Negro League Stats

[diesel230]

Should MLB recognize Sadaharu Oh as the career Home Run leader?

One Japanese player was allowed in the league before 1995. Tons of talented Japanese players played in a league with set schedules, complete and verified statistics.

[Syd Thrift]

When I see people make such horrifically bad, slippery slope style arguments, I immediately understand that they’re unserious and are hoping people take anger seriously even if it’s wrapped in laughable ideas.

[Pelican]

Originally Posted by diesel230

Pelican,You seem to validate each of my points by the end of each paragraph you write. You say they have accurate and thorough stats and in the same sentence say you cant speak to the incomplete Willie Mays Home Run game, just one of the hundreds of incomplete games and box scores that will be added. "Early" seasons is what's lacking? Willie Mays is to early? They do not have accurate and complete boxscores. Not good enough to be official MLB stats compared to MLB statistics from the same era. Seamheads has done fantastic work. A variety of sources. Do you know who the sources are? I have read about some of the sources. Boxscores in home ledgers. Good enough maybe to confirm a previous source? Possibly. To use as MLB official statistics? Negative.

Again, you validate my point. Barnstorming, semi pro, PCL teams, MLB tour players, Cuban teams. Amazing talent for sure. Random games with no league structure and dedicated layed out schedule. Gmaes were played and canceled on a whim.

Never once did i say the talent was lacking or the level of competition wasn't there. I'm sure the best Black players were just as good and better than the white superstars of their time. Just like every decade after integration. Its pretty obvious.

You guys can do all the mental gymnastics needed to tell yourselves this makes sense. In the end, it does not.

So if you agree the black players were just as good, what exactly is your agenda, challenging the stats?

[diesel230]

When I see people make such horrifically bad, snarky comments without the funny, without bringing any points or ideas to the table, I immediately understand that they’re unserious and not very intelligent, just hoping to be included.

[LansdowneSt]

Originally Posted by diesel230

ItalyProf and Lansdowne

We know who won the 1994 Batting title because we had Elias sports, Televised games, the press, our eyeballs and brains, fans at the games etc....

Now please tell me how we know how the now 3rd highest Batting Average ever, in his whopping 73 At Bats for the season, Lyman Bostock won the 1941 Batting Title.
And please, lol saying "Seamheads" is not the answer

I thought you were implying there were less than the 501 AB (or whatever the 3.1 AB per scheduled 162 games is) due to the 1994 strike and asking how it is calculated when there are less than 501 AB. I answered the question sincerely and literally showing how BA is calculated as I thought 1994 was chosen for that reason.

[diesel230]

Pelican, i don't think anyone is challenging the idea that the players weren't talented. It's the amount of games, the talent level of some of these barnstorming teams they played and the fact that a lot of these games were exhibition in nature. On top of of all that, the statistics and record keeping were shoddy at best. Saying "Seamheads" for some reason seems to mean with 100% certainty that Hoss Walker went 1 for 4 in August of 1941, even though a quick search will show that even the people accumulating the Negro League Stats say its not very accurate.

The new ERA season leader is based off of 28 innings. You know that small a sample size throws the whole thing off.

[diesel230]

Originally Posted by LansdowneSt

I thought you were implying there were less than the 501 AB (or whatever the 3.1 AB per scheduled 162 games is) due to the 1994 strike and asking how it is calculated when there are less than 501 AB. I answered the question sincerely and literally showing how BA is calculated as I thought 1994 was chosen for that reason.

Thanks bud. Keep up the great work!!

[uruguru]

Originally Posted by diesel230

Pelican, i don't think anyone is challenging the idea that the players weren't talented. It's the amount of games, the talent level of some of these barnstorming teams they played and the fact that a lot of these games were exhibition in nature. On top of of all that, the statistics and record keeping were shoddy at best. Saying "Seamheads" for some reason seems to mean with 100% certainty that Hoss Walker went 1 for 4 in August of 1941, even though a quick search will show that even the people accumulating the Negro League Stats say its not very accurate.

