Baseball for the Thinking Fan

Login | Register | Feedback

btf_logo
You are here > Home > Baseball Newsstand > Discussion
Baseball Primer Newsblog
— The Best News Links from the Baseball Newsstand

Wednesday, August 30, 2006

Sabermetric Research: Birnbaum: The color bar: discounting Babe Ruth’s accomplishments

Sudder points out the latest from Phil Birnbaum…

Morally, of course, there’s a big difference between segregation by race and segregation by league. But the question is not a moral one; it’s an empirical one. If we don’t discount Mike Schmidt’s career because he didn’t have to face American Leaguer Jim Palmer, why do we discount Hank Greenberg’s career because he didn’t have to face Negro Leaguer Satchel Paige?

If you consider the Negro Leagues like a third major league, equal in talent to the AL and NL, the parallel is exact. It doesn’t matter which pitchers are in which leagues, so long as the talent level is about the same. Schmidt didn’t have to face Palmer, but, because of that, he had to face Tom Seaver that much more. And since Greenberg didn’t have to face Paige, he had to face Lefty Grove a few more times. As long as the leagues are even, the segregation is irrelevant.

Repoz Posted: August 30, 2006 at 10:29 PM | 100 comment(s) Login to Bookmark
  Tags: negro leagues, sabermetrics

Reader Comments and Retorts

Go to end of page

Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.

   1. AndrewJ Posted: August 30, 2006 at 11:44 PM (#2162161)
But the Negro Leagues were almost certainly worse than the majors. The highest percentage ever of major league players who were black was in 1974, when there were about one-third as many black players as non-black. But in the 30s, there were probably at least half as many Negro League players as white major leaguers (as far as I can quickly estimate). So the Negro Leagues were, proportionally, somewhat diluted.

Given that blacks make up about one-tenth the population, that sounds right.

But.

Throw in the dozens of whites-only minor leagues in the 1930 and the absence of a Negro minor league system ... and now we have 200 or so Negro Leaguers versus, what, 400 major leaguers plus 3000-4000 white professional ballplayers? Isn't that a more accurate comparison?
   2. Matt Clement of Alexandria Posted: August 31, 2006 at 12:26 AM (#2162220)
I don't understand this post. I fear I might be unfair, but I've read through the argument five or six times now, and all I see is a whole bunch of language games and unearned, inexplicable moral superiority.

Can anyone explain the point here?
At first, it sounds pretty reasonable – there were many pitchers, better than those Ruth actually faced, who would have certainly been good enough to play in the American League, if not for segregation. But since Ruth didn’t have to face them, the pitchers he did face were a little worse, and his stats were inflated for that reason.

But by that logic, everyone’s stats are inflated. Sure, Ruth didn’t have to hit against black pitchers. But Rogers Hornsby never had to face Walter Johnson, arguably the best pitcher ever. That’s because, according to the rules at the time, Hornsby was segregated to the National League for all of Johnson’s career.
This is just cheap logic games. The point, obviously, is that the competition would be better if the leagues were integrated. Birnbaum wastes a whole lot of space devising bizarrely even-handed language to talk about segregation, but I don't see why I should care if the Negro Leagues are considered a "third league" or whether I should see the Cardinals as "segregated" from the Athletics.
Now, in fairness to those who make the Babe Ruth argument, they’re implicitly assuming that integration would lead to the elimination of the Negro Leagues, or their conversion into minor leagues. You’d pull Josh Gibson and Oscar Charleston into the majors, they would displace marginal major-leaguers, and the white stars would stay put. Under this scenario, yes, of course, integration would have made improved the major leagues. Dizzy Dean’s job would be harder, as he’d be pitching to better hitters overall.
Of course everyone "implicitly assumes" that integration leads to the end of the segregated all-black league. That's what actually happened. The only argument anyone has ever made in regards to the quality of the segregated white leagues was that integration leads to improved competition. Birnbaum never addresses the actual argument. He tosses out a bunch of speculation, and then concludes:
So compared to these, how important is the dropping of the color bar to the question of league improvement? It’s probably a drop in the bucket. Well, maybe it’s a cup in the bucket, or even a pitcher in the bucket – it would take a bit of research to know which. But in any case, to cite segregation as the only factor, or even the only important factor, while ignoring all the rest, doesn’t make sense. It attempts to answer the question by what feels good morally, rather than by a full accounting of the the evidence.
What full accounting? He hasn't done it. He's just rhetorically downplaying the effects of segregation without data, as much as these strawmen he's arguing against are "ignoring" all the other factors in league quality. The tone of moral superiority in this post, cleaved to a desire to downplay the effects of segregation of baseball, really rubs me the wrong way.
   3. Zach Posted: August 31, 2006 at 12:32 AM (#2162228)
One thing that the Bonds steroid years convinced me of is that a Ruthian talent will show through regardless of context. The numbers might be a little different in different eras or with different competition, but a player like that can't not stand out.
   4. Slivers of Maranville descends into chaos (SdeB) Posted: August 31, 2006 at 01:07 AM (#2162307)
Of course everyone "implicitly assumes" that integration leads to the end of the segregated all-black league. That's what actually happened. The only argument anyone has ever made in regards to the quality of the segregated white leagues was that integration leads to improved competition.

What Birnbaum is saying is that it's not integration per se that improved the level of competition, it's the contraction following integration that improved the level of competition. The same amount of talent spread among fewer teams. And that the true competition level is judged by (total talent/total ML roster space).

Seen in that light, we shouldn't divide the level of competition into pre-integration and post-integration, and say that anybody who played before that line should be discounted. Instead, integration should be seen as part of a continuum of events that affected either the talent level or the number of ML roster spots. Things like the closing of the American Association, the expansion of the Rockies and Devil Rays, the decline of black ball players since 1974, the increase in Latin players, etc. etc. So instead of one event, we have dozens, of which integration was an important part but not orders of magnitude above the others, and certainly not the only part.
   5. Slivers of Maranville descends into chaos (SdeB) Posted: August 31, 2006 at 01:09 AM (#2162311)
Of course everyone "implicitly assumes" that integration leads to the end of the segregated all-black league. That's what actually happened. The only argument anyone has ever made in regards to the quality of the segregated white leagues was that integration leads to improved competition.

What Birnbaum is saying is that it's not integration per se that improved the level of competition, it's the contraction following integration that improved the level of competition. The same amount of talent spread among fewer teams. And that the true competition level is judged by (total talent/total ML roster space).

Seen in that light, we shouldn't divide the level of competition into pre-integration and post-integration, and say that anybody who played before that line should be discounted. Instead, integration should be seen as part of a continuum of events that affected either the talent level or the number of ML roster spots. Things like the closing of the American Association, the expansion of the Rockies and Devil Rays, the decline of black ball players since 1974, the increase in Latin players, etc. etc. So instead of one event, we have dozens, of which integration was an important part but not orders of magnitude above the others, and certainly not the only part.
   6. Crispix Attacksel Rios Posted: August 31, 2006 at 01:13 AM (#2162327)
Throw in the dozens of whites-only minor leagues in the 1930 and the absence of a Negro minor league system ... and now we have 200 or so Negro Leaguers versus, what, 400 major leaguers plus 3000-4000 white professional ballplayers? Isn't that a more accurate comparison?

I think there were also a lot of black players in the Mexican league. Does anyone know what those numbers tended to be?
   7. Slinger Francisco Barrios (Dr. Memory) Posted: August 31, 2006 at 01:43 AM (#2162411)
If you consider the Negro Leagues like a third major league, equal in talent to the AL and NL, the parallel is exact. It doesn’t matter which pitchers are in which leagues, so long as the talent level is about the same. Schmidt didn’t have to face Palmer, but, because of that, he had to face Tom Seaver that much more. And since Greenberg didn’t have to face Paige, he had to face Lefty Grove a few more times. As long as the leagues are even, the segregation is irrelevant.

Seductive, but it doesn't wash. It's not Grove vs. Paige. It's, say, Jim Walkup or Harry Kelley vs. Paige.
   8. jwb Posted: August 31, 2006 at 02:05 AM (#2162484)
Jim Walkup: 462 IP, 260 BB
Ouch!
   9. Minus Ice Posted: August 31, 2006 at 03:42 AM (#2162709)
Matt Clement - Fantastic post. My sentiments exactly.
   10. Eraser-X is emphatically dominating teh site!!! Posted: August 31, 2006 at 03:59 AM (#2162729)
Yeah, I agree with the other respondents.

This articles tone is even more maddening when you consider that the Negro Leaguers had to endure NOT being treated like a third major league.

I mean, I understand his point and SdeB articulates it better--if next year MLB expands to 60 teams, it's doubtful that official records will discount statistics even though they will be against weaker competition.

But I do believe the actual moral ground is to include all of these changes into the analysis--like many members of the HoM work hard to do, not like the author of this piece who I feel is saying something more like, \"#### changes, people are idiots to point this out!"

At the very least, we owe it to the players and moreover, the great sport of baseball whose history has been perverted by the stain of segregation.
   11. ERROR---Jolly Old St. Nick Posted: August 31, 2006 at 04:13 AM (#2162749)
Matt's post is long enough to quash this moron completely, but not so long that it gives his argument more attention than it deserves. What's next on this man's agenda, a study arguing that Soviet jewelry didn't have it so bad?
   12. Shredder Posted: August 31, 2006 at 04:18 AM (#2162756)
Seen in that light, we shouldn't divide the level of competition into pre-integration and post-integration, and say that anybody who played before that line should be discounted.

Who's discounting them? Getting to the end of that piece, he seems to be making the point that there are a lot of factors that changed the game and the record books, and integration is overrated as one of those factors.

That may indeed be the case, but what he doesn't get into is why it's overrated.

But in any case, to cite segregation as the only factor, or even the only important factor, while ignoring all the rest, doesn’t make sense.


But it does make sense depending on the context of the discussion. Most of us here can have an intelligent discussion on how much integration changed the record book in comparison to expansion, or internationalization, or advances in technology and equipment. But those things are gradual, and if you're having this discussion with someone who's not a baseball geek, they may require more explanation. Integration is quick to explain, and easy to understand.

Of course, recently this argument has most likely popped up in discussions of the impact of steroids on the record book. If someone advances the argument that records set by guys on the juice are illegitimate, a counter-argument is that the game is so different in so many ways than it was 80 years ago that comparing counting stats across generations may already be a fool's errand. And integration is a quick and easy example for that counter-argument. In addition, specifically as it relates to Bonds, if someone says "Bonds should be stricken from the record book because of steroids", it's very easy and effective to respond "Well Bonds would never have had the chance to set the record if he'd played in Ruth's era."

So to sum up, I've never run across anyone who discounted pre-integration records. I would agree that integration's effect on the stat sheet may be overstated. But I'd argue that the reason it's cited more than the other examples he gives, despite their relative impacts, is because of the simplicity of making that argument and the effectiveness of getting one's point across in the process.
   13. Gonfalon Bubble Posted: August 31, 2006 at 04:21 AM (#2162762)
What's next on this man's agenda, a study arguing that Soviet jewelry didn't have it so bad?

Hey, Soviet jewelry really didn't have it so bad. Some of those Faberge eggs are gorgeous!
   14. Mister High Standards Posted: August 31, 2006 at 04:23 AM (#2162766)
Matt's post is long enough to quash this moron completely


This guy isn't a moron. He has done TONS... literally flipping tons of great research and is the editor of SABR's By the Numbers Newsletter. He is a really good reseacher... of course I don't really get his point - but he is far from a moron.
   15. ERROR---Jolly Old St. Nick Posted: August 31, 2006 at 05:02 AM (#2162786)
This guy isn't a moron. He has done TONS... literally flipping tons of great research and is the editor of SABR's By the Numbers Newsletter. He is a really good reseacher... of course I don't really get his point - but he is far from a moron.

I'll grant your point, but you sure as hell couldn't prove it by what he's written here. The most chartible thing one could say about him on the basis of this piece is that like many saberites, he suffers from the disease of reflexive contrarianism, which shows up completely in the chalk on the blackboard tone of the article. But if it's only a temporary affliction, I'd be glad to see what else he's written.

If someone advances the argument that records set by guys on the juice are illegitimate, a counter-argument is that the game is so different in so many ways than it was 80 years ago that comparing counting stats across generations may already be a fool's errand.

A more perfect non sequitur can scarcely be imagined. A moral point about steroids cannot be countered by a technical one about other changes in the game. And if the underlying counter-argument is really that Ruth's records are illegitimate because of segregation, that is another argument altogether, but either way you resolve it, it in no way legitimizes records gained in part by steroids. And BTW I'm not saying that Bonds's records should be "stricken from the books," only that in their totality they are bogus in terms of what they represent.
   16. Shredder Posted: August 31, 2006 at 05:11 AM (#2162788)
And BTW I'm not saying that Bonds's records should be "stricken from the books," only that in their totality they are bogus in terms of what they represent.

Well that's fine. I don't want to turn this into a debate on that point. It's been done to death. We agree that they shouldn't be stricken from the books. As to one's opinion that they're bogus, that's a value judgment that each person is free to make for themselves. And I'll leave it at that.
   17. Eraser-X is emphatically dominating teh site!!! Posted: August 31, 2006 at 05:14 AM (#2162791)


This guy isn't a moron. He has done TONS... literally flipping tons of great research and is the editor of SABR's By the Numbers Newsletter. He is a really good reseacher... of course I don't really get his point - but he is far from a moron.


We all are morons in certain areas and geniuses in others. Some folks are just better at not making brilliant "discoveries" and patting themselves on the back while they look like idiots because they lack the self-awareness to know when they are in an area that they suffer from general ignorance or a blindness of perspective.

Others are needlessly long-winded and write endless run-on sentences rather than just getting to the point of what they are trying to say, and in such a way, dilute their points, rendering it difficult for others to extricate their main points, even though in many cases they might actually agree with them, which is really a shame, in my humble and concise opinion.
   18. Shredder Posted: August 31, 2006 at 05:21 AM (#2162792)
A more perfect non sequitur can scarcely be imagined. A moral point about steroids cannot be countered by a technical one about other changes in the game.

And I'd agree that it's a poor counter to the argument that from a moral standpoint, juicers have a less valid claim to records than someone who played pre-integration.
   19. David Nieporent (now, with children) Posted: August 31, 2006 at 06:10 AM (#2162808)
I'll grant your point, but you sure as hell couldn't prove it by what he's written here. The most chartible thing one could say about him on the basis of this piece is that like many saberites, he suffers from the disease of reflexive contrarianism, which shows up completely in the chalk on the blackboard tone of the article. But if it's only a temporary affliction, I'd be glad to see what else he's written.
I see. So when people you disagree with agree with the majority view, it's "groupthink." When people you disagree with disagree with the majority view, it's "contrarianism." But of course when <u>you</u> disagree with the majority view, you're not "contrarian"; you're a brave dissenter from groupthink.
   20. Dag Nabbit at ExactlyAsOld.com Posted: August 31, 2006 at 06:31 AM (#2162812)
Nice post by MCoA at #2. That being said, I've met Phil and he's a good guy who generally does good work. This is the first time I've been really disappointed by anything he's done.

This guy isn't a moron. He has done TONS... literally flipping tons of great research and is the editor of SABR's By the Numbers Newsletter. He is a really good reseacher... of course I don't really get his point - but he is far from a moron.

