highest-ranking players, and it's not my bias towards Cardinals. It's my willingness to consider peak, prime and rate when considering Dizzy Dean. You know, if you use pretty much ANY system of career accumulation, you won't just lose Dean. You will also lose Sandy Koufax. There is not enough difference between the two (14 Win Shares, career), and Koufax is not right at the in/out line of the Hall. But you will lose him, as well as Dean.

5) I have GRAVE doubts about assigning minor league credit, because there are SO many players who would deserve it. Take Rabbit Maranville. In 1925, the Cubs moved him to the Cardinals, who put him in the minors. In 1926, they started Tommy Thevenow at short and left Rabbit in the minors. In 1927, Thevenow turning out to not be even the hitter Rabbit was, the Cards tried out 19-year-old Heinie Shuble, with Rabbit in the minors. In 1928, they hired Bill McKechnie, the manager who most cared about fielding in the history of the game. He looked at Thevenow. He looked at Shuble. He brought Rabbit Maranville back from the minors, restarting his major league career. So, for about 2 1/2 years, Rabbit was in the minors, waiting for the kid candidates to not turn out to be Hall candidates. So, you want a couple of years at the start of someone's career, where he was backed up behind Ken Keltner or somebody? OK. I want 2 1/2 years minor league credit for Rabbit. McKechnie buried Hank Sauer's bat in the minor leagues, because Sauer had a lousy glove. I want those years back for Sauer, and there are like 7 of those guys at Sauer's level, much less missing "only" three years. That's why I don't do things like that. There are just too many candidates. War. Negro Leagues. Those are the only two I pay any serious attention to. I'm not going to go reevaluate the 300 or so players who were trapped in the minors because the team's management didn't know what it was doing. Julian Javier was buried behind Bill Mazeroski for a few years. It's a huge list of these guys.

6) Electing the Baker Bowl to the Hall, in the players' wing, is absurd. And that is exactly what would happen if Gavvy Cravath got elected. His candidacy isn't him. It's the Baker Bowl.

5) I have GRAVE doubts about assigning minor league credit, because there are SO many players who would deserve it.

A lot of players may deserve minor league credit, but for most guys it's a relatively... minor thing.

For Maranville, his stint in Rochester was well into his decline phase. He wasn't missing prime seasons. He's not in my consideration set and a couple of league average-ish seasons aren't moving him much.

There are maybe half a dozen guys who move up significantly with minor league credit - Lefty O'Doul, Babe Adams, Gavvy Cravath, Luke Easter, and maybe a couple others. Most other players get a season or two that may fill out their career. It's probably not going to be enough to push a guy onto your ballot unless he was already close.

Personally, I don't worry about a little smidgen of minor league credit for guys outside my consideration set, which is currently about 250 guys. Maybe 20 of those guys get minor league credit, and maybe four are anywhere close to ballot territory. And I think I'm being pretty thorough.

The new numbers will definitely impact my ballot. Ernie Lombardi had made my preliminary ballot at number 13, but this adjustment will drop him off — and in fact may drop him off my list of top 6 catchers. At present, my guess is that Urban Shocker will reach the number 15 slot, and Chet Lemon will probably make the vacated spot in my top 40.

My top six catchers will now likely be: Schang, Munson, Posada, Porter, Tenace, Kendall. It’ll be a tussle between Kendall and Lombardi for the last slot.

bjhanke, you seem a little cranky today. I hope everything is going OK with your sheltering and social distancing. It's pretty anxiety producing!

1)...Could you do me the favor of just telling me what the column headings mean?
Sure thing.
EST BBREF EST EST EST EST DRA NET NET
NAME BSR BSR DPR LFR CFR RFR OFR NODRA DRA
NAME = Name ;)
EST BSR = My estimated baserunning runs
BBREF BSR = BBREF's rBaser (their estimate of baserunning runs)
EST DPR = My estimated double-play avoidance runs (BBREF doesn't do any such estimations for years they don't have complete data for, so theirs is always 0)
EST LFR = My estimated throwing runs in LF
EST CFR = My estimated throwing runs in CF
EST RFR = My estimated throwing runs in RF
DRA OFR = DRA's estimate of throwing runs across the entire outfield
NET NO DRA = ( EST BSR - BBREF BSR ) + EST DPR
NET DRA = ( EST BSR - BBREF BSR ) + EST DPR + ( (EST LFR + EST CFR + EST RFR) - DRA OFR)

To put those last two into plain English, NET NO DRA is the gain/loss in runs for running and GIDPs compared to BBREF's estimates. NET DRA also adds the gain/loss of outfield throwing runs compared to DRA.