The new ERA season leader is based off of 28 innings. You know that small a sample size throws the whole thing off.

There are a lot of issues.

Sample size is one of them, obviously

Another issue is the talent level of the league. No one can or should argue that the top NLB players would not have been top MLB players, but how deep was that talent over the entire league? If the depth was thinner, then that could arguably pad the stats of the top NLB players.

To give you an idea of what I mean, I selected a year at random: 1930. In that year, there seemed to be 15 NLB teams and 16 MLB teams. In other words, 48% of what we now consider to be "major league quality" teams were composed of black players. But in 1930, they represented only 9.7% of the total US population. So clearly the talent level of the NLB was thinner, and that would have inflated the stats of the top players.

And after 1947, the NLB talent level absolutely cratered. There were guys making the NLB all-star teams in the 1950s that were equivalent to milb talent and could not crack MLB rosters. So do we count pre-integration NLB players differently from post-integration players?

Race is such a loaded topic in the country so it's hard to imagine having a numbers-oriented conversation without it degenerating into name-calling.

Mixing NLB stats with MLB stats doesn't make any statistical sense because they were created by two completely segregated player pools. It has the same mathematical rigor as adding in Japanese stats.

But obviously, it's being done to redress a social injustice so ultimately none of the other arguments even matter.

[Pelican]

Originally Posted by diesel230

Pelican, i don't think anyone is challenging the idea that the players weren't talented. It's the amount of games, the talent level of some of these barnstorming teams they played and the fact that a lot of these games were exhibition in nature. On top of of all that, the statistics and record keeping were shoddy at best. Saying "Seamheads" for some reason seems to mean with 100% certainty that Hoss Walker went 1 for 4 in August of 1941, even though a quick search will show that even the people accumulating the Negro League Stats say its not very accurate.

The new ERA season leader is based off of 28 innings. You know that small a sample size throws the whole thing off.

Okay, understood. BTW I did not mean to be "snarky" or those other things you said - if that was directed to me. My apologies to you and others on the forum, if I came off that way. It was not my intent to belittle anyone or to lose my sense of humor. You make valid points, and I don't mean to question your sincerity or integrity or sense of humor, either!

My understanding is that the stats against barnstorming teams are not included in those now deemed official by BaseRef and MLB. There was an effort to limit the stats to games played between the various Negro League teams, even though the schedule was not balanced, and (like MLB at the time) there was a considerable difference between good and bad League teams.

If the barnstorming stats were included, there would be plenty of games played, but games played against semi-pro and town teams would be misleading and not equivalent in quality of opposition to MLB schedules. So we have the limited sample size of League games against equivalent opposition.

With respect, I think you underestimate the value of the painstaking work that has been done in accumulating stats from various sources, and thus the ultimate reliability of the stats compiled by Seamheads. Actually, for Negro League games, it is possible to know that a given player went 1 for 4 in a League game, and who the opposing pitcher was, where the game was played, lineups, pitchers. Records include doubles, triples, stolen bases, total errors. It becomes harder to calculate fielding chances, assists, putouts, errors by player. That's a weakness.

Was the Negro League record keeping up to MLB standards of the time? Of course not. Hence the effort to sift through box scores and accounts in the newspapers of the day and team records. And there are still some gaps, particularly in the early days. Point taken. We work with what we have found.

Agreed 100% that a sample size of 28 innings is too small for calculated stats like ERA. The same would hold true for batting averages in only 28 games. Sure, those figures provide a window into how good (or not) players were; but over a limited period. Anyone, with the possible exception of me in middle school, can go on a hot streak. In a long season, everybody regresses to the mean. Career stats are more meaningful, due to the large sample size. And of course Negro League players suffer by comparison for cumulative stats, due to far less games played. We all could debate a long time about how to make the comparison. Which is part of the fun, for me.