To expand. Aside from being the editor of the BtN Newsletter, he's also - according to the head o SABR's stats committee, the guy who really does all the work for that committee. He's also involved with SABR's Baseball Research Journal. When a statistical article comes down the pike, he's the guy that's asked to review it and weigh in on its worthiness. He doesn't have the only say, or even final say, but he does play a notable role. He's also given presentations at a few SABR-Cons, and his work has inspired other studies.


This article is, alas, extremely disappointing. Let me focus on one sentence here:

If you consider the Negro Leagues like a third major league,

First off, no. It wasn't equal in play, and certainly not in pay. Nor in facilities, attendence, or organized structure. The original Negro Leagues collapsed in the Depression, survived partially through barnstorming, before being resurrected later in the 1930s.

equal in talent to the AL and NL

No. Heck, I sat in on a Negro Leagues committee meeting in Toronto and even the honchos there wouldn't say that.

There's only one reason to consider it to be a major league - there were players of major league caliber there, but not all of them, not even most of them, not in the same circumstances as white major leagues, and certainly not by choice. To draw an analogy, it would be like if Bud Selig announced today that Canadians are forever banned from MLB, and from now on the best Canadian players would have to play in the Northern League or something like that. Would that make the Northern League a major league? There would be MLB quality players in it, but it would hardly be a major league.

Long as I'm here:

But by that logic, everyone’s stats are inflated. Sure, Ruth didn’t have to hit against black pitchers. But Rogers Hornsby never had to face Walter Johnson, arguably the best pitcher ever. That’s because, according to the rules at the time, Hornsby was segregated to the National League for all of Johnson’s career.

Problem: Hornsby's Cards and Johnson's Senators were dipping into the same talent pool for players. Nothing prevented Hornsby from being picked up by an AL team nor Johnson by an NL team. Or Schmidt and Palmer, as he later uses as examples. The AL and NL operated as two leagues in terms of scheduling, but as one league when it came to getting talent. That is ever so incredibly not the case with the Negro Leagues.
   21. Starring RMc as Bradley Scotchman Posted: August 31, 2006 at 11:02 AM (#2162846)
Aw, geez, not this again...
   22. Phil Birnbaum Posted: August 31, 2006 at 11:55 AM (#2162851)
A few replies:

Comment #4 (from SdeB) summarizes my argument better than I ever would have been able to do it myself.

Shredder (#12) writes, “… if you're having this discussion with someone who's not a baseball geek, they may require more explanation. Integration is quick to explain, and easy to understand.”

Agreed. Integration made the majors stronger as a one-time effect, and it’s a good example of one of the events that changed the level of competition (making it stronger). My point is that the hypothetical conclusion “… therefore, the overall quality of play must have been weaker then” is not necessarily true. And my second point would be that now that blacks are a much smaller proportion of players, you have to make somewhat of the same argument about today’s play being weaker.

He adds, “I've never run across anyone who discounted pre-integration records.” I’ve read that in print many times. A couple of weeks ago I read someone who said that he could never consider Ruth the best player of all time because of the quality of play before segregation. I should have saved the link.

I agree with Eraser-X (#17) that I can be a moron on this point even though I may be a good researcher. But on #10, I don’t understand what “…when you consider the Negro Leaguers had to endure NOT being treated like a third major league” has to do with the question.
   23. Phil Birnbaum Posted: August 31, 2006 at 12:23 PM (#2162855)
Here’s a different argument that might make the same points better.

Suppose, inexplicably, that MLB decided to segregate. Noticing that about 10% of players are black, they decide that three teams will be all black – call that the new “National League”, and the other 27 teams “American League” will be all white. Will the quality of play change? If the overall caliber of the two groups of players is the same, the quality probably will not change. The exact same players have jobs.

Now suppose that MLB creates *five* all-black teams instead of three. Now, the quality of play will decrease in the new black National League, since two teams’ worth of minor leaguers have to be called up. And the quality of play will *increase* in the white American League, since two teams’ worth of marginal players are sent to the minors.

Finally, suppose MLB creates only *two* all-black teams. Now, the caliber of play on the black teams is higher, because three teams’ worth of players are squeezed into two teams. But the white American League has to call up one team’s worth of minor leaguers to fill some slots. So the quality of play goes down in the white league and up in the black league.

Three cases, three different effects. The impact of segregation on quality of play *depends*. Quality could stay the same, it could decrease, or it could increase. Because of that, the argument that says “because Roy Halladay doesn’t have to face Ryan Howard, the quality of play must be worse in the white league” is not valid. The conclusion doesn’t follow from the premise.

Now, suppose two years later, MLB sees the folly of its ways and decides to re-integrate both leagues. Again, whether the quality of play (that the white players see) increases is the reverse of which of the three cases was true before.

But, if MLB decided to integrate *and at the same time* get rid of the National League, then the quality of play would increase substantially. But (as SdeB pointed out in comment #4) that’s because of the contraction, not the integration.

One final hypothetical: suppose that in 1947, the major leagues had said: “We acknowledge the shameful legacy of segregation of the last half-century. As partial restitution to the Negro Leagues, their players, and their fans, we are announcing that ten Negro League teams are being admitted to the majors, which will now consist of 26 integrated teams.”

In that case, we could all agree that integration would not have improved the quality of play. Right?
   24. Sam M. Posted: August 31, 2006 at 01:17 PM (#2162894)
Phil, I love hypos more than most -- they're my stock in trade (I'm a law professor, after all). But before dealing with those, let me ask you a simple question. Do you not agree that the most following is the most likely scenario had there been no color line before 1947:

There would have been more players competing for the same number of jobs on the same number of major league teams.

That is, the black players would have had an opportunity to compete for roster spots on the 16 existing major league teams -- opportunities denied to them -- but there would have been no expansion, no additional teams, and no greater number of roster spots?

If that is so, then surely you agree that this increased competition for the same number of jobs would have squeezed out the worst of the white players, the ones the immortals most readily and happily feasted on, and in their place would have come the best players from the Negro Leagues? And this would have had some impact on the ability of the immortals from the pre-1947 era to dominate, to compile the eye-popping stats they compiled.

Certainly, we lack the ability to quantify this effect, but logic tells us the effect almost certainly existed, and that the massive injustice done to the African-American players was also an injustice (lesser, to be sure) done to Babe Ruth and Ty Cobb and Lou Gehrig and Al Simmons and Lefty Grove, for we will never know what they might have done with the opportunity to assemble a career against the true, best competition their era had to offer. Tom Seaver and Hank Aaron faced one another, as did Koufax and Mays, and all their contemporaries since 1947. They know, and so do we. I don't know about the Babe -- but I have no doubt that whatever his numbers would have looked like in a just world (and they'd have still been awesome) they wouldn't have been as good as they were. We DO have to discount them. With all due respect, you've done absolutely nothing to convince me otherwise.
   25. Phil Birnbaum Posted: August 31, 2006 at 01:49 PM (#2162923)
Hi, Sam (#24),

You write, "We DO have to discount [Ruth's stats]. With all due respect, you've done absolutely nothing to convince me otherwise."

Yes, I agree that you have to discount Ruth's stats. I mean, there was probably a 20% increase in eligible MLB players once black players were included. And, as you note, the number of teams didn't increase. So, yes, you have to do some discounting.

And you also have to discount today's stats for the fact that there were three times (!) as many black players in the leagues as recently as 1974. Again as I wrote, that's 2/3 as much discounting for Derek Jeter as for the Babe.

And don't forget to discount the Babe even more, becuase there the US population, and therefore the number of potential star players, grew by over 50% between 1920 and 1960, with no increase in the number of roster spots. That's two and a half times the adjustment you have to do for segregation.

I agree with you that racial segregation was a massive injustice. But it is NOT a massive cause of increased quality of play from Babe Ruth's time.
   26. BDC Posted: August 31, 2006 at 02:10 PM (#2162952)
we don’t discount Mike Schmidt’s career because he didn’t have to face American Leaguer Jim Palmer

He did, though :) Palmer got Schmidt to pop up in the fifth inning of Game Three of the 1983 Series, and Palmer went on to win the game.

As to the main point of the thread: sure, you have to discount Babe Ruth's raw numbers. Unlikely that he would hit .375 with 50 HR and 150 walks on a regular basis even in today's game, let alone the 1960s or 70s. But relative to his contemporaries of both colors, and relative to other great players across eras, you have to assume he's right up there. Ruth established himself as the best white player in baseball for a long time. In every era, the best white players and the best black players have performed at pretty much the same level, and one assumes they would do that whether leagues were segregated or integrated. I've never seen a reason to assume that you'd take Babe Ruth, or Oscar Charleston for that matter, transport them via time machine to 2006, and see them play at the level of Kevin Mench or Gary Matthews Jr --especially if some kind of "segregation discount" is one's major reason for assuming that deflation. Ruth and Charleston would both be major stars today. The exact OPS+ they would ring up is somewhat irrelevant to the global issue here.
   27. Sam M. Posted: August 31, 2006 at 02:11 PM (#2162953)
And you also have to discount today's stats for the fact that there were three times (!) as many black players in the leagues as recently as 1974.

No, Phil -- that's a very different sort of phenomenon (some members of the population deciding against pursuing a baseball career), one you don't discount for today any more than you'd discount for the voluntary choices young men made in the 20s or 50s to not become baseball players. Demographic shifts that bring about changes in player populations (e.g., fewer black, but more Hispanic players over the last quarter-century) don't require the same sort of adjustments. We can safely assume (unless you have arguments and evidence to the contrary) that there is currently no significant excluded population of players who are better than a significant number of those currently in the major leagues, whose presence would change the competitive environment confronting the best players.

There are a whole slew of reasons why there has been a decline in the number of African-American players -- but I think it is safe to say that the very best African-American baseball players are getting to the majors. The problem, it seems to me, is that a lot of the best A-A athletes aren't becoming baseball players in the first place. But if they aren't becoming baseball players, then they're certainly never developing into major league-quality baseball players, and they certainly can't be said to be of a caliber that they should displace Chris Woodward or Joe McEwing. Some bad individual decisions aside, the guys in the majors right now are the best players that the teams can stock their rosters with.

THAT is the difference: Derek Jeter is facing the best baseball players, the best competition. Maybe Tiger Woods would have been an ace pitcher instead of a golfer in 1960; we'll never know. Let's say he could have been. But Mark Redman is not keeping a job because Woods is excluded; Redman is keeping a job because Woods is not (having made the choice not to become a baseball player many years ago) good enough. Redman is therefore the best competition for Derek Jeter to face. Even if we might wonder what it might be like if some of the great athletes in other sports HAD selected baseball, they didn't . . . so they aren't competition for Derek Jeter, and don't impact his numbers.

The case for discounting Ruth's numbers is completely different. The color line meant he didn't face players who would have affected his numbers.
   28. Phil Birnbaum Posted: August 31, 2006 at 02:29 PM (#2162972)
Sam (#27),

You're making a moral argument, not an empirical one. You're arguing that you only discount for injustice, you don't discount for "voluntary choices."

My point is about the quality of play. Regardless of whether Josh Gibson was barred from the majors, or whether he was offered a contract with the Yankees but decided to become an accountant instead, the effect on quality of play is exactly the same.

Shredder is going to think I hired you to make the exact point I'm arguing against, after he wrote that he's never heard it said. :)

Phil
   29. Guapo Posted: August 31, 2006 at 02:39 PM (#2162983)
Suppose, inexplicably, that MLB decided to segregate. Noticing that about 10% of players are black, they decide that three teams will be all black – call that the new “National League”, and the other 27 teams “American League” will be all white. Will the quality of play change? If the overall caliber of the two groups of players is the same, the quality probably will not change. The exact same players have jobs.

I'm not sure it's that easy. I think the quality of play in both leagues will decrease. I think the issue is you're treating all players as fungible.

Let's assume in MLB there are only 30 shortstops who are replacement level or above, and 6 of them (20%) are black. Now, in the AL you've got 27 shortstop positions, and only 24 replacement level SSs. So you're going to have to fill that position with sub-replacement level players, or you can convert other players from "surplus" positions (getting to that in a sec).

Meanwhile, in the NL, you've got 6 good SSs for 3 spots. 3 of those guys are either going to the bench, the minors, or they'll have to move to other positions to fill talent shortages.

Obviously, if blacks make up 10% of the players and 20% of the shortstops, there must be a countervailing shortage of blacks at some other position- let's say P. MLB had 300 pitchers going into this exercise, but only 5% (15) were black.

In the AL, you now have 270 pitcher spots and 285 pitchers. Those 15 extra guys are either going to the minors, or you can try and convert them to SS to fill the SS holes. Good luck with that.

In the NL, you have 30 pitcher spots and 15 pitchers, meaning each team is scrambling to fill half their staffs. You'll have to convert players from other positions or bring up minor leaguers to fill their positions.

This is why racial segregation is economically inefficient, boys and girls.
   30. Phil Birnbaum Posted: August 31, 2006 at 02:45 PM (#2162991)
Guapo (#29),

Yes, very good point, didn't think of that.

I guess you'd have to change my argument to something like, "suppose MLB observes that 20% of shortstops are black, and decrees that all and only teams in the NL Central must have a black shortstop ..." and so on.

Or, you could just add "assume black players and white players are equally suited to all positions."

Either way, you can show cases where segregation makes the quality of play better, worse, or the same, which was the idea.

But point well taken.
   31. ERROR---Jolly Old St. Nick Posted: August 31, 2006 at 02:49 PM (#2162995)
Thank you, Sam. You've spared me a lot of energy. I still don't think he gets it. This one quote of his pretty much cinches it:

And you also have to discount today's stats for the fact that there were three times (!) as many black players in the leagues as recently as 1974. Again as I wrote, that's 2/3 as much discounting for Derek Jeter as for the Babe.

As if the pre-1947 Major Leagues would have welcomed all those black Latinos who have more than replaced the numbers of those African Americans who may have gone onto other pursuits. The "empirical" distinction between the talent pool of Ruth's day and our day has to do with skin color, not place of birth. Duh.

<u>This guy isn't a moron. He has done TONS... literally flipping tons of great research and is the editor of SABR's By the Numbers Newsletter. He is a really good reseacher... of course I don't really get his point - but he is far from a moron</u>.

<i>I'll grant your point, but you sure as hell couldn't prove it by what he's written here. The most charitable thing one could say about him on the basis of this piece is that like many saberites, he suffers from the disease of reflexive contrarianism, which shows up completely in the chalk on the blackboard tone of the article. But if it's only a temporary affliction, I'd be glad to see what else he's written.


I see. So when people you disagree with agree with the majority view, it's "groupthink." When people you disagree with disagree with the majority view, it's "contrarianism." But of course when you disagree with the majority view, you're not "contrarian"; you're a brave dissenter from groupthink.</i>

Whatever you say, David. You're the only one using the word "groupthink" here. And I have nothing against "contrarianism" per se, only in a case like this where it seems to be so completely forced. In this it does remind me of much that I've read in Slate, with its tone of "everything you thought you knew is wrong." And while 98% of sabr and sabr-influenced material doesn't fall into this category, I have no hesitation in placing this piece into it. I'm glad to hear that apparently it's an aberration.

