MLB Recognizes Negro Leagues

31 Replies

You may have read about MLB’s decision to officially recognize several Negro Leagues as major leagues for the 1920 to 1948 seasons. So, now the debate begins about interpreting the statistics for those players, teams and seasons.

More after the jump.

Probably the most complete source of Negro League statistics can be found at seamheads.com, including WAR metrics derived from the basic stats. Over at FiveThirtyEight.com, those WAR metrics were used in a snappy little article as the principal basis of comparison between Negro League players and those who played in what were formerly the only major leagues. I commend the article (which includes interactive stats presented with impressive and insightful graphics) to your attention. If you’re like me and know very little about the Negro Leagues, it’s a great way to be introduced to some of their star players.

Thanks to Bob Eno for suggesting this post, and for bringing the FiveThirtyEight article to my attention.

0 0 votes

Article Rating

31 Comments

Oldest

Newest Most Voted

Inline Feedbacks

View all comments

Bob Eno (epm)

3 years ago

Thanks to Doug for the shout out (and for alerting me to the post). I think I’ll have pushed up a field of daisies before sabermetrics gets its arms around integrating Negro League stats with MLB’s, but I’m sure the effort will be increasingly interesting as initial attempts to set parameters become more nuanced through competing analyses. It’s a project so much more constructive than, say, trying to work on comparability measures in light of PEDs. I wonder if a subgroup of sabermetricians will make a specialty of this project and give the rest of us more tools to work… Read more »

Doug

3 years ago

Reply to Bob Eno (epm)

Had there been a more “enthusiastic” reception by major league teams to integration, there might be a reasonable body of data to work with in determining how to normalize negro league and major league stats, using the statistics of players with careers straddling that period. Sadly, the slow pace of integration by many major league teams really leaves us with relatively few such examples, making the task of normalization all the more difficult.

Last edited 3 years ago by Doug

Paul E

3 years ago

Reply to Doug

Doug,
Normalization is an impossibility. Stats are sketchy….rosters were thin. Even if we all agreed that Satchel Paige was the greatest of all time regardless of skin pigmentation, from a statistical perspective, are his “normalized” numbers (W-L, ERA, ERA+, SO/BB, etc….) more like Grove, Feller, Johnson, Spahn, Seaver, or Maddux?

Mike L

3 years ago

Reply to Paul E

Maybe the more universal good of righting a historic wrong is more important than trying to make fine gradations on stats.

Paul E

3 years ago

Reply to Mike L

MLB and corporate America are in the same boat and (IMO) merely looking to avoid boycotts and cancel culture. I seriously doubt either’s sincerity in any of this recent “onboarding”

Mike L

3 years ago

Reply to Paul E

I’m not really sure what Manfred’s intent here was, but there’s no doubt in my mind that segregation was a moral wrong. Personally, I’d be more focussed on correcting it going forward than trying to create an equivalence in the past through gestures.

Paul E

3 years ago

Reply to Mike L

Mike L
I’m sure that as individual US citizens we all agree that segregation is morally wrong, however, I have reservations about corporate America and MLB being concerned about anything beyond the bottom line. Maybe I’m naive (or half-blind) but I believe the majority of segregation nowadays in the US suburbs and urban areas is due to monetary/economic differences and not flat-out racism. (Sorry, I can’t call them inequities).

Mike L

3 years ago

Reply to Paul E

Specifically, I was talking about segregation in baseball and not housing. As to the rest of it, I’d repeat what I said above, which is that gestures about the past are just that. Sports is a business, but it’s also a niche business which markets its past as a core part of its DNA–places like the Hall of Fame are examples. MLB is making a business decision as much as a “moral” one to do this.

Mike L

3 years ago

Reply to Bob Eno (epm)

Bob, would you be concerned that suggesting a “correction” rubric might actually support an adverse inference–try too hard to normalize, and the entire project looks too synthetic and diminishes all?

Bob Eno (epm)

3 years ago

Reply to Mike L

Hi Mike – Well, first off, I hadn’t meant to argue that integrating the stats was a good thing, just that it was going to happen and that I was sure it would be interesting to see how the process is handled by people with skills in baseball history and stats. Paul says above that normalization is an impossibility, but I think what he meant might have been that it can’t be done well. I don’t have any doubt that there will be multiple competing models for normalization. I expect that ones more complete in considering the changing historical details… Read more »

no statistician but

3 years ago

Since Bob Eno has made a brief reappearance, I’ll chime in, too. I’ve been dreading the raising of this topic here ever since its appearance on the national scene. Why? I don’t think anyone can honestly claim to be able to interpret the stats available at Seamheads 1) in any way approaching an impartial manner as far as comparing is concerned; 2) in any way that can possibly allow for the irregularities of scheduling, season length, availability of reliable box scores or any box scores at all; 3) in any way that accounts for the fluctuations of roster players and… Read more »

Bob Eno (epm)

3 years ago

Reply to no statistician but

It’s been a long time since we were on the same string, nsb. I was dismayed when you announced you didn’t plan to continue posting comments, and your comment here illustrates why: as always, your points are all good. But one of your valid points is that while we can see the difficulties of integrating these incommensurable stats, it’s inevitable that attempts to do so will be coming our way. As I wrote in response to Mike’s comment, I think that after a while we’ll start to become interested in how clumsy initial attempts are superseded by increasingly nuanced analyses,… Read more »

bells

3 years ago

Reply to no statistician but

I find this to be a fascinating and complicated topic as well, and I’m compelled to respond to the thoughts you’ve laid out here nsb (no offense taken if you limit your comeback to one comment and respond no further, was nice to read your thoughts). I see two strains of critique in what you’re saying – first, the ‘it’s not the band I hate, it’s their fans’ critique of the integration of these statistics and the effect it will have on arguments. I, too, dislike the possibility that someone will interpret statistics with too much simplicity and lack of… Read more »

Bob Eno (epm)

3 years ago

Reply to bells

I dropped back in just in case any of my earlier comments had generated a reply that I ought to acknowledge, and although there were none, I’ve been treated to one of the most thoughtful and provocative posts I’ve ever seen on HHS. I think what bells is doing is suggesting that when we play with statistics, we’re capable of elevating our level of play and, in fact, routinely do so. The function of simple stats, like leader boards, isn’t the end of play: they’re something we play with when we add the high heat of interpretation, which is really… Read more »

Doug

3 years ago

Under the heading of For What It’s Worth, these are the only position players I could find with 100 MLB games and 100 Negro League games up to 1948. Only two of the eight players had much more than a major league season’s worth of games in the Negro Leagues.