For example, this debate has motivated me to work on a database of post-season head-to-head matchups between/among MLB and Negro League players, including Winter Leagues with mixed teams. For some years, we have dozens of games with good box scores played in MLB stadiums before large crowds. To me it's a great, largely untold, story. Those involved (some white stars refused to play blacks) simply shelved the prevailing racism for a month or so. Money, I would imagine, prevailed over bigotry.

These matchups continued over decades, from the early 1920's to the late 1940's. They are the only "crossover" evidence of head-to-head competition we have. Yes, the sample size is small, compared to a MLB (or Negro League) regular season. But this evidence can't be ignored, as it provides a statistical "check" on how these players measured up versus one another.

I hope you'll forgive me for questioning your motives. I actually don't think we are far apart on this. I'm overjoyed to see the available stats recognized; but even I would take them with a grain of salt. The HTH evidence suggests that grain would be small; but we will never really know for sure. So we'll keep talking about it, right?

[Pelican]

Originally Posted by uruguru

There are a lot of issues.

Sample size is one of them, obviously

Another issue is the talent level of the league. No one can or should argue that the top NLB players would not have been top MLB players, but how deep was that talent over the entire league? If the depth was thinner, then that could arguably pad the stats of the top NLB players.

To give you an idea of what I mean, I selected a year at random: 1930. In that year, there seemed to be 15 NLB teams and 16 MLB teams. In other words, 48% of what we now consider to be "major league quality" teams were composed of black players. But in 1930, they represented only 9.7% of the total US population. So clearly the talent level of the NLB was thinner, and that would have inflated the stats of the top players.

And after 1947, the NLB talent level absolutely cratered. There were guys making the NLB all-star teams in the 1950s that were equivalent to milb talent and could not crack MLB rosters. So do we count pre-integration NLB players differently from post-integration players?

Race is such a loaded topic in the country so it's hard to imagine having a numbers-oriented conversation without it degenerating into name-calling.

Mixing NLB stats with MLB stats doesn't make any statistical sense because they were created by two completely segregated player pools. It has the same mathematical rigor as adding in Japanese stats.

But obviously, it's being done to redress a social injustice so ultimately none of the other arguments even matter.

It's another good observation. My guess is that the Negro League talent was better suited to a single eight-team league. Fifteen teams would thin it too much.

However, in comparing populations, it is important to focus on how many young guys were playing baseball = the available talent pool, rather than the total population. The farther back you go, the more baseball was a working class game (okay, I know that the origins were essentially in private clubs; but I'm talking Twentieth Century, not Nineteenth Century here). Look at all of the first generation Irish and Italian and German and Polish immigrants playing. It was also largely a rural game, and that helped in the development of black players. My impression is that there was a large talent pool for black and white players from these sources.

As far as overall talent, in both MLB and the Negro Leagues, there was a vast disparity between teams in the 1920's and 1930's, in each. There were the great players, the solid everyday players, and marginal guys who occupied the bench, filled in for injured players, mopped up from the bullpen. As a Phillies fan, I am painfully aware of how bad the team was, most years. And even the nightly A's crashed and occupied the cellar after 1931. Somehow the MLB teams survived the Great Depression, whereas teams and whole leagues folded in the less-secure world of black baseball.

One huge difference is that MLB developed a strong system of farm teams in the 1930's and 1940's, creating an unending source of new talent, training, development. The Negro Leagues had, at times, minor affiliate leagues, but that structure was weak. There was no equivalent "pipeline" for developing players within an organization. Guys would more commonly get a tryout by word of mouth, after playing for a semi-pro or industrial league or town team.

The only real comparison we have would be the annual October barnstorming series between white and black teams, basically all-star teams on each side. There are also Winter League stats, where white, black, and Latino players played on the same teams. Again, this was mostly the better players, and not every star played Winter ball. At least at the high-talent level, there were not significant discrepancies between white and black (or Latino players deemed too dark in color to play in MLB) players in these matchups. But it's a smaller sample size than a MLB Season of 154 games. It's what we have, and it at least suggests that stats compiled in segregated leagues can be compared - in some fashion.