And BTW, David, if you can find one (1) post of mine where I've ever termed anyone, in any context, a "brave dissenter," I'll buy you a 10 course meal at the restaurant of your choice and a signed copy of Atlas Shrugged for Hanukkah. Words like "brave dissenter" are laughably out of place in the context of any discussion of baseball. There are no penalties here other than an occasional boot from The Bootman, and even that's only a brief suspension.
   32. bunyon Posted: August 31, 2006 at 02:54 PM (#2162998)

And BTW, David, if you can find one (1) post of mine where I've ever termed anyone, in any context, a "brave dissenter," I'll buy you a 10 course meal at the restaurant of your choice and a signed copy of Atlas Shrugged for Hanukkah.


Come on, Andy, you hate the Braves.

(Do you really have a signed copy of Atlas Shrugged (assuming signed by author and not Nathanial Brandon)? That would make (for me) a fine mother's day gift.)
   33. Phil Birnbaum Posted: August 31, 2006 at 02:59 PM (#2163006)
Andy,

I'm not sure I understand your comment, but the best I can figure it, it's this:

1. The number of [American?] Blacks dropped by two-thirds; but
2. The number of latinos (some of whom are Black) increased enough to compensate for the missing black players.

You're agreeing with me ... you're saying that you can't simply look at the numbers of black players and draw a conclusion about the quality of play, becuase *there are other important factors*, like the increase in Latino players!

That's true, both those factors need to be considered. And you also need to consider Japanese players, and population increases, and expansion, and less interest in baseball among kids, and all that.

Unless I'm misunderstanding your point.
   34. ERROR---Jolly Old St. Nick Posted: August 31, 2006 at 03:01 PM (#2163007)
(Do you really have a signed copy of Atlas Shrugged (assuming signed by author and not Nathanial Brandon)? That would make (for me) a fine mother's day gift.)

No, and that means I'm really at serious risk, since I'd be out between $1500 and $12,500 if David could ever take me up on my offer, ho ho.
   35. ERROR---Jolly Old St. Nick Posted: August 31, 2006 at 03:04 PM (#2163014)
Phil, the point is that for empirical purposes, in comparing the present day to pre-1947, a black is a black is a black, and the number of blacks has increased, not decreased, since the 70's. What does the country of origin have to do with anything?
   36. Sam M. Posted: August 31, 2006 at 03:05 PM (#2163018)
You're making a moral argument, not an empirical one. You're arguing that you only discount for injustice, you don't discount for "voluntary choices."

Phil,

I don't think I am making a moral argument. I'm making a distinction between one phenomenon that existed throughout the entire history of baseball (the sport can only choose among young men who have decided, at some much earlier point in their lives, to take up the game and develop their skills sufficiently to get noticed as "baseball players"), and one that existed only until 1947.

It has ALWAYS been true that baseball teams could only select the Best Available Players. A lot of kids got killed in WWI. That affected the quality of play in the 20s. I'm not suggesting we discount Babe's numbers for that, any more than we should discount for the decline in the percentage of A-A kids taking up baseball in the last 25 years, or give a bonus for the surge in the Hispanic talent coming into the game in that same period. Those things are so complex and difficult to sort out (more sports to choose from, growing population base, international players, wars . . . .) that I think you just have to put it all to one side for these purposes.

But the color line is a unique and quite different phenomenon: the ONLY occasion where we know there was a pool of already-qualified players (not ALL the Negro Leaguers, of course, but a sub-set of them), demonstrably better than a pool of the Major Leaguers then holding jobs (again, not all of them, but some percentage of the worst of them), who were excluded. It was not just a matter of the possibility that if some other, unknown and unidentifiable group of people had taken up the game, the relevant population would have changed and the game along with it. That was true in 1910, 1920, 1930, 1940, etc. In this one era ONLY, a specific group of players DID take up the game, were good enough to have changed it, and were denied the opportunity.

There is, of course, a moral argument about that. But there is also an empirical one. It is not that you "only discount for injustice." It's that this era has a phenomenon unique to it that you must discount for, whereas the "voluntary choice" issue is one that is common across baseball history, and thus requires no discounting for any particular time frame.

It may well be that one particular era suffers more from the "voluntary choice" issue. (I wouldn't really call it "voluntary choice," by the way -- since it also encompasses things like wars that simply keep people from ever taking up baseball or making it to the majors. But you get the idea . . . . The terminology is convenient, so let's go with it.) But I doubt it, given the wide variety of factors that are encompassed under that umbrella. I suspect every era has taken its hits in various ways. And if there is a case to be made for heavier discounting for one era, like today, comparable to the impact of the color line, I certainly haven't heard it made effectively. The decline in A-A players, after all, can be argued to have been balanced by overall population growth as well as the influx of international players.

Sam
   37. ERROR---Jolly Old St. Nick Posted: August 31, 2006 at 03:06 PM (#2163022)
A simpler way of putting it is this: What percentage of today's Major Leaguers would have been racially ineligible in Babe Ruth's day? That's the only germane empirical number.
   38. Shredder Posted: August 31, 2006 at 03:18 PM (#2163039)
Shredder is going to think I hired you to make the exact point I'm arguing against, after he wrote that he's never heard it said. :)


I screwed up when I used the word "discount". That's my bad, and you'r right to correct the point on that. I explained that poorly, and it sort of changes the tenor of that whole post. I was thinking along the lines of someone arguing the pre-integration records should be, well, not so much ignored as perhaps placed in their own category. I've never heard anyone advance that theory seriously, and it certainly wasn't clear that that's what I was getting at. That's what I get for posting that late.

So addressing simple "discounting", does Ruth get "points off" for playing before integration? I think it's an example, like the many that you cited, of how the game has changed, and any attempt to compare records across generations has to take that account. My interpretation of your column (and please correct me if I'm wrong) is that integration had such a nominal effect that Ruth's records should not be discounted solely because of integration, and that because there was already a sort of segregation between the leagues, failure to integrate the negro leagues was really inconsequential. I don't think I'd agree with that, and I definitely think his records need to be viewed in that historical context. We don't know how much a difference integration made, and reasonable minds can disagree. But ultimately, since I'm not making sense, I'll just agree with what Sam said.

The color line meant he didn't face players who would have affected his numbers.

What about the reserve clause? I'm no historian, so maybe there's been studies done that I haven't read, but to what extent did teams like the Yankees stuff players in the minors who could have been on other major league teams?
   39. kthejoker Posted: August 31, 2006 at 03:32 PM (#2163059)
If for every Major League at-bat you could measure every lurking and dependent variable which led up to that particular battle (batter vs. pitcher), you would have to include race. It may not explain 100% of the battle (certainly gender has had a much stronger effect on which person is in the batter's box and which person is on the mound for any at-bat), and it explains much less of the battle since integration, but clearly race had an impact on that battle pre-1947 (and, let's be realistic, well into the 1950s, too.)

Certainly other factors such as age/cohort ("Why Babe Ruth Never Faced Pedro Martinez" is a pretty short article), league separation ("Why Babe Ruth Rarely Faced Carl Hubbell"), roster positions ("Why Joe McEwing Gets To Face Greg Maddux"), population expansion ("How Billings, Montana's Dave McNally Made The Pros"), and the rise of other sports ("Bizarro Jim Brown, Hall of Fame Leftfielder") have all had their place. But so does race, and if you held all of those other points equal, race still had an effect on all of those pre-integration at-bats. To suggest that because it was not the only factor that it is not an "important" factor is disingenuous and wrongheaded.
   40. Phil Birnbaum Posted: August 31, 2006 at 03:51 PM (#2163087)
Hi again, Sam,

You write (#36), "the sport can only choose among young men who have decided, at some much earlier point in their lives, to take up the game and develop their skills sufficiently."

So what you seem to be saying is that when black youth choose not to pursue baseball, we don't have to account for that, because they're not in the pool.

That's not right. The result of having black athletes choose other sports is that the pool of available talent is shallower. And that means the quality of play is lower.

Look at it this way: instead of barring black players from the majors, what if the United States had explicitly banned black children from learning baseball. Josh Gibson would have lived, but never played. Are you suggesting that, in that case, you WOULDN'T have to discount Ruth's stats, because now he faced the best players available?

That makes no sense to me. Ruth would have had exactly the same advantage either way. Whether the players aren't in the pool because (a) they were barred, (b) they never learned the game, (c) they chose other sports, or (d) they were never born, the effect is the same on quality of play. And that's the question we're asking.
   41. Phil Birnbaum Posted: August 31, 2006 at 04:13 PM (#2163112)
Shredder (#38),

You write, "My interpretation of your column ... is that integration had such a nominal effect that Ruth's records should not be discounted solely because of integration ..."

Correct. There are several other important factors to be considered also.

" ... and that because there was already a sort of segregation between the leagues, failure to integrate the negro leagues was really inconsequential."

Not exactly. My point was that segregation between the leagues is an example of how segregation is not *in and of itself* proof of a lower quality of play. It was perhaps a weak point, and furthermore I didn't get it across very well, but the fact is that quality only increased because the majors pulled players from the Negro Leagues *instead of absorbing the teams*. If the majors had absorbed the Negro Leagues in full, like the NHL absorbed the WHA in 1979, the quality of play would have DECREASED.

As SdeB (#4) put it, all that matters is the pool of talent and the number of roster spots. It doesn't matter WHY the pool of talent increased.

So I think I otherwise agree with what you're saying, Shredder, except that part about agreeing with Sam.
   42. ERROR---Jolly Old St. Nick Posted: August 31, 2006 at 04:34 PM (#2163131)
Look at it this way: instead of barring black players from the majors, what if the United States had explicitly banned black children from learning baseball. Josh Gibson would have lived, but never played. Are you suggesting that, in that case, you WOULDN'T have to discount Ruth's stats, because now he faced the best players available?

That makes no sense to me. Ruth would have had exactly the same advantage either way. Whether the players aren't in the pool because (a) they were barred, (b) they never learned the game, (c) they chose other sports, or (d) they were never born, the effect is the same on quality of play. And that's the question we're asking.


Which again makes me wonder why you make the distinction between African American players and other black players when it comes to comparing the relative sizes of the talent pools. You can even leave race completely out of it if you want and just concentrate on the numbers.

Ruth's day: White Americans and a tiny smattering of white Cubans. No non-whites, even though individual players and teams of non-whites in both the continental U.S. and in Latin America had demonstrated their abilities in match after match with Major Leaguers going back to the turn of the century.

Today: The whole world of baseball players, identified and grabbed up by the Major Leagues as never before, and enticed to America with unheard of riches.

At this point, make everyone the same color and just start counting. The relative sizes of those two talent pools are like night and day.

You can't just count the U.S Negro Leagues and leave it at that. You can't ignore the fact that beyond the shores of the U.S. there's an enormous talent pool out there today that was completely untapped in earlier times, and in truth barely tapped on any systematic scale until the late 70's. The leader boards of today are just as topheavy with Afro Latin names as they were topheavy with African American names a few generations ago, and none of these players could have played in Ruth's day. All of that missing talent sure as hell wasn't being duplicated by the pool of white farm boys and stickball players, and you haven't even addressed this point, let alone resolved it. It seems to interest you not at all, at least not to date.
   43. Sam M. Posted: August 31, 2006 at 04:45 PM (#2163147)
Look at it this way: instead of barring black players from the majors, what if the United States had explicitly banned black children from learning baseball.

That policy would have been like the color line in one way, and like the "voluntary choice" cluster of factors in another -- i.e., it would have been a hybrid.

It would have been like the color line in that it would have been specific to the one era, rather than generalized across time. It it this factor you are missing, I think, when you try to compare it to assessments of the talent pool for the 60s or 70s or 90s: EVERY era has factors that may affect the depth of the talent pool, including Ruth's era. Until you show me otherwise, with a fairly high burden of proof, my assumption is those factors balance each other out. A decline in A-A's playing the game in the 90s? OK. Thousands of young men died in World War I, seriously depleting the talent pool in the 20s, and in World War II, having the same effect in the 50s.

But the color line (and your hypothetical about barring black children from learning baseball)? Those were unique to pre-1947 baseball, and on top of the factors that this era shares in common with post-Robinson time frame. Thus, they stand above and beyond as additional factors affecting the assessment we should make of the performance of the players of that era.

But it would have been different from the color line -- and more like the "voluntary choice" factors -- in a second way. Unlike the color line, it wouldn't have resulted in a group of players of major-league quality standing there, better than some of those white guys wearing major league uniforms. We'd instead have the Tomb of the Unknown Baseball Players -- a great unanswerable mystery. There'd be a group of young men who didn't ever take up baseball, like the Tiger Woods example in my earlier post. And we'd know that Babe Ruth did, indeed, face the best possible players his era had to offer -- as we know he did not in the real world.

If that were the case, then my argument WOULD be genuinely only a moral one. Then, his numbers would be the product of injustice -- but empirically, they would be the stats that reflected the performance the best hitter of the time would produce against the best pitchers he could face, period. But empirically, that simply isn't true of the real world. In the real world, black children DID learn baseball, some of them DID become great players, great enough to have earned (but not obtain) major league jobs, and great enough to have affected Ruth's stats had they gotten the chance.

There's just no two ways around that.
   44. Phil Birnbaum Posted: August 31, 2006 at 04:56 PM (#2163159)
Andy (#42),

Aren't you making my point? You're pointing out that the integration of 1947 is only a small part of the increase in the talent pool since Babe Ruth's time. That's my whole argument!

I don't understand what you're saying about players from other countries. Sure, go ahead and add them to my argument -- it doesn't change the conclusion.

Didn't the influx of hispanic players come much later than the integration era? That is, if you want to talk about Latin players, don't you have to discount Jackie Robinson's stats too, since full "integration" of those guys didn't come until a couple of decades later?

Also, I don't understand what your response has to do with the excerpt of mine you quoted, and you didn't address the question I was asking.
   45. Phil Birnbaum Posted: August 31, 2006 at 05:02 PM (#2163167)
Sam (#43),

"EVERY era has factors that may affect the depth of the talent pool, including Ruth's era."

My point exactly!

"Until you show me otherwise, with a fairly high burden of proof, my assumption is those factors balance each other out."

All those umpteen other factors balance each other out? Because you say so? And why all those other factors EXCEPT integration? How do you figure the umpteen factors balance out, but the umpteen-plus-one factors don't?

Because segregation was unique to that era? You immediately go on and mention three more factors that were unique to their own eras!

And, by the way, you didn't address my original question. If blacks had been barred from baseball from birth, and Josh Gibson never played, would you still argue that Ruth's stats have to be discounted? If so, why not also discount Derek Jeter's stats 2/3 of the way?
   46. Shooty would run in but these bone spurs hurt! Posted: August 31, 2006 at 05:03 PM (#2163168)
There's no denying Babe's greatness, but his iconic status surely was aided by not having to compete with the Oscar Charleston's and Josh Gibson's and Cristobal Torriente's of the world. The question for me is not that Babe wouldn't have hit 60 homers--I think he'd have come close enough--but how many other players who were excluded in a very covert way from attempting the same thing, players who I think most of us can agree were just as skilled and physically talented. It's not Babe's fault, but there will always be that uncertainty for me that he was really the best. It's a shame but it's the nature of our history. Shame can be constructive, though. I hope.
   47. Ned Garvin: Male Prostitute Posted: August 31, 2006 at 05:29 PM (#2163236)
I think, when you try to compare it to assessments of the talent pool for the 60s or 70s or 90s: EVERY era has factors that may affect the depth of the talent pool, including Ruth's era. Until you show me otherwise, with a fairly high burden of proof, my assumption is those factors balance each other out.