2) I also do not support Ben Taylor. I think tops out at the borderline, and I think we have several eligible players who exceed it. One's mileage may vary, however, and we can't close the book on Taylor yet because we may not have his complete career data at this time. There may be winter league data yet to come. I don't know if that's the case, I'm only saying it is possible.

3)...But if they gave access to their database, I'd have those numbers and rankings done in a week.
I have great news for you! Click through for the complete BBREF WAR database for hitters and pitchers.

6) This is a question of great interest, obviously. Outside the Baker Bowl, Cravath's line was 273/366/419. That's a .781 OPS. The NL from 1912-1920 had an away OPS of .632, which is considerably worse than Cravath's away OPS. I looked up players from 1912-1920 within a 1,000 PA of Cravath's 1912-1920 total (4,239) and within 20 points of his away OPS and got this list:
Bobby Veach: .797 and 175 batting runs in 5059 PA
Heinie Groh: .768 and 162 batting runs in 4803 PA
Zack Wheat: .786 and 161 batting runs in 4952 PA
Edd Roush: .792 and 136 batting runs in 3461 PA
Sherry Magee: .775 and 123 batting runs in 3936 PA

So as an away hitter, Cravath was as good a hitter during this stretch as four HOMers and a should-be HOMer (IMO) were altogether.

Based on the league's splits during this period, we would have expected Cravath to be an .871 OPS hitter at home. Who was Ann .871 OPS hitter during this say 1912-1920 time period? There were only two.
Gavvy Cravath: .871 and 223 batting runs in 4239 PA
George Sisler: .874 OPS and 192 batting runs in 3282 PA.

If I widen out just a little (ten points of OPS), we add Eddie Collins: .851 OPS, 323 batting runs, 5789 PA. Well, that's more PA than the 1,000+/- Cravath I was shooting for, but you get the idea.

So if Cravath had a merely normal home/road split, he'd be among the best non-Cobb/Speaker hitters of the era. I don't think he's a creation of Baker Bowl. I think it's very probably he's a highly skilled player who changed his game at home, and who on the road showed the talent and skills of a high-performing player, the likes of which we have elected before.

As promised, here' a look at how my running and GIDPing estimates came out in light of BBREF's new numbers. It's expressed as whether you'd have been better off using BBREF's old numbers or my numbers in evaluating a player.

Hey Dr. C -

How difficult is it for you too add more players to the yearly MLE spreadsheet found here?

The two I'd be most interested in would be Willard Brown and Roosevelt Davis.

There are a half dozen other guys that I'd like to see, but if it takes more than like 45 seconds, it's not a big issue:
Andy Cooper
Leon Day
Luke Easter
Sam Jethroe
Judy Johnson
Spottswood Poles

Thanks!

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 often favoring 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. Am currently treating 19th century pitchers pretty much equally as post-1900, but 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. 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.
2. Buddy Bell. Best WAR at 3B. Have decided to trust the metric for him.
3. Wally Schang. Among best C WAR, also hit well.
4. Jeff Kent. Best WAR at a middle infield position and hit well, can't in good conscience rank him below guys like Sosa or Johnson.
5. Kenny Lofton. Best available CF. Not as much hitting as I'd like, but lots of WAR at a premium position.
6. Bobby Abreu. Best WAR among available RFs, better than Sosa.
7. Bob Johnson. Best WAR among available LFs.
8. 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.
9. Vic Willis. Good pitcher WAR, best after McCormick.
10. Sammy Sosa. Better WAR than I remembered. Happy to give him some benefit of the doubt given his treatment by the BBWAA.
11. Vern Stephens. I value hitting at a premium position highly, so I'm ranking him here.
12. Tommy John. Pretty much all compiling, minimal peak. But racked up plenty of WAR this way. I'm on board.
13. Thurman Munson. Better all around than Lombardi, shorter career.
14. Sal Bando. Second best WAR at an under-represented position.
15. Mickey Welch. Next best pitcher WAR after John by a decent margin.