Mike L

3 years ago

Reply to Doug

Doug, that table reminds me of how old I am getting. Striking to see Jim Gilliam, who I saw play in the mid 1960’s when the Dodgers were constantly in the World Series or on Game of the Week, was a Negro League player in 1946-48.

Doug

3 years ago

Reply to Mike L

Gilliam’s stat line for 1946-48 is instructive, with 105 OPS+ yielding 6 WAR in fewer than 600 PA. He must have been a hell of a defender, or (more likely) there was a huge drop-off between average and replacement level players.

Mike L

3 years ago

Reply to Doug

Thinned out rosters after WWII was over?

Doug

3 years ago

Reply to Mike L

By way of comparison, Randy Johnson (2.9 WAR, 1982-84) has the most career WAR with less than 600 PA and 105 OPS+ or lower. For single seasons, Mark Belanger (6.5 WAR, 1976) and Lance Johnson (6.1 WAR, 1993) have the most WAR with the same criteria, and 15 other players have 5 WAR or more in such a season. But, only one of those (Willie Randolph, 1976) was under 25 when he did it, whereas Gilliam was a teenager.

Last edited 3 years ago by Doug

Doug

3 years ago

Reply to Mike L

Some random factoids on these eight players. Doby – first player (1952) to lead his league in SO and Runs, with 100+ of each Irvin – last NL player (1951) to make post-season debut with 4 hits in a World Series game Thompson – last NL player (1950) with two IPHR in a game Jethroe – only modern era player with 100 R and 35 SB in each of first two seasons Minoso – last player (1954) to lead his league in TB with fewer than both 20 HR and 30 doubles Campanella – first catcher with multiple seasons of… Read more »

Last edited 3 years ago by Doug

Richard Chester

3 years ago

Reply to Doug

It appears that Yankees’ GM, George Weiss, was the main reason they were slow to integrate. Yankees’ owners Dan Topping and Del Webb let him run the club while they remained in the background. Weiss had a cold-hearted personality. Curiously Topping also owned a team of the same name in the All-America Football Conference from 1946-1949 and he had no qualms about African-American Buddy Young being on the team.

Paul E

3 years ago

Reply to Richard Chester

During Weiss’ 13 full years as the Yankees’ general manager from October 1947 to October 1960, the team won ten AL pennants and seven more World Series titles, compiling a regular-season record of 1,243–756 (.622 W-L %). Maybe he figured. “if it ain’t broke, don’t fix it”? I dunno

Buddy Young could fly 🙂

Paul E

3 years ago

Reply to Doug

Doug,
Is this a lack of statistical information or did these guys really only play 40 games per season in the Negro Leagues? If the latter, they must have played a ton of ‘exhibition’ games or their skill sets couldn’t have improved much.
Re Gilliam, by “debut” season, I imagine you don’t necessarily mean “rookie” year. Richie Allen led the NL in triples with 13 in 1964 – his ‘rookie’ year. He had 24 PA’s in 1963…?

Doug

3 years ago

Reply to Paul E

Yes, I meant first season, not rookie year. As to Negro league seasons, Seamheads says their database is built from the ground up, so all of their season totals are based on contemporary box scores and game accounts. None of their totals are based on season totals that might be published elsewhere. So, presumably, those totals (on Seamheads) will increase as time and resources permit. That said, I really don’t know how many games Negro league teams typically played; I would imagine most games were on weekends to accommodate players on visiting teams who had, or wanted to have, other… Read more »

Last edited 3 years ago by Doug

Author

Doug

3 years ago

For those who may be interested, I have updated a post from 5 years ago on batteries with 200 game starts together. You can find the revised article here.

Richard Chester

3 years ago

Reply to Doug

Not on your list is Dazzy Vance/Hank DeBerry with 246 starts together. Vance started 349 games so the percentage is 70.5%. Check me out.

Doug

3 years ago

Reply to Richard Chester

Thanks, I will check it out.

DeBerry’s 1926 season is notable. His .383 SLG that year is the highest without a HR of any player scoring in less than 15% of 40+ TOB (w/o ROE).

Last edited 3 years ago by Doug

Doug

3 years ago

Reply to Richard Chester

You got it exactly right, Richard. 246 starts.

But, it wasn’t 70.5%. For the seasons they were active, Vance started only 275 games. So, DeBerry was his catcher in 89.5% of Vance’s starts, the biggest number of the group, And, Vance was DeBerry’s pitcher in 48.6% of DeBerry’s starts. That is the ultimate personal catcher.

Richard Chester

3 years ago

Reply to Doug

My spreadsheet shows 37920 starting pitcher/starting catcher combinations, AL, NL and FL from 1901-2019.

Author

Doug

3 years ago

Reply to Richard Chester

So, that was the only one I missed?

Richard Chester

3 years ago

Reply to Doug

37919/37920= 99.997%, not bad at all. Actually 97920, due to missing data, is a fraction of 1% on the low side.