Of course, every era has many factors changing the talent level available to the major leagues. Maybe today Dave Winfield plays football, for example. And it does go both ways - baseball is not the national pasttime anymore (by the original meaning anyway), so in general, urban and rural youth play less baseball. How much do the myriad factors affect the talent pool? I have no idea of course. But to assume that:

a) They cancel each other out exactly, except for segregation

and

b) Segregation was a larger factor than any of these

is a huge jump with absolutely no evidence that I know of. I'm not saying this is wrong, but I am saying none of us has any evidence either way. How much would segregation have affected the major leagues? Someone threw out 10% above, that seems reasonable if that is the amount of the population restricted. Can wars affect the talent level that much? The existence of large football and basketball leagues? Population increases? Expansion? Increased Latin players? I would not count any of these (and more) out.

Segregation was one of many factors which affected the talent levels, and it could have been the dominant one, or it could have been lost in the noise. The main differences are that it was a huge moral issue, and it was very tangible. If Oscar Charleston dies at age 4, or chooses to play hockey, or loses a limb at war, or anything else, then nobody notices his absence in baseball, like Sam M. has been implying. But if he grows up and dominates the Negro Leagues, then everyone can see that MLB is missing out on a great player, and MLB pitchers are getting off lightly by facing a replacement-level batter instead of Charleston. It is very clear, but not necessarily more impactful than if Charleston had dreamed of being a dentist his entire life, and never played baseball.


The question for me is not that Babe wouldn't have hit 60 homers--I think he'd have come close enough--but how many other players who were excluded in a very covert way from attempting the same thing, players who I think most of us can agree were just as skilled and physically talented.

This is nitpicking, but the number of Negro League players equally talented and skilled as Babe Ruth was probably about 1, plus or minus 1. How many white players (picking from a much larger pool) were equally talented and skilled as Ruth?

And to continue nitpicking, those players DID get the chance to hit 60 HRs, just not in the same venue. And they DID play against white players, and clearly showed their worthiness.

I do agree with the larger point of course - that black players did not get to compete in the same official league as the white players, so it is difficult to tell exactly how they compare to the great white players. And as a result, they are relatively forgotten by the public, which is a shame.
   48. HowardMegdal Posted: August 31, 2006 at 05:48 PM (#2163273)
"And why all those other factors EXCEPT integration? How do you figure the umpteen factors balance out, but the umpteen-plus-one factors don't?"

While quantifying these factors seems to be impossible, I'd point out the difference between all the other factors, such as wars, population changes, etc. all kept POTENTIAL ballplayers out of the major league talent pool. Those in the negro leagues were ACTUAL ballplayers. In other words, we don't know what percentage of kids killed in WWI would have followed the career path of the professional ballplayer- but among negro leaguers, that percentage is 100%.

It's like asking if a pool of junior high school baseball players is a valuable as a pool of 24-year-old in-prime major leaguers. The exclusion of that pool, particularly when representing 10-12% of the U.S. population (is this number way off?) would be a huge factor to consider. I agree we can't quantify how huge relative to the other factors. But it's hard to imagine that it isn't the far biggest factor of them all, at least for me.
   49. ERROR---Jolly Old St. Nick Posted: August 31, 2006 at 06:08 PM (#2163319)
Aren't you making my point? You're pointing out that the integration of 1947 is only a small part of the increase in the talent pool since Babe Ruth's time. That's my whole argument!

Phil, we're not making the same argument at all.

Yours is that race is but one small part of the different contexts of Ruth's era and today.

Mine is that race is a huge difference---the biggest difference by far when it comes to affecting the difference between the talent pools of the respective eras.

First you dwell on the fact that African Americans are only a third as plentiful in relative terms as they were in 1974, and start talking about "discounting" Jeter's statistics by 2/3 because of that---not literally, but as a way of analogizing the decline of African Americans since 1974 to the exclusion of all blacks in Ruth's time.

And when I make the obvious rejoinder that the total percentage of Ruth era-ineligibles playing today--which is the only relevant point of racial comparison--is higher than it's ever been, you say that I'm "making" your point! Simply mind boggling.

Didn't the influx of hispanic players come much later than the integration era? That is, if you want to talk about Latin players, don't you have to discount Jackie Robinson's stats too, since full "integration" of those guys didn't come until a couple of decades later?

Finally an argument that makes empirical sense. And yes, since in effect the early part of the Jackie Robinson era was nearly as segregated as the Ruth era, and even the latter part of his playing career was still nowhere near as demographically inclusive as it is today, I certainly do discount the statistics from the 1950's somewhat. I've argued many times here that there's been a vast improvement in the quality of Major League baseball since the 1950's, largely due to the enormous expansion of the talent pool.

And the extent to which the statistics of the 1950's should be discounted, compared to the statistics of the Ruth era, is indeed a debatable question. I'd tend to say that absent interleague play, the AL stats are more suspect than the NL's, since nearly all of the top black talent in the 50's and 60's wound up in the NL. But this is something we can never really resolve with any precision.

But what is beyond debate is that a big majority of the best position players of today, and a significant percentage of pitchers, would not have been eligible for the Major Leagues of Babe Ruth's day. And by playing semantic games by distinguishing between African Americans and black Latinos, all you're doing is reinforcing the thought that in your argument here, you're in over your head.
   50. Sam M. Posted: August 31, 2006 at 06:18 PM (#2163343)
I have no idea of course. But to assume that:

a) They cancel each other out exactly, except for segregation

and

b) Segregation was a larger factor than any of these

is a huge jump with absolutely no evidence that I know of.


The ONLY safe assumption is that they roughly cancel each other, because they are all the same in kind. They all take out potential players from the overall, great population mass in unknown quantities before they ever become anything resembling baseball players. And these factors have been affecting the talent pool since major league baseball began -- every era, always. Thus, until someone does empirical work that establishes some other, better assumption, we can ONLY assume a general rough equivalence, because we have no basis for any adjustment otherwise.

But the color line is manifestly different. It is unique to that ONE era, unlike the other things being brought up. Thus, it must affect the population of pre-1947 players in addition to the factors that were already affecting it, the factors that period has in common with the other eras. It doesn't matter if the color line is bigger than the other factors; the point is that it is an additional factor that ONLY THIS ERA HAS. If the other eras are anywhere near in balance on the factors they share, the presence of another, significant, era-specific factor makes for an impact on the players' performance that applies only to the pre-1947 leagues.
   51. Phil Birnbaum Posted: August 31, 2006 at 06:58 PM (#2163443)
Sam (#50),

Well, yes, integration affects only one era. So what?

Look, the talent pool has been increasing, say, 1% every year, due to population increases. The talent pool increased 20% ONE TIME based on a specific era-specific event.

Ignoring compounding, that's 85% (since the Babe's time) population, and 20% integration.

You seem to be saying that all those 1% figures "cancel out". They can't cancel out -- they're all in the same direction! (If I steal $20 from you today, and then I steal $20 from you tomorrow, are we even?)

If, by cancelling out, you mean that *other factors* cancel out the 85%, what might those factors be, and how do you know they add up to exactly 85%? Further, why don't they cancel out the 20% from integration? Do demographic trends "choose" which other demographic trends they cancel out?

Be specific. What cancels out what?
   52. Phil Birnbaum Posted: August 31, 2006 at 07:20 PM (#2163492)
Andy (#49) says

“… race is a huge difference---the biggest difference by far when it comes to affecting the difference between the talent pools of the respective eras.”

Not true.

The increase in US population from 1920 to 1950 was about 50%. The increase in talent in the 50s due to integration of American blacks was about 20%. The increase in talent due to integration of non-American blacks was small – call it 10%, which is likely very high. That means that population beats integration, 50% to 30%. And that doesn’t even include changes since 1950.

I think your error is this:

“… the total percentage of Ruth era-ineligibles playing today … is the only relevant point of racial comparison …”

This is completely incorrect. The adjustment to Ruth’s stats must be based on the black talent pool at the time, not the black talent pool now. To see why, imagine that 100 years from now, the majors are 100% black. By your standard, 100% of players were ineligible in Ruth’s time, and so Ruth’s stats have to be discounted to zero!

That’s absurd, obviously. The percentage of black players today is completely irrelevant. All that’s relevant is the available talent per roster slot today versus the talent per roster slot in Ruth’s day.

You are focusing on the idea that a large percentage of players today would have been ineligible in the 1920s. That is true. But it doesn’t impact the mathematics of what Ruth would have done in the context of the talent available in the 1920s. The number, percentage, or country of origin of blacks playing today is completely and utterly irrelevant to the question under discussion.
   53. Sam M. Posted: August 31, 2006 at 08:34 PM (#2163700)
Be specific. What cancels out what?

Phil, I am saying that the pre-1947 immortals didn't face the best possible competition, and thus their accomplishments must be discounted somewhat -- and that the discount is significant. The factor that is involved here is different than any other era, because it is the only era in which major league quality players, already trained, developed, and proven, were excluded and lesser quality players were kept on the job -- thus allowing the superstars to inflate their performance against lesser competition.

You have argued, in response, that even in subsequent eras, the immortals have faced diluted or lesser competition, because of "other factors" that have kept some players who would have been better than the worst major leaguers from ever getting to the majors -- from ever becoming baseball players in the first place. Instead, they become (my example) Tiger Woods, or Kellen Winslow.

My answer to this is simple: these other factors balance out over time. Whatever factors keep nascent baseball superstars at the age of 10 from ever getting to be baseball prospects at the age of 18 kept them out in 1910, 1920, 1930, 1940, 1950, 1960, 1970, 1980, 1990, 2000, and they keep them out today. Those factors include:

* injuries and disease;
* other sports;
* wars;
* lack of interest in sports (or greater interest in other things).

We're not talking here about the population of baseball players (whether that grows or not, at 1% or whatever) -- we're talking here about the population of the baseball players that never were. Aren't we? The point is whether there are enough of these -- the "never weres" -- to create a significant discounting effect. And my point is that there is simply no reason to believe THIS pool of players -- the ones winnowed out by voluntary choices or cruel fate, who never got to be of major league quality -- is any different now than it was in 1925 or 1960. And if it's no different over time, then we don't have to discount it for any particular era.

You tell me. Why should we discount it for any era, the way I'm saying the color line discounts pre-1947 performance? What reason is there to believe that there is an equally serious depletion effect, and for what time period, that is comparable to that exclusion? Because I simply see no evidence that any era has seen that sort of effect, unique to that era -- and it has to be unique to that era to justify any sort of discounting.
   54. Kiko Sakata Posted: August 31, 2006 at 08:42 PM (#2163714)
The factor that is involved here is different than any other era, because it is the only era in which major league quality players, already trained, developed, and proven, were excluded and lesser quality players were kept on the job -- thus allowing the superstars to inflate their performance against lesser competition.

Actually, there was one other "era" when this was true: during World War II, when a whole slew of major-league baseball players were in the armed forces. I have no point beyond that, I just thought maybe it would encourage a cogent and relevant point from somebody else.
   55. rr Posted: August 31, 2006 at 11:13 PM (#2163883)
It seems clear there is a qualitative difference, in that, as Howard and others said, we have a case here such that actual ballplayers--star players--who wanted to play MLB, presumably, were unable to do so. That is a very different thing than socio-demographic changes altering the talent pool.

Quantifying it, of course, is another matter, and depends on several assumptions related to creating the hypothetical and projecting the stats. If we posit a situation in which Landis, a great man in many ways, has an Ebenezer Scrooge-moment WRT race and orders MLB to integrate in October 1921, it seems reasonable to assume that by the end of the decade, MLB would have 30-50 black players; and they would be the very best players from the Negro Leagues--HOFers and All-Star types.

Then, of course, you have the stat translation problem, which, as far as I know, could only be done speculatively, given the records available.

That said, it seems possible that you could run a sim program with All-Star type players gradually substituted into the various teams, in place of the worst white guys, and run a zillion sim seasons, and see what you got.

Intuitively, it seems likely to me that such an adjustment would reduce Ruth's stats slightly, but not greatly. Maybe he hits 55 HR in 1927 instead of 60.
   56. Phil Birnbaum Posted: September 01, 2006 at 12:53 AM (#2164010)
Sam (#53):

OK, I see what’s happening … you’re leaving out the increase in the talent pool. Forget the “never weres” who were actually born. Consider the “never weres” who were NOT born.

There were 50% more American men in 1960 then there were in 1920. Even all else being equal, there must have been about 50% more men in 1960 who were of equal caliber to the players of 1920. These cannot “cancel out”. Suppose that at any point in time, 3 out of a million people have enough skill to play baseball at the 1920 level. In 1920, there were 106MM people, so about 300 players were good enough to play. In 1960, there were 179MM people, so there about 530 players at that level.

In 1920, only about 240 of the 300 were white, and allowed to play. This means that, in 1920,

60 players were lost to the majors because of color;
230 players were lost to the majors because of not having been born yet.

The fact that segregation was “unique to that era” has no bearing on its relative importance to the quality of play. And the fact that population growth is roughly the same in every era is completely irrelevant.

Plus, if you add in players from other countries, the disparity gets worse. There were probably millions of candidates from Latin countries in 1960, many, many more than in 1920. The talent pool is absolutely huge now compared to then. The increase in talented players these days is orders of magnitude higher than the increase in talented players in 1920 if you included blacks.

The problem is, you don’t see it. There may be, say, six times as many 1920-talent players now as then, but only twice the roster spots. If that’s true, then for every roster spot, there would be two other players, either in the minors or released, who would have been good enough to play in 1920. That pales beside the relatively small number of black players who were barred.

As for injuries, war, competing sports, and lack of interest, you claim their effects are constant. I gave several examples of why they might have been different in different eras. There was no Tommy John surgery in the 30s. In the 20s, it may have made good financial sense to become a pharmacist instead of a shortstop prospect. On the other hand, US blacks seem to be losing interest in baseball, which goes the other way.

You may be right that all these other things add up to the same amount – maybe interest in other sports has grown at exactly the same rate as Tommy John surgery, so that the net in 1930 is the same as the net now. That would be quite a coincidence, wouldn’t it? I prefer to think of them as unknowns instead of assuming them away.

But you may have a point, and these factors, taken together, MIGHT be minor. But the birthrate increase, and the scouting of other countries, accounts for a huge increase in the talent pool.
   57. Minus Ice Posted: September 01, 2006 at 01:23 AM (#2164054)
Intuitively, it seems likely to me that such an adjustment would reduce Ruth's stats slightly, but not greatly. Maybe he hits 55 HR in 1927 instead of 60.

And over the course of his career, it becomes a more significant decrease.
   58. rr Posted: September 01, 2006 at 01:36 AM (#2164082)
And over the course of his career, it becomes a more significant decrease.

Well, if one buys my reasoning, which no one may. But let's say Ruth's record was 672 instead of 714. It is also likely that Jimmie Foxx hits fewer HRs as well.

It's a fun what-if in that Willie Mays would have stuck around, or tried to, to break the record if it had been in the 670s. 1961 would have been different, too. But then, would the number of blacks playing in the 1960s have been higher if the game had been integrated in the 1920s? And if so, would that have changed the quality of play in the 1950s and 1960s? It's the old time-travel-movie trick of one change having numerous effects.
   59. ERROR---Jolly Old St. Nick Posted: September 01, 2006 at 03:39 AM (#2164279)
Phil, the more you post the better you explain yourself, and I do appreciate your willingness to get past some of the initial strong reactions and concentrate on your message.