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

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

All required disclosure players are on ballot or within top 40. None of the newcomers make my ballot, and only Mark Buehrle (whom I see as similar to Andy Pettitte) and Tim Hudson (a slight bit lower than Buehrle) make my top 40.

Incorporates latest changes in BBRef WAR. Am expecting this will be my final ballot for 2021, barring any major BBRef WAR changes. Please treat it as such in case I'm not able to post it to the 2021 ballot thread.

Dr. Chaleeko - I wasn't cranky when I posted up before; I just had taken in a LOT of comments here, and was trying to respond to several of them at once, in as few words as possible. Now, as regards comment #196 - I generally like your work (you're one of the people whose name on a comment usually means quality work), and I REALLY appreciated your translating the column headings for me. I was in real trouble without that. But there are some problems with comment #196, mostly about Gavy Cravath.

1. I may not have been clear enough about having access to a database. I don't mean access to READING a database, or using a menu to construct one of the SQL queries that the site allows through its menu pages. I mean that I want to PROGRAM against the database, without limit. I programmed for money for 25 years. It's frustrating to want to write a program against a giant database, but to know full well that I will never have the privileges that I'd need. I don't blame the sites. If they turned me loose to program, I'd chew up huge chunks of processing power going through play-by-plays to get stuff I want. I could not possibly be worth it. So, I don't blame them. But sometimes, I have to say that the reason I don't have a piece or three of info that I'm commenting on is that none of the available database queries do that particular thing, and no sensible database owner would allow me to program against it.

2. If you're going to use OPS to defend what you believe is a serious HoM candidacy, don't cite the mean. The HoM is not in the business of inducting everyone above the mean. If it were, there would have to be thousands of candidates, and we'd have to induct 50 a year. The mean, in a Hall discussion, is just a distraction. I will concede at any time that Cravath was a noticeably better hitter than average.

3. Your list of other HoM candidates in #6 is far too weak. The best of them is Heinie Groh, and three of them have better numbers than Cravath. In other words, they are the absolute bottom of the in/out line. What we need for a good comparison is a list of all the HoM right fielders and their OPS numbers. I'm going to look up the list and render another comment here, but I won't get that done tonight.

4. The comparison to Sisler is another problem. Sisler a) didn't enter MLB until a few years after 1912, b) was a kid starting his career right out of college, rather than a veteran in his prime and short decline, and c) was a pitcher when he came up. That's not a good match for Cravath. And Kid Sisler has slightly better OPS numbers anyway.

5. Using Eddie Collins' numbers from 1912-1920 is using what? 40% of his career, balanced against Cravath's ENTIRE career? Even if you give Cravath 2 years MLE time for playing in Minneapolis, that's a very bad match.

6. Eddie Collins was a Gold Glove second baseman. Sisler was a Gold Glove 1b through 1922 (much worse when he tried to come back with double vision). At the time, 1B was a more important position than RF, where Cravath played. And Cravath has terrible defensive numbers in RF, the least-important defensive position of his day. If you're trying to make a HoM case for Cravath by comparing him to Sisler and Collins, defense is a big hurdle.

7. My problems with Cravath supporters have mostly to do with them freaking out over the homer crowns and the resultant black ink. Well, Gavy Cravath has the worst home/road HOMER splits I can find in the 20th century. 3.85-1. The SECOND-highest such split is Fred Luderus, at 3-1. Luderus was the lefty power hitter in the Baker Bowl in a career that is almost an exact match for Cravath's years. And he's not close to Cravath's split, despite being in the same ballpark. If Cravath supporters were to just ignore the black ink and focus on what Cravath could actually do, I wouldn't get so cranky about it, partially because the candidacy would lose a large portion of its power. I DO want to thank you for NOT getting into the black ink.