My problem with stressing raw population increases over the racial factor can be expressed by considering two hypothetical scenarios for the Major Leagues of 2006:

(1) A talent pool drawn from the same demographic base as in 1927, but with the increase in the white U.S. population since that year;

(2) The talent pool as it exists today, with full scouting of the baseball playing world.

We agree that (2) provides by far the greater talent pool. We differ on the emphasis on the different factors, i.e. the racial component vs. overall population increase.

Statistically speaking, your argument is not completely implausible, but two counter-arguments seem far more persuasive to me:

First, the overall difference in population between the Ruth era and today, with respect to the talent pool, is heavily weighted by the increase in population of non-whites from all over the world. The rate of “native white” increase in the U.S. population since 1930, by itself, would barely be enough to overcome the expansion factor. The birth rate and immigration rate among non-whites reduces the “native white” percentage of the U.S. population with every passing day. A Major Leagues of today drawn from the Ruth era demographic pool, even with the increase in that pool’s population, would scarcely be a Major League at all by today’s actual standards of competition.

And second, there is the empirical evidence for the argument that the influx of non-white talent has added to the quality of Major League baseball above and beyond the mere numbers of non-white players. Any one of many measuring rods can be cited: MVP awards; Rookie of the Year awards; top ten leaders in nearly every batting category, baserunning category, and several pitching categories; you name it. This is a sticky point to raise since it implies a racial factor in talent distribution, which I certainly wouldn’t argue, but empirically speaking, facts are facts.

Put these together, and I would argue that the racial component is paramount. The fact that the increase in non-white population (as opposed to their eligibility per se) is responsible for the increase in quality of play is irrelevant, since none of these newly born players would be eligible anyway if segregation were still in place. And no increase in the white population by itself could ever make up for that exclusion.
   60. Eraser-X is emphatically dominating teh site!!! Posted: September 01, 2006 at 04:43 AM (#2164347)
Ok, re-reading, now I'm really confused about what Phil's argument actually is.

Thanks by the way for taking the time to join the discussion!

Are you arguing that we should completely discount Babe Ruth and every other player from that eras statistics because there are other factors involved and when you add all of those up, you end up with a MUCH MUCH larger talent pool in the present day?

And that's not enough, then we can get into factoring in all of the other issue like improvements in technology, conditioning, game strategies, etc., right?

That's what I see when we cease to treat segregation as a special case, and I can see some merit to that approach.

But the other issue I run into is: How is this argument going to be used? It's more than just a fun armchair or mathematical exercise right? Ultimately, people will use these arguments to make judgements and advocate actions with morale implications. We will never be able to fill in all of the gaps of the talent appraisal puzzle, but we will certainly not let that stop us from trying. How do we avoid morale issues when bridging those gaps?

As we do this, how do you suggest we get around the fact that we CAN'T rely on statistics as faithfully for Negro Leaguers because they were lost to history due to a terrible atrocity which has wiped out a small, but significant part of our great pasttime's history?

When we consider Effa Manley's HoF candidacy, tensions rise and there are concrete points made on either side. But the key issue is that every person on the face of the Earth has different circumstances and different opportunities to affect the game of baseball. How do we account for that in evaluating them? We get to plenty of questions that there is no data to solve for, "What was her effect on the game, if any?" "How meaningful, if at all, was it for someone of her background to have that effect?" "How did her ethnicity or gender affect her ability to impact the game and how in God's name are we supposed to assess that?"

We get so buried in unanswerable questions, that we need to remind ourselves what's really so amazing about the great game: it is a mirror for our own principles, quirks and values.

If I believe that minorities are being overcompensated for whatever ill-treatment they might have received, I probably am going to start by seeing her election as a farce. If I believe the opposite, I'm probably going to start from the position that it's about time...

In the end, I don't believe that being right, or even "the truth" is really the most important thing in these cases where the questions are unanswerable.

I believe that Ruth was an inferior player because that supports my faith in a belief that I believe with create a better world--that humanity is better off when the pool we are drawing from is larger and has more diverse perspectives that allow for many more approaches to whatever problem comes along whether it be "hitting a baseball" or "educating a child".

I can understand how that approach would be deeply troubling to an academically trained-mind--I myself find it deeply troubling on that level. But I think it does reflect a reality--we should pursue the best answers on a grander scale to our questions and we should do so with a full understanding of the finite limits of our raw intellect, but the infinite potential of our shared human wisdom.
   61. Phil Birnbaum Posted: September 01, 2006 at 01:09 PM (#2164468)
Hi, Andy,

If you want to compare the conditions of Babe Ruth's career to the conditions today, the only things that matter are

(a) the size of the talent pool
(b) the number of roster slots.

Race is irrelevant, except as it affects the size of the talent pools. I don't know where you're going with this increase in white vs. non-white population, but *it doesn't matter*.

What matters is that segregation in Babe Ruth's time made the talent pool 20 percent smaller. This is the effect of the color bar on Babe Ruth's career. It wouldn't matter if segregation was by religion, or by country, or by place of birth, or by players whose names started with the letter "S".

In 1947, the talent pool effectively got 25 percent larger as a one-time shot. However, the talent pool based only on population statistics has grown by 150 percent since Ruth. And that's only Americans. Include talent in other countries -- regardless of race, which doesn't matter anymore -- and, you're even higher. Suppose that other countries have half the talent pool of the US. Now you're up to a 275 percent increase (250 * 1.5) - 100).

Consider the talent pool (including blacks) in Ruth's time to be 100. The talent pool now is 375. If 25% of the talent in Ruth's time was black, and therefore barred, Ruth had the advantage of "25" points of dilution.

But Ruth also had the advantage, compared to today, of 275 points of dilution. The effect of demographics is ELEVEN TIMES the effect of segregation.

And so my point is that it is somewhat dishonest to say that empirical considerations force you to have to devalue Ruth for segregation but not for demographics.

I repeat that the proportion of blacks in the talent pool today has NO BEARING on the question. You seem to feel that the large number of blacks in the majors today amplifies the effect of segregation in the 1920s. It does not. The effect of segregation on Ruth's stats is exactly the talent pool of blacks *at the time*. The talent pool of blacks later has nothing to do with the effects of segregation on Ruth, even though (as you note) it is true that they wouldn't have had the right to play in Ruth's time.
   62. Phil Birnbaum Posted: September 01, 2006 at 01:31 PM (#2164493)
Let me try an analogy.

I run a golf tournament in my neighborhood. Prize is $100. It's limited to whites only. 200 players from the neighborhood vie for a spot in the tournament. The top 100 players tee-off. A good golfer named "Babe" wins regularly.

Ten years later, under pressure from the mayor, I allow black golfers to enter. Now, 250 players from the neighborhood vie for a spot in the tournament. The top 100 players tee-off.

Twenty years later, I raise the prize from $100 to $10,000,000. Now, a million players vie for a spot in the tournament, some from Spain and Fiji and other far-away places. The top 100 tee-off. Phil Mickleson wins with a 65. Tiger Woods is second at 66. Vijay Singh is third at 68.

An observer notes that the quality of play sure has improved over the last 30-40 years -- it's not the same game at all. "You know," he says, "we can't compare Babe's string of tournament victories to today. We have to discount Babe's record a bit. Because of segregation."

And that would be a bullcrap conclusion. Babe's record is certainly not comparable to today's golf, but that's not because of integration -- it's because prize money is up so high that the talent pool is now in the millions instead of the hundreds. The effect of integration -- 250 people -- is a drop in the bucket compared to the effects of raising the prize money -- 999,750 people.

But you have to think a little bit to see the 999,750 people. If you're a casual golf viewer, it was 100 players before, and it's 100 players now. The 999K people below the surface are invisible, compared the 50 blacks who were barred in the first place (like "Josh," who was a damn good golfer, but died before the tournament opened up). Nonetheless, the effect of segregation is literally only a very small fraction --50/999,750 -- of the improvement.

I can prove it. In the segregated tournament, the top score was usually about 84. In the integrated tournament, the top score was usually about 83. In the third stage, with the big prize, the top score was 65. Isn't it obvious that demographics are the main reason that Babe's accomplishment can't be compared to today?

Sadly, in baseball, both pitchers and hitters improve, so even after the demographics, the raw statistics look almost identical -- the improvement is hidden. So that's two reasons it's easy to ignore the effects of demographics -- you don't see the pool of players directly, and the statistics don't show the improvement.

However, it's there. And no matter how much you admire Josh Gibson and Oscar Charleston -- and they were great players indeed -- they are only a small part of the reason that Babe Ruth's statistics can't be compared to today's.
   63. Matt Clement of Alexandria Posted: September 01, 2006 at 01:46 PM (#2164505)
But, Phil, you don't know that. As Bill James pointed out, the best playright in Topeka today is not nearly as good as the best playwright in Stratford-on-Avon in the 16th century, even though Topeka's a bigger city. The relationship between overall population and skill level of the top .001% at any skill is not going to be linear. It's too weird for that.

IF we accept your premise that the relationship between quality and population is linear, then Babe Ruth sucked. The quality of play, as a linear function of population with the opportunity to play in MLB, is maybe four times worse in 1925 as compared to today. Babe Ruth's AL was about 1/4 as good as today's game - something like low-A ball, perhaps.

Obviously, that can't be correct. We don't know the relationship between population and quality of play. It's a weird thing.

We do know the relationship between segregation and quality of play. It's very obvious, becuase you replace real players with real players. As such, we point to segregation because it's much more knowable and tangible. This is perfectly reasonable, and your moral superiority in the article was highly off-putting. We don't treat hypothetical differences in competiton in the same way as we treat real differences in competition.

The simple way of putting it is that Rogers Hornsby didn't play against the best players of his day, while Robbie Alomar did. That's quite analytically separable from the question of whether there are millions of Dutch boys who might just turn into the next Bert Blyleven if only soccer weren't so popular.
   64. Phil Birnbaum Posted: September 01, 2006 at 01:51 PM (#2164509)
Hi, Eraser-X (#60),

Yes, exactly, your first few paragraphs summarize my argument accurately.

And what I said on my original blog post is that if you argue that we should discount Ruth's stats *only for the effects of segregation*, you are making a moral argument, not an empirical one. The moral part is choosing to adjust for one reason for a smaller talent pool and not for another, larger, one.

Suppose you were to say, "I can't consider Babe Ruth's statistics at face value to today's statistics because of my moral objections to the color bar. However, once an adjustment is made for that, I am comfortable taking those new stats at face value, even though other changes in the talent pool require a far larger adjustment. That's just the way I feel." To that, I have no objection whatsoever.

But if you argue that segregation is the only factor, or the biggest factor, in the empirical adjustment of Ruth's record to be comparable to today's, well, then I object. Because that's just not right, no matter how correct your moral arguments.

Your concern about how this conclusion "will be used" suggests that you are addressing larger moral questions than I am. Which is fine. But I'm not arguing those moral issues here.

And besides, what's wrong with acknowledging that segregation comes out as a small factor in any accounting of Ruth's stats compared to today's ... but, nonetheless, that fact in no way mitigates the shame and injustice that black players endured?

I mean, it's just the way of the universe that not every fact turns out in such a way to easily reinforce our moral arguments. It doesn't mean our arguments are wrong, just that we need to find other ways to reinforce them.
   65. Phil Birnbaum Posted: September 01, 2006 at 02:10 PM (#2164528)
Hi, Matt (#63),

Four times the population doesn't necessarily mean four times the quality of play. The quality of play will be higher, but not linearly.

Consider contracting the majors to 1/4 of their current size, or 7 teams, so that only the best 25% of players get roster spots. Those teams won't hit four times as many home runs, or strike out four times as many batters. But they'll be substantially better.

Imagine it the other way. Suppose you barred players whose names started with A-M. Would the league be "twice as bad" in any statistical sense? I don't think it would. It would be worse, but not THAT much worse. Remember when Bill James said that you could add lots of expansion teams without anyone noticing a drop in quality?

As for the playwright argument, yes, of course, cultural reasons factor in too. But if you want to argue that 10 times the population doesn't mean 10 times the number of potential baseball players, the burden is on you to explain why 1920s black players were the equivalent of London playwrights (and not Topeka playwrights). You can't just say, "Hey, I saw an airplane once, and therefore you can't assume that things fall when you drop them."

Do you have any reason to believe that young men in 1960 had less baseball skill than young men in 1920? Because, unless you do, it follows that the number of potential players at skill level X is exactly proportional to population.

And remember my argument about what if blacks were prohibited from playing baseball at all? By your methdology, since Josh Gibson would never have existed as a ballplayer, there would be no adjustment necessary to Ruth's stats (since you can't assume more potential ballplayers = better baseball). By my logic, Ruth's adjustment would be exactly the same either way, because I would extrapolate how many Gibsons there would have been. Which do you think is the better method?
   66. ERROR---Jolly Old St. Nick Posted: September 01, 2006 at 02:26 PM (#2164547)
Phil,

Again, you express your point more clearly with every post. It's now apparent even to a moron like me what you're trying to get across.

But for the purposes of evaluation, Matt's point remains: The simple way of putting it is that Rogers Hornsby didn't play against the best players of his day, while Robbie Alomar did. If you keep the all-white talent pool but retain the white population increase, the Major Leagues since 1930 would have stagnated at best, as the native white U.S. population since then has barely increased over the rate of ML expansion: A little more than double vs. a little less than double.

But if all you're claiming is that the increase in the non-white population is more important than its inclusion per se, in terms of its effect on the quality of play, I grant the point---although to raise it still seems to beg the question of why the sina qua non of opening the door in the first place would not be considered more critical than the ensuing result (the flood of non-white talent), which would have been irrelevant without the initial act of inclusion itself. IOW first things first. I've run out of analogies to express this point, but it remains nevertheless.

Matt,

Excellent post as usual, but you should really provide us with some sort of a warning when you serve up such frightening links. This is supposed to be a civilized site.
   67. Matt Clement of Alexandria Posted: September 01, 2006 at 02:28 PM (#2164550)

Imagine it the other way. Suppose you barred players whose names started with A-M. Would the league be "twice as bad" in any statistical sense? I don't think it would. It would be worse, but not THAT much worse.
It would be worse exactly by the difference between Joe Mauer, Carlos Guillen, Barry Bonds and the rest of the excluded ballplayers, compared to the replacement-level N-Z players. We would know quite a lot about that. It would be a real exclusion of real people.
And remember my argument about what if blacks were prohibited from playing baseball at all? By your methdology, since Josh Gibson would never have existed as a ballplayer, there would be no adjustment necessary to Ruth's stats (since you can't assume more potential ballplayers = better baseball).
No, it would be a different kind of adjustment. Then there would be, I think, three kinds of adjustments.

Real, tangible adjustments: accounting for real baseball players who were barred from playing
More hypothetical adjustments: accounting for the moral abomination of a race-based rule preventing children from playing a game
Fully hypothetical adjustments: accounting for differences in overall population

What I'm saying is that you are conflating two kinds of adjustments for league quality - adjustments that deal with entirely real exclusions involving real people about whom we can know, and adjustments that deal with hypothetical questions of what types of ballplayers are produced by different sized populations.
   68. Phil Birnbaum Posted: September 01, 2006 at 05:37 PM (#2164793)
Andy,

"The simple way of putting it is that Rogers Hornsby didn't play against the best players of his day, while Robbie Alomar did."

So what? The difference between Hornsby and Alomar is not the talent they didn't face, but the talent they DID face.

You run a race in which blacks are not allowed to compete. You finish 356th. Then, nuclear war wipes out everyone on earth except you and 500 one-legged people of assorted races. You run another race. You finish first.

In the first case, you didn't play against the best runners of your day. In the second case, you did. Do you really want to argue that's all that matters, that your 356th place finish needs to be made to look WORSE to properly compare to your first place against a hundred one-legged hopping people? Because that's what you're arguing.

The "playing against the best players of his day" is competely irrelevant, except as a moral argument.
   69. Phil Birnbaum Posted: September 01, 2006 at 05:56 PM (#2164813)
Matt,

Whether the exclusions are "real" or not is irrelevant.

Look at my golf example (#62). Integrating added 50 people to the talent pool and lowered scores by 1. Demographics added 999,000 people to the talent pool and lowered scores by 17.

You would argue that when comparing Phil Mickelson to Babe Ruth, only the integration matters, because we can identify the 50 players who weren't allowed to play earlier. But we should just ignore the other 17 stroke improvement, because those 999,000 people weren't born in Babe Ruth's time, and are "hypothetical".

That's got no logic to it at all.

Suppose you had $1 as a kid, and later in life you have $100,000. Were you poorer as a kid than now? No, of course not, you will argue. The $99,999 difference didn't exist then -- the bills were all printed last week -- and so the difference is entirely "hypothetical". The only adjustment we should make is when the neighborhood bully stole 20 cents from you in grade 3, because we can identify the two dimes.

Come on.

If the population in Ruth's time had been the same as the population in Alomar's time, the level of pitching he faced would have been substantially better -- much more than the marginal improvement he would have seen if 1920s black players had been included. The fact that you can't identify these 200,000,000 hypothetical 1920s people does not mean their IMPACT is hypothetical.

Some day, the US will have 200MM more people than it does now. These people have no names yet. But we can figure out pretty well how much food they will eat, how much more housing they will take up, how many more hospitals and doctors we will need to treat them, and, yes, how much the quality of baseball will improve when there are now 500MM people fighting for roster spots instead of just 300MM.
   70. ERROR---Jolly Old St. Nick Posted: September 01, 2006 at 06:34 PM (#2164857)
Phil, I'm not making a "moral" argument at all, and I'm not talking about hypothetical races involving one-legged nuclear war survivors. I'll leave that one to the DAV.

Again, your whole argument about population increases being paramount rests on the assumption of integration.

While the raw numbers support the irrefutable argument that the gigantic increase in the non-white talent pool is the greater "cause" of the superior game on the field of today, those numbers wouldn't have any relevance at all without the prior policy of inclusion.

"The talent that [Hornsby and Alomar] DID face", as you put it, was first and foremost a choice of policy, and then and only then a byproduct of fertility rates. You can't have 100 grandchildren without the grandparents first having thrown away their birth control devices.

When you can't have B without first having A, it's hard for me to see how you can say that B is somehow more important than A. This has nothing at all to do with any "moral" issue, only with the natural sequence of events.
   71. ERROR---Jolly Old St. Nick Posted: September 01, 2006 at 06:50 PM (#2164877)
If the population in Ruth's time had been the same as the population in Alomar's time, the level of pitching he faced would have been substantially better

Not necessarily, if the competition were still restricted to whites, and if there were 30 teams instead of 16. The U.S. native white population from Ruth's time to ours has not outpaced the 30/16 ratio by all that much. I'd say that whatever quality jump there might have been would be more due to factors such as conditioning and nutrition than to population increases.

If you want to hypothesize a doubling of the population along with no expansion, then of course that's another story, but in that case the point would be so obvious it would scarcely be worth wasting words on.
   72. Matt Clement of Alexandria Posted: September 01, 2006 at 06:55 PM (#2164884)
Look at my golf example (#62). Integrating added 50 people to the talent pool and lowered scores by 1. Demographics added 999,000 people to the talent pool and lowered scores by 17.

You would argue that when comparing Phil Mickelson to Babe Ruth, only the integration matters, because we can identify the 50 players who weren't allowed to play earlier. But we should just ignore the other 17 stroke improvement, because those 999,000 people weren't born in Babe Ruth's time, and are "hypothetical".
But you made those numbers up. You seem to be citing the large, real number in an example as evidence, but you made up those numbers.

We don't know how population size affects the quality of the game. The quality of the game is impossible to measure particularly well. I have no idea whether the best white people of today are that much better than the best white people of the 20s - I tend to doubt the effect is that large.

One thing we can understand, that affects game quality, is the effect of integration and segregation. We know that Oscar Robertson and Pop Lloyd were banned, we know that Minoso and Robinson were banned for a time. These are tangible, historical facts. We place a lot more weight on them due to their reality.

It's not irrelevant at all. It's precisely material to the discussion - if you think that the effect of population on baseball is so concrete, why haven't you made it so? You're merely asserting that the effect is large, with no evidence, then grouping it with the relatively known effect of integration, in order to downplay the importance of integration. This is why I reacted harshly in the first place: you're downplaying the importance of segregation and integration without any facts. As such, your choice is arbitrary, and as such, it's political.
   73. Phil Birnbaum Posted: September 01, 2006 at 07:10 PM (#2164902)
Matt,

Well, OF COURSE I made those numbers up. I made those numbers up to show an obvious example of how an increase in population can overwhelmingly trump the effets of segregation, even though all of the population increase is invisible unless you look for it.

You write, "We don't know how population size affects the quality of the game." Well, that's deliberately keeping yourself ignorant to cling to your point.

We don't know the details, although we could probably estimate them, but it seems pretty obvious to me that the more people in the pool of potential players, the better the quality of play.

But maybe we'll have to agree to disagree. Here's what our disgreement boils down to:

Phil: When the pool of population from which baseball players doubles, the quality of play will increase substantially.

Matt: We don't know that's true, and so we have to assume the effect of doubling the population is exactly zero.

Matt, your whole argument hangs on that one point. I guess we'll have to leave it at that.
   74. Phil Birnbaum Posted: September 01, 2006 at 08:10 PM (#2164956)
Andy,

You write, “I'm not talking about hypothetical races involving one-legged nuclear war survivors.

That argument of mine was a serious point, which, I think, pretty much destroyed your entire “didn’t get to play against the best players of his day” argument. If you want to ignore the rebuttal and take your ball and go home, that’s fine. But trying to make me look silly by quoting it out of context in a mocking tone is kind of insulting.
Is it your lawyer training? When both the law and the facts are against you, bang on the table?

In any case, I’ll assume that you’re abandoning the “didn’t get to play against the best players of his day” argument and proceed.

You write, “again, your whole argument about population increases being paramount rests on the assumption of integration.”

Not exactly. It rests on the fact that integration actually occurred. The comparison between Hornsby and Alomar is actually between two comparisons:

1. How would Hornsby’s stats differ if he played in an integrated 1920s?
2. How would Hornsby’s stats change if he played in the same conditions as Alomar?

That’s how we tell if integration is a large effect when comparing Hornsby to Alomar. What’s the difference between Hornsby and integrated Hornsby, and what’s the difference between Hornsby and the actual Alomar?

The question of what integration did for Alomar is NOT relevant to this question.

”Those numbers wouldn't have any relevance at all without the prior policy of inclusion.”

This is true, but not relevant. The question ASSUMES the fact that integration occurred, and asks what part of the difference between the segregated Hornsby and the integrated Alomar is caused by integration. The “what if” applies only to Hornsby and not Alomar.

Put another way, the question asks the difference between a segregated Hornsby and today’s actual Alomar. What caused Alomar to be integrated is a completely separate issue.

Put a third way, if integration didn’t happen, the question would make no sense at all. You have to assume integration for the question to be meaningful in the first place.

I don’t know how else to explain this. Here, let me try nuclear war so you can ignore it. :)

“If the US had not entered WWII, the Germans would have won and might have nuked the entire country, in which case there would be no Robbie Alomar. Therefore, these numbers would have no relevance at all without the prior policy of entering the war. Therefore, it’s not demographics that led to there being twice as many potential ballplayers in 1960 as 1920 – it’s warmongering.”

“I don’t know how you can say warmongering is not relevant. When you can’t have demographics without first going to war, it’s hard for me to see how you can say that war is somehow more important to the Alomar question than demographics.”


In other news, one of the important reasons home run rates are up in the last decade is that Lyndon Johnson didn’t ban baseball in 1966.

My point is that yes, Alomar wouldn’t exist without integration, but that does not mean integration is materially responsible for playing conditions. There is an infinity of things that baseball wouldn’t exist without, like Lyndon Johnson’s benevolence, the existence of food on this planet, and trees to make bats out of. This doesn’t make them relevant. It just so happens that because we’re talking about integration, it SEEMS like it’s relevant in this context. But it’s not. The "but for" test doesn't work.

I'm having trouble phrasing this ... perhaps someone else can explain this properly, someone who took a philosophy course on cause and effect (or probably someone who's just smarter than me). On this one, my only proof is to refute your principle with obvious counterexamples, which you probably don’t want to take seriously.

Alternatively, maybe you can try answering the question in some way instead of just criticizing my answer, so that we can flesh out your principle. Here’s the question:

What method should we use to adjust Hornsby’s career to make it comparable to Alomar’s? How much of that adjustment is due to the fact that Hornsby played in a segregated era while Alomar didn’t, and how much of it is due to other factors?

My answer, which you have already seen, is look at the pool of eligibles in both eras, and compare the differences. The difference between 1920 segregated and integrated is the effect of segregation; the difference between 1920 integrated and 2005 actual is the effect of demographics. Compare the two.

Now, I know you disagree with my method and conclusions. What’s your method and conclusion?
   75. Phil Birnbaum Posted: September 01, 2006 at 08:23 PM (#2164970)
Wait, Andy, I think I see what's going on.

I'm asking this question: if we discount Hornsby's career to compare to Alomar's playing conditions, how much of that discounting is due to adding in that era's black players, and how much to other causes?

You're asking this question: if we augment Alomar's career to compare to Hornsby's playing conditions, how much of the augmentation is due to the elimination of today's black players, and how much due to other causes?

These are two different questions, with answers that could, in theory, be very different. (And only in the second case is today's racial proportion relevant to the answer.)

Am I right about which question you're answering?
   76. Phil Birnbaum Posted: September 01, 2006 at 08:45 PM (#2164990)
Sorry, Andy, I misspoke ... it was Sam who's the law professor, not you. Apologies to Sam too.
   77. Srul Itza Posted: September 01, 2006 at 09:01 PM (#2165003)
Phil:

Just wanted to say I admire you for wading into the debate, and not being put off by the typical "this guy is a moron" posts.

I also agree with what you are saying. There is a lot of "talking past each other" going on this thread. So, nu, what else is new?

The problem that you have pointed out is a subset of one that has been discussed in numerous other threads -- the host of great imponderables that make it difficult to compare players across eras, and make it difficult to determine the level of play as compared from one era to the next.

Leaving aside chemical assistance, the fact is that improved sports medicine and training, and the ability to reach a larger portion of the planet, has resulted in continual increases in various measures of athletic ability (strenght, speed, etc.). Exactly how much -- to the fourth decimal point, if you please -- does that translate to baseball? A thousand imponderables.

The fact is that, because of increased U.S. and global wealth, and increased global travel possibilities, along with increases in world population, the potential pool of world-class athletes in has increased markedly. Exactly how much -- to the fourth decimal point, if you please -- does that translate to baseball? A thousand imponderables.

How is that affected by different opportunities now available -- e.g., careers in other sports? How is that affected by the advancement of various social groups who now no longer see baseball, or athletics, as a route out of poverty or out of small town America -- factors which now, instead drive players in places like the Dominican Republic?

The end of the color bar SHOULD have been different, in one way: The suddenness of the influx, because of the immediate availability of a large number of highly talented, Major League ready talent. In FACT, it did not happen that way, because even after Jackie Robinson's debuts, many teams were still reluctant to hire African American players, and/or deliberately chose to limit the number of African American players on their rosters. Many people here (and I don't doubt them, I just have not studied up on this myself) have posited that the National League's greater willingness to accept such players created an imbalance between the leagues, but even taking that into account, the inclusion of African American players into MLB was actually far more gradual than the available talent would seem to have mandated.

Coming back to the original premise -- lack of complete integration is just ONE factor in adjusting the playing conditions of Babe Ruth to the playing conditions of Ted Williams to the playing conditions of Barry Bonds. We can ASSUME, with a reasoanble degree of certitude, that if Babe Ruth had played in a 100% integrated league, but still limited largely to United State players, his statistical dominance would not have been the same. Trying to put a number on it, though -- much less translating it to today's Globally infused, NBA/NFL/Expansion team deleted, jet travelling, night-game playing, diffently equipped time -- is probably beyond our abilities. At least to the four decimal place level.
   78. ERROR---Jolly Old St. Nick Posted: September 01, 2006 at 10:36 PM (#2165057)
Phil, the more I read what you're saying, the less sure I am that we're really disagreeing all that much. I think it simply boils down to a semantic difference: I don't see how a consequence (the flood of non-white talent into the Majors caused by the population explosion) can be more important than the enabling cause (integration), since the consequence couldn't have happened (or wouldn't have been relevant) independently of the cause. You say that we have to take integration as a given and proceed from there, and that in doing so we can see that the most important factor in baseball's improvement has been the numerical explosion in the now-integrated talent pool.

As you note above, these aren't incompatible positions.

Now that that's out of the way,

What method should we use to adjust Hornsby’s career to make it comparable to Alomar’s? How much of that adjustment is due to the fact that Hornsby played in a segregated era while Alomar didn’t, and how much of it is due to other factors?

My answer, which you have already seen, is look at the pool of eligibles in both eras, and compare the differences. The difference between 1920 segregated and integrated is the effect of segregation; the difference between 1920 integrated and 2005 actual is the effect of demographics. Compare the two.


Rather than try to formulate some specific methodology, I'll give an empirical guess as to the effects of integration and population on Hornsby.

If baseball had incorporated the Negro Leagues of 1920 into its existing 400 roster spots, I'd guess that after a period of shakeout, remedial coaching, and the end of holdout teams, the top 50 to 100 black players would have replaced the worst 50 to 100 white players. I base this on what happened to the Majors between 1947 and the late 1960's. An important assumption of mine is the elimination of the sort of quota systems you had for the first 20 years or so of formal "integration." Without this elimination there isn't really a true test. I added the line about remedial coaching because in many cases the raw black talent hadn't been exploited to its fullest, and I am going on the (somewhat utopian) assumption that the Majors would want to make efficient use of its newly discovered material, as opposed to giving it lip service, as did many teams through the 60's.

Given this set of assumptions, I'd imagine that Hornsby would have wound up somewhere around .330 to .340 rather than .358. If OTOH a more limited form of integration had taken place, he might have been closer to .350.

And if Hornsby played today? More like maybe .310 to .320. And from this you can infer what you want about the relative effects of integration and population base. Hornsby was a great, great talent, but I'd have a hard time believing that he could translate that into Albert Pujols or or Wade Boggs numbers against recent competition.

Of course all of this is based on little more than guesswork, hunches, and a mix of thoughts about the evolution of talent in other sports combined with the uniqueness of baseball skill sets, specifically the batting eye. As others have said, there are almost too many variables to consider.

I'm leaving work now, but before I go I'd like to throw a question back to you:

What sort of improvement would you see between 1920 and 2006, given the growth of population, if the Major Leagues had remained segregated, and had expanded to 30 teams?

I realize that this is an almost absurd question in real life, since it either supposes an improbable maintenance of a Jim Crow society or a flukish solitary holdout of baseball, an equally absurd thought. But try the exercise anyway, using whatever extra assumptions you wish. My one guess would be that pitching would be far more dominant than it is in the Majors of 2006. And in a league like that, I'd guess that a transplanted Hornsby would hit between .340 and .350. Not as good as the historical Hornsby in his Jim Crow league, but better than a Hornsby in a hypothetical integrated Major Leagues of his time---and way, way better than a Hornsby in the real Major Leagues of today.
   79. Eraser-X is emphatically dominating teh site!!! Posted: September 01, 2006 at 11:45 PM (#2165105)
Phil, thanks for responding and clarifying. If that is the tack you are taking I would agree with you in an absolute sense.

Where I think MCA and Andy's perspectives are very instructive is when you consider how people actually evaluate ability and why they are interested in MLB.

There are almost always some people who go on and on about how we should stop paying to see ballgames and just go watch Little League where the game is "purer", whatever that means. But people don't want to see bad "pure" baseball, they want to see the best baseball available.

That doesn't mean the baseball we will perfect in the year 2904 when Mike Marshall XXX discovers how baseball can generate a field of infinite happiness and properity and everyone decides to play baseball. It doesn't mean having races after a nuclear war.

It means "Who are the best players on the planet and where do they play?"

That's why something like segregation is so disconcerting--it's a lot different from people having the chance to play baseball and deciding to become a professor instead. It's a conscious attempt to avoid having the best players in the world on the same field at the same time.

And it doesn't have to just be direct segregation--prejudice and cultural miscommunications can create similiar dynamics.



That's why the greatest baseball in the history of the world was played this spring. It wasn't the highest playing level, and not all of the best players actually did play.

But for the first time in the history of the world, pretty much anyone who wanted to compete with the best players in the world on the same field could do so and do so without having to adjust to specific cultural mores or conventions.

And what did we learn?
1) Small sample size is a #####.
2) At least a few of the world's best players don't play in MLB.
3) Baseball is a sick, sick, game! (that's like Michael Jackson's "bad" ;))

So we aren't going to yell at kids who go to law school and demand that they play baseball to improve the level of the game, but we can do everything in our power to allow people who want to play baseball at the highest level and the skills to do so to partake in that?

Ending note: I realize the WBC does have a sort of exclusionary policy in terms of the fact that entries are limited by country (sort of), so that's an issue to contemplate as well :)
   80. Phil Birnbaum Posted: September 02, 2006 at 06:45 PM (#2165810)
Hi, Andy,

You say, "[With integration] I'd imagine that Hornsby would have wound up somewhere around .330 to .340 rather than .358. ... And if Hornsby played today? More like maybe .310 to .320."

Okay, let's say .335 and .315. That means that you think integration was about 23 points, and everything else another 20 points. Correct?

"And from this you can infer what you want about the relative effects of integration and population base."

Well, what do *you* infer? Doesn't your estimate seem to imply that other factors are almost as important as integration?

Or are you arguing that even the second 20 points are due to integration, since they wouldn't have happened without it?

The reasoning is more important than the numbers.

You then ask me, "What sort of improvement would you see between 1920 and 2006, given the growth of population, if the Major Leagues had remained segregated, and had expanded to 30 teams?"

Well, you didn't answer my question, but I'll answer yours. Actually, I don't know what improvement I'd see, but in terms of the talent pool, the US population went up about 160% between 1920 and 2000, while the number of roster spots increased by only 88%. 260 divided by 188 is 1.38. Assuming the white population grew at the same rate as the US as a whole, I would conclude that the talent pool per roster spot is 38% larger now, and so there would be a material improvement in quality of play.

Which is still irrelevant to the question of how to interpret Rogers Hornsby's stats. I anticipate, though, that you'll argue that anything over 38% should be considered the result of integration, and counted against Hornsby.

I still say that's incorrect. Here's one more parallel, this time a much closer one.

A young man called Ron L. earns a few bucks a week doing odd jobs. Unfortunately, he turns to crime and gets caught. In jail, he makes license plates for $1.25 an hour. While in prison, though, he discovers he makes a superb baseball player. On his release, he signs with the Tigers and earns $1,000,000 a year.

What is the "more important" cause of the increase in Ron L.'s salary? I argue that it's the baseball talent. You are arguing that it's prison, because if he hadn't been released from prison, he never would have been able to make that much money. I am unable to convince you that it's the baseball talent. You make your point by asking me what the quality of play would be like if all former criminals were still in jail. I agree that it wouldn't change, but that the question is irrelevant.

So we're stuck.

How can we test our respective theories? You arrange to go to jail, eventually get released, and see if you make more money. I'll develop as much baseball talent as Ron L., and see if I make more money. That should settle the question about what's the "more important" cause of the pay increase.

Same for our little debate. You take over the National League. Institute segregation for a decade, then apologize and recind it. I'll introduce baseball to countries that don't play it, draft their best players, and see if I get an increase in quality of play. Then we'll see whose quality of play is better, and which factor is more important.
   81. Phil Birnbaum Posted: September 02, 2006 at 06:49 PM (#2165812)
Saul (#77),

Thanks for the comments! I agree with what you're saying.

Just to clarify that I am not trying to quantify the impacts of segregation and demographics (although I might try to do an estimate in the future, just for fun). All I'm trying to do is point out that segregation is only a (probably small) part of the difference between then and now.
   82. Phil Birnbaum Posted: September 02, 2006 at 07:04 PM (#2165829)
Eraser-X (#79),

"Where I think MCA and Andy's perspectives are very instructive is when you consider how people actually evaluate ability and why they are interested in MLB."


I agree with that 100%. Obviously, I have been arguing that MCA and Andy's arguments are fatally flawed, and that, as a fact, the color bar was responsible for only a small part of the lower quality of play compared to today.

But you're absolutely right -- we learn a lot about what MCA and Andy (and other posters) think is important. I'm right there with Andy when he says that the players of the segregation era didn't have a chance to play with the best players of their time. Andy said that is indeed a tragedy, for both the barred players and, to a lesser extent, the players who were deprived of the opportunity to play against them. I couldn't agree more.

I have been arguing that segregation is a small part of the dilution in quality of play between then and now. But segregation is a very large part of the dilution in RESPECT for the quality of play between then and now. And you, and Andy, and MCA, have given excellent reasons why.
   83. Srul Itza Posted: September 02, 2006 at 07:22 PM (#2165862)
Thanks for the response Phil. By the way, it's Srul, not Saul. If your bubbe is still around, she might recognize it as a yiddish diminutive for Yisrael.

I did not think you were trying to quantify. I think it is impossible to do so, because there are just so man factors on either side. I think I agree with you that, the effect of disqualifying 13% of the American population + X% of the Latin American population from even having the chance to play baseball, may be small in comparison with (60 years of population growth + changes in diet and training + changes in equipment + changes in sports medicine + the effects of globalization) - (loss of potential players through rising income and greater economic opportunites for certain social groups + greater popularity of other sports opportunity). There are, of course, numerous other factors I have omitted.
   84. ERROR---Jolly Old St. Nick Posted: September 02, 2006 at 07:45 PM (#2165885)
You say, "[With integration] I'd imagine that Hornsby would have wound up somewhere around .330 to .340 rather than .358. ... And if Hornsby played today? More like maybe .310 to .320."

Okay, let's say .335 and .315. That means that you think integration was about 23 points, and everything else another 20 points. Correct?

"And from this you can infer what you want about the relative effects of integration and population base."

Well, what do *you* infer? Doesn't your estimate seem to imply that other factors are almost as important as integration?

Or are you arguing that even the second 20 points are due to integration, since they wouldn't have happened without it?


First, do note that the premise of that .330 to .340 figure was that the integration wasn't token (a la the pre-1960 NL and the pre-1970 AL), but real. Without that, I said it would have been closer to .350.

But to your question about ranking the two factors: My answer would be "depending on what you mean by 'more important.'"

Obviously no integration, therefore no non-white population in the majors, population rise or no population rise. So in that sense I still say that integration was more important.

But using your premise, and strictly by the numbers, of course the numerical expansion of the talent pool would be critical. I've never argued otherwise, only that since integration was required to unlock the talent box, integration was the key factor.

Which is why I said in the last post that this is really just a matter of semantics. The rest of your post just re-iterates this. Once your premise is accepted, I don't have any quarrel with your point about the supremacy of population increases. The exact numbers are really beside the point, as they are based largely on hunches, and incorporate other factors such as training and coaching advances.

But if you keep all those non-white players locked up in jail, unless you start teaching baseball to a lot of Europeans and Australians, and however many Castilians you can find in South America, you're going to have a pretty hard time goosing up the talent level of the Major Leagues all that much over Babe Ruth's time. There just aren't that many white people outside the U.S. who play baseball.

If OTOH you want to pose a theoretical world where all the world's white people have grown up with baseball the way they do in Southern California, then of course the quality of play would grow proportionately---as long as further expansion didn't mirror this rise in the talent pool.

And if Ron L[eFlore] is out of jail but still stuck in a Jim Crow society, he ain't making no million dollars playing baseball, talent or no talent.
   85. Matt Clement of Alexandria Posted: September 02, 2006 at 10:23 PM (#2165969)
I have been arguing that segregation is a small part of the dilution in quality of play between then and now. But segregation is a very large part of the dilution in RESPECT for the quality of play between then and now. And you, and Andy, and MCA, have given excellent reasons why.
But you haven't cited a lick of evidence for that. You have stated the increase in population, and other factors, but you haven't made the slightest effort to quantify them. This is what I meant by arbitrary, above. As best as I can tell, you feel that integration is a smaller factor. That's fine, as a feeling. But if you're going to heap condescension on those who disagree, as you did in the original article, you need slightly more than a feeling, and made-up numbers in a hypothetical.
Phil: When the pool of population from which baseball players doubles, the quality of play will increase substantially.

Matt: We don't know that's true, and so we have to assume the effect of doubling the population is exactly zero.
This is just insulting. Come on.

I said that we don't know what the effect of increasing the population by 3-4 times is, as balanced by the expansion of the leagues. Is the effect of population so "substantial" that it dwarfs the effect of integration? Maybe. I don't know. You don't know. No one has cited any evidence in either direction. Only one of us - that'd be you - has asserted that people who think that integration was a big deal are just doing what feels good.

Given that the effect of segregation is relatively knowable and concrete, we place a set amount of weight on it. The effect of population is practically unknowable and not at all concrete, and so we, quite reasonably, don't make major assumptions about the size of the effect. You have chosen to make the assumption that the size of the population effect dwarfs integration. I think that's unsupported by evidence. I think that common sense has a lot going for it here - let's focus on what we know, and not get too trapped in the hypotheticals.
   86. Matt Clement of Alexandria Posted: September 02, 2006 at 10:36 PM (#2165978)
But using your premise, and strictly by the numbers, of course the numerical expansion of the talent pool would be critical. I've never argued otherwise, only that since integration was required to unlock the talent box, integration was the key factor.
This is a good point. Insofar as Phil is talking only about what adjustments we make to Babe Ruth's statistics, this isn't a big deal. Insofar as we're talking about the effects of integration on baseball, it's important to note that integration was in fact a major causal factor in the increase of hte population of possible ballplayers. Phil's argument goes only to one aspect of the historical debate over the effects of segration on the quality of play in baseball.
   87. Phil Birnbaum Posted: September 03, 2006 at 04:00 AM (#2166171)
Hi, Matt, (#85)

But you haven't cited a lick of evidence for [your conclusions]. You have stated the increase in population, and other factors, but you haven't made the slightest effort to quantify them.


Huh? I have quantified them repeatedly. I'll do it again:

1. The proportion of blacks in MLB was highest at 27%. It is now 9%. I will assume that in Rogers Hornsby's time, the proportion of blacks would have been 20% if not for integration. (Feel free to suggest a different number if you prefer.) That means that the talent pool of players capable of playing at the actual level of Hornsby's time was 20% higher than actual.

2. By 2000, the population of the USA had grown to 265% of its original level. That means the talent pool of current players capable of playing at the actual level of Hornsby's time is 165% higher than it was then. (Feel free to suggest a different number if you prefer. I was under the impression that you were suggesting ignoring this factor (and calling it zero) because you didn't know how to quantify it, but you called that insulting. My apologies for the insult. What's your number, and what's your logic?)

3. Feel free to choose an arbitrary increase in talent pool of non-US players, which was miniscule in 1920 but is very large now. The population of the US was 100MM in 1920. The population of non-US countries where players are scouted is probably more than 100MM. But we'll leave this out for now.

So: increase in talent pool caused by segregation: 20%.

Increase in talent pool caused by population: 165%. Increase in number of roster spots: 88%. Therefore, increase in talent pool per roster spot: 40%.

So, even without considering non-Americans, population appears to have twice the impact of segregation.

I do not "feel" that integration is a smaller factor. It is a conclusion I draw from the sizes of the talent pools.

Now, you write,

The effect of population is practically unknowable and not at all concrete, and so we, quite reasonably, don't make major assumptions about the size of the effect.


As I wrote, I don't think it's "practically unknowable" at all. I think it is reasonable to assume that the talent pool is directly proportional to the population. Furthermore, I think it is incumbent on you to give reasons why it might be larger or smaller, and how much so, if you think otherwise.

But I point out that ANY REASONABLE ASSUMPTION about how talent pools affect the caliber of play will lead to a conclusion that segregation is a minor factor.

There were 10 million blacks in the US in 1920. For segregation to be even EQUALLY as impactful as demographics, you have to assume that there was as much baseball talent among ten million 1920s blacks as 37.5 million people in 2000 (7.5 million of whom would be black). Because that's the ratio: almost 4 to 1.

And that doesn't factor in non-Americans, unscouted then, heavily scouted now. Add those in, and what ratio do you wind up with ... 5 to 1? 8 to 1? 10 to 1?

So I can see why you don't want to address the demographics issue and why you want to "stick to what we know" -- because you would have to make patently absurd assumptions for your conclusions to stand up. Better to pretend the question can't be answered, and call me condescending for trying to make reasoanble estimates.

You claimed I insulted you by saying you thought the effect of demographics was zero. Well, okay, you never said that explicitly. But by demanding we ignore the question, you're doing the same thing. "How much does an elephant and a hen weigh? Well, the elephant won't fit on my scale. And it's absurd to think you can estimate it. Let's stick to what we know, which is the hen's weight. Therefore, only the hen is important. Conclusion: an elephant and a hen weigh three pounds."

This is not intended as sarcasm ... that's actually the way I see your argument working.

In any case, I will rephrase our difference:

Phil: When the pool of population from which baseball players are taken doubles, it is a good default assumption that the number of men who meet a certain talent level will likely also double. Therefore, demographics is at least as important as segregation.

Matt: We don't know that, I don't accept that doubling is a reasonable assumption, and I remain so agnostic on the question that I don't believe we should even estimate it. Only what we know counts. Therefore, only segregation is important.

Is this now a fair summary of our differences? If you want to agree to disagree on this, and leave it at that, it's fine with me.
   88. Phil Birnbaum Posted: September 03, 2006 at 04:04 AM (#2166175)
Srul (#83),

Oops, sorry about that. The only Yisrael I knew was just "Yisrael". Although my mother sometimes calls me "Pinyaleh."
   89. Eraser-X is emphatically dominating teh site!!! Posted: September 03, 2006 at 04:30 AM (#2166190)
1. The proportion of blacks in MLB was highest at 27%. It is now 9%. I will assume that in Rogers Hornsby's time, the proportion of blacks would have been 20% if not for integration. (Feel free to suggest a different number if you prefer.) That means that the talent pool of players capable of playing at the actual level of Hornsby's time was 20% higher than actual.


This is super-picky and not really relevant, but wouldn't the talent pool be 25% greater if blacks would have made up 20%?


If the potential talent pool was only 20% greater than actual, it would mean that blacks only comprised 16% of the potential talent pool (blacks included).

I'm not too good at math, so if I'm wrong, let me know...

Back to the substance of your argument...


You claimed I insulted you by saying you thought the effect of demographics was zero. Well, okay, you never said that explicitly. But by demanding we ignore the question, you're doing the same thing. "How much does an elephant and a hen weigh? Well, the elephant won't fit on my scale. And it's absurd to think you can estimate it. Let's stick to what we know, which is the hen's weight. Therefore, only the hen is important. Conclusion: an elephant and a hen weigh three pounds."


This is where I don't think you are being fair because you choosing to deal with absolutes and not functional significance and MCA's argument, from my appraisal seems to be more about functionals.

Basically, very few people even are interested in absolute baseball talent. In fact, think about that for a minute--without human assumptions of value being introduced to the game, what is it but just a game with a ball and stick?

In the world we've created where baseball is not only meaningful, but completely imbued with the beliefs and values of our society, we don't care about absolute talent. We care about seeing the best ballplayers in the world playing ball together.

This might not be utterly true. Some might ponder the loss of talent we've had based on the number of abortions in the country.

But in general, we don't think, "Pujols is good, but only 17% as good as he could have been if my father had developed a serum to allow human beings to grow an extra pair of arms!"

We assume that we are dealing with players who exist, not players who could have existed. I believe that's even clear in the case of the Negro League players. Most people who have considered the NLe issue are upset by the lack of opportunity for great players to compete. But I've seen few studies or arguments that seek to grant Negro League players added credit for their tougher playing conditions or smaller talent pool that was drawn upon for their coaches, managers, and executive staff.
   90. Phil Birnbaum Posted: September 03, 2006 at 05:15 AM (#2166212)
Hi, Eraser-X,

First, for the math question, you're right that if the talent pool of blacks was 20% of that of whites, there would be 25% as many blacks as whites in the integrated league. But that's not the number I'm concerned with ... I'm asking how many more players there are capable of playing *segregated ball* if you include blacks. And that answer is the 20%.

Hope that made sense, I may not have explained it right.

Second, you say

But in general, we don't think, "Pujols is good, but only 17% as good as he could have been if my father had developed a serum to allow human beings to grow an extra pair of arms!"


Then why are we thinking, "Ruth [statistically] was good, but 105% as good as he would have been if baseball had been integrated?"

And if we can think that, what's wrong "Ruth was good, but 115% as good as he would have been if baseball had drawn from today's talent pool?"

[Note: numbers above are made up.]

Finally, to be honest, I don't understand what you mean about absolutes and functional significance, so I can't comment on that.
   91. Eraser-X is emphatically dominating teh site!!! Posted: September 03, 2006 at 06:34 AM (#2166225)

Then why are we thinking, "Ruth [statistically] was good, but 105% as good as he would have been if baseball had been integrated?"

And if we can think that, what's wrong "Ruth was good, but 115% as good as he would have been if baseball had drawn from today's talent pool?"


Because the first scenario could have easily become reality. The players existed and they were ready to play. At any point, someone could have said "C'mon Satchel, why don't you start tomorrow?"

The second scenario does not constitute a scenario that is remotely plausible with human intervention. We can transport Babe Ruth in our minds or our computers, but if we don't, we have lost nothing concrete. There were no games that Ruth could have played against Pedro Martinez that we missed out on.

When we consider that scenario, we have to start answering questions that seem literally unanswerable like, "How would Babe Ruth have utilized modern conditioning?" "How would Babe Ruth have responded to the fact that he may have had increased access to different drugs?" or even, "Would Babe Ruth have ever even played a game of baseball in the modern age?"



Another interesting divergence is the importances of accurate record-keeping and the arbitrary decision of authorities of what constituted MLB. The main reason why we don't consider the NLe a major league is because there are not comprehensive formal records of each game and no one before has called it a major league. Obviously which reason is more weighty differs from person to person.

But why do we consider the American Association or the Union Association major leagues, but not NPB or the Korean Leagues or the NNL?

What if we find a dusty old scorebook that meticulously grants us accurate data of the '27 Ruth/Gehrig tour where they integrated some Asian American stars into their teams, is that a major league?

I'm sincerely asking--What makes a major league? Is it the product on the field? Or is it the fact that there is someone, somewhere keeping score and preserving the data? Or is it whether or not there is some suit thousands of miles away administering the teams in an orderly way?




Oh, and this was the math I did basic on your scenario:

Segregated league=.8 Integrated League
So Integrated league=1.25 Segregated League.

So there's 25% more talent in the Integrated League?
   92. Phil Birnbaum Posted: September 03, 2006 at 02:53 PM (#2166298)
Eraser-X,

I meant

Integrated League=1.2 Segregated League.
   93. Matt Clement of Alexandria Posted: September 03, 2006 at 03:20 PM (#2166310)
As I wrote, I don't think it's "practically unknowable" at all. I think it is reasonable to assume that the talent pool is directly proportional to the population. Furthermore, I think it is incumbent on you to give reasons why it might be larger or smaller, and how much so, if you think otherwise.
But as I showed above, it clearly isn't, unless you want to argue that Babe Ruth played in the functional equivalent of the Carolina League. I don't think it's directly proportional at all. I don't know what the relation is. You think this is a far simpler question than I do.

I think this is the core of our dispute, actually. You think that "directly proportional to population" is a reasonable answer. I don't. You already agreed with me that the 1925 AL isn't the 2005 Carolina League, so I'm not sure anymore what you mean by "directly proportional".

How much better is baseball now than in 1925? Please explain how you use the proportional increase in population to estimate this. Please tell me which contemporary baseball player is statistically equivalent to Babe Ruth under these adjustments. As best as I can estimate, Babe Ruth's stats come out somewhere in the range of maybe Gary Sheffield, at best, based on any "proportional to population" theory of league quality. I find that very unlikely.

You continue to use analogies which are gratingly condescending, and which assume your position is right before the analogy has been constructed. It's a very off-putting debating technique. If we were only measuring the total population of possible ballplayers, then elephant weight might be applicable. But we're not. We don't know the relation of size of population to league quality. Your assumptions on the topic seem to me to lead inexorably to the conclusion that Babe Ruth's stats are relatively unimpressive compared to today's players.
   94. Phil Birnbaum Posted: September 03, 2006 at 03:57 PM (#2166325)
Matt,

But as I showed above, it clearly isn't, unless you want to argue that Babe Ruth played in the functional equivalent of the Carolina League. I don't think it's directly proportional at all. I don't know what the relation is. You think this is a far simpler question than I do.

As I said in post 65: Four times the population doesn't necessarily mean four times the quality of play. The quality of play will be higher, but not linearly.

I argue that the talent pool is proportional to population. Quality of play increases, but not necessarily proportionally, with talent pool.

Let me try this one more time.

If you integrate the 1920s, you add the best baseball players out of 10,000,000 American blacks. You get about 50 to 100 players, say. Maybe a couple of superstars, a few stars, a good number of everyday players, and some reserves.

If you increase the US population from 1920 to 2000, you get the best baseball players out of 150,000,000 Americans. You get fifteen times a couple of superstars. You get fifteen times as many stars. You get fifteen times "a good number" of everyday players. You get fifteen times some reserves.

After accounting for the increased number of teams, you still get many, many more players than if you simply integrate the 1920s.

Make any assumptions you want about the MAGNITUDE of improvement: the fact is, adding more players makes for a better league than adding fewer players.
   95. Phil Birnbaum Posted: September 03, 2006 at 04:03 PM (#2166328)
And, Matt,

You continue to use analogies which are gratingly condescending, and which assume your position is right before the analogy has been constructed. It's a very off-putting debating technique.


I have no idea why you would find my analogies condescending. They are deliberately made vivid to more obviously illustrate what I believe are the logical fallacies in your argument. May I respectfully suggest that you might be missing my points?
   96. vortex of dissipation Posted: September 03, 2006 at 09:26 PM (#2166664)
Interesting thread. One question.

A Sandy Koufax or Bob Gibson never had to face Sadaharu Oh or Shigeo Nagashima. However, a Johan Santana or Roger Clemens has to face Ichiro or Hideki Matsui. Should we discount the achievements of a Koufax or Gibson because of that? If the argument is that Ruth did not face all of the best baseball players active during his time, wouldn't that be equally true of Koufax and Gibson?
   97. ERROR---Jolly Old St. Nick Posted: September 03, 2006 at 11:20 PM (#2166740)
Interesting thread. One question.

A Sandy Koufax or Bob Gibson never had to face Sadaharu Oh or Shigeo Nagashima. However, a Johan Santana or Roger Clemens has to face Ichiro or Hideki Matsui. Should we discount the achievements of a Koufax or Gibson because of that? If the argument is that Ruth did not face all of the best baseball players active during his time, wouldn't that be equally true of Koufax and Gibson?


I'd call that a very good question, since Gibson and Koufax both played in a league which was somewhat in a halfway stage between the 1920's and 2006, from two standpoints.

First, it was roughly the chronological midpoint.

Second, the National League of the mid-late 1960's was almost fully integrated with respect to African Americans, but only beginning to get the influx of worldwide talent in the quantity that we have today. And WRT to Phil's point, the total population pool was somewhere in the midpoint as well, once you factor in expansion.

What also makes it interesting is that while many of us (well, some of us) have vivid memories of Koufax and Gibson, none of us likely saw Ruth play at his peak. This is bound to color some of our opinions.

Putting this all together seems nearly impossible to me, especially if you lower the pitcher's mound and introduce all that extra weight training, even discounting steroids.

Part of me has to assume that Koufax and Gibson's stats would suffer from having to face the added roster depths of today, and from the smaller parks, the better conditioning, and (yes) by the big expansion of the worldwide population, which has to raise the quality of the game.

But then the part of me which saw Koufax and Gibson pitch dozens of games, and absolutely dominate the best American League teams nearly every time they stepped on the mound---that part of me has a hard time believing that they wouldn't do just about as well today as ever.

And trying to reconcile these contrasting hunches leads me down the dread road of intangibles. The intangible factors of spirit and determination which always seem to be found in supersized quanitities in the superstars of every era. The factors which always seem to allow them to raise their games to meet the quality of the opposition, as Koufax and Gibson always did in the World Series.

Which leaves me in a quandry: Objectively, you have to discount their statistics. They didn't play against the level of talent (on average) that they'd have to face today.

But if they had to face the best teams today with the money on the line, I wouldn't bet a plugged nickel against them.

Or, to be honest, against Babe Ruth, either.
   98. Phil Birnbaum Posted: September 04, 2006 at 12:52 AM (#2166804)
Hi, Vortex,

For my part, yes, I agree with you that you do have to take that into account.
   99. Spencer Benedict Posted: September 04, 2006 at 01:35 AM (#2166832)
Part of me has to assume that Koufax and Gibson's stats would suffer from having to face the added roster depths of today, and from the smaller parks, the better conditioning, and (yes) by the big expansion of the worldwide population, which has to raise the quality of the game.

Gibson and Koufax would be better conditioned as well. They would probably also pitch about 220 innings per year.
   100. ERROR---Jolly Old St. Nick Posted: September 04, 2006 at 01:48 AM (#2166845)
Gibson and Koufax would be better conditioned as well. They would probably also pitch about 220 innings per year.

Hard to imagine either one of those two pitchers being in any better condition than they were back then. They were both superb physical specimens with little if any visible excess weight. And it's not easy to condtion arthritis out of your pitching elbow.

Your 220 innings estimate, OTOH, seems about right, especially if Koufax had been consciously trying to reduce his IP in order to preserve his arm.

You must be Registered and Logged In to post comments.

 

 

<< Back to main

BBTF Partner

Dynasty League Baseball

Support BBTF

donate

Thanks to
ERROR---Jolly Old St. Nick
for his generous support.

You must be logged in to view your Bookmarks.

Hot Topics

NewsblogMets targeting A’s Billy Beane as candidate to run baseball operations
(2 - 9:33pm, Sep 19)
Last: Walt Davis

NewsblogHey, pal, WEEKEND OMNICHATTER's eyes are up here, for September 17-19, 2021
(94 - 9:28pm, Sep 19)
Last: Random Transaction Generator

NewsblogRemorseful Thom Brennaman deserves to work again now; He’s already suffered 14 months of hell
(47 - 9:25pm, Sep 19)
Last: Howie Menckel

NewsblogNBA 2021 Playoffs+ thread
(4431 - 9:21pm, Sep 19)
Last: tshipman

Sox TherapyIt's A Sprint, Not A Marathon
(14 - 9:13pm, Sep 19)
Last: pikepredator

NewsblogEmpty Stadium Sports Will Be Really Weird
(14003 - 8:51pm, Sep 19)
Last: Mayor Blomberg

NewsblogManny, Tatis have heated dugout exchange
(5 - 8:09pm, Sep 19)
Last: Howie Menckel

NewsblogShohei Ohtani has sore arm, might not pitch again this season
(34 - 7:26pm, Sep 19)
Last: Hombre Brotani

NewsblogBest average in Majors ... but no batting title?
(23 - 5:50pm, Sep 19)
Last: Walt Davis

NewsblogOT Soccer Thread - Transfer! Kits! Other Stuff!
(272 - 5:02pm, Sep 19)
Last: AuntBea odeurs de parfum de distance sociale

NewsblogCan Milwaukee’s Corbin Burnes Deliver MLB’s First Sabermetric Pitching Triple Crown?
(1 - 4:58pm, Sep 19)
Last: Tim M

NewsblogBaseball Pension Tension
(16 - 2:09pm, Sep 19)
Last: Ziggy: social distancing since 1980

Newsblog8th? BBTF Central Park Softball Game: SEPT 18, 2021
(224 - 12:19pm, Sep 19)
Last: Howie Menckel

NewsblogThe Brown Bombers, a Little-Known Black Baseball Team, Shook Colorado’S Segregated Sports World
(3 - 9:33am, Sep 19)
Last: kirstie819

NewsblogOT - August/September 2021 College Football thread
(202 - 2:30am, Sep 19)
Last: Lance Reddick! Lance him!

Page rendered in 1.3011 seconds
48 querie(s) executed