I totally realize that race will always be touchy subject. I don't wade into these discussions lightly. These were, after all, human beings we are comparing. We'll never unravel the effects of the color bar, even if we can eliminate it in OOTP Seasons before 1947. But the value in these discussions to me is that we are talking about it at all, wherever we come down, for valid mathematical reasons based on quality of opposition. I have tried to be careful not to offend anyone's sensibilities on this. But we shouldn't shy away from debate, just because it's complicated. It's healthy to talk about this, as long as we can keep the tone respectful. I will try to do better with that, going forward.

[sprague]

Pelican's post above is clear, and very useful to the overall discussion of this subject.
I don't envy the developers on this one, having to do their best to make honest evaluations of statistics from various leagues with very different overall talent pools and schedules (even though the top 10 players of each league would be comparable) and keep people happy at the same time. I wish them well with their continued efforts.

[Reed]

Back to OOTP. To make most everyone happy I hope OOTP makes inclusion of the Negro Leagues/players and the stats optional somehow.

Of course many black players were as good as the white players back then. My question would be the stats and how stable were the teams and the non stars on the teams.

Just like I question the the inclusion of pre 1903 white MLB teams/players. Those teams weren’t stable but I have the option of not including the players and stats previous to 1901 or 1903 in my Random League. If I had to include “those” players/stats (white-pre 1901) in my random league, I would stop playing random. If I had to include NeL players/stats prior to 1947 I wouldn’t include those years in my random.

Bottom line, it’s just a game so do whatever you want. If I don’t like it, I won’t buy it.

[David Watts]

Originally Posted by Reed

Back to OOTP. To make most everyone happy I hope OOTP makes inclusion of the Negro Leagues/players and the stats optional somehow.

Of course many black players were as good as the white players back then. My question would be the stats and how stable were the teams and the non stars on the teams.

Just like I question the the inclusion of pre 1903 white MLB teams/players. Those teams weren’t stable but I have the option of not including the players and stats previous to 1901 or 1903 in my Random League. If I had to include “those” players/stats (white-pre 1901) in my random league, I would stop playing random. If I had to include NeL players/stats prior to 1947 I wouldn’t include those years in my random.

Bottom line, it’s just a game so do whatever you want. If I don’t like it, I won’t buy it.

I just wish there was an easy way to erase all history prior to 1901. A way that doesn't require me to set up unwanted minor leagues. Start prior to when I want to and delete things. Just let me start in 1901 with all stats set to zero. One of the main reasons I play random debut so much is I don't have to look at all that pre1900 garbage.

Disclaimer: I understand it's not garbage to everyone. I've just been frustrated for a long time that it's so hard to erase all history.

I really wish we had the option when setting up a league to choose whether we want players that missed the season we are creating to return. If we could simply choose to have them stay missing, erasing history would work like a charm(just as it did prior to OOTP17), At one point, I remember Markus saying he would look into making that an option. The fact that it never took effect, makes me believe it's not a simple change. That option, if it ever added, would basically make OOTP the perfect game. I love using historical players, but I want the league to be my own.

[Thundercrack]

Originally Posted by uruguru

There are a lot of issues.

Sample size is one of them, obviously

Another issue is the talent level of the league. No one can or should argue that the top NLB players would not have been top MLB players, but how deep was that talent over the entire league? If the depth was thinner, then that could arguably pad the stats of the top NLB players.

To give you an idea of what I mean, I selected a year at random: 1930. In that year, there seemed to be 15 NLB teams and 16 MLB teams. In other words, 48% of what we now consider to be "major league quality" teams were composed of black players. But in 1930, they represented only 9.7% of the total US population. So clearly the talent level of the NLB was thinner, and that would have inflated the stats of the top players.

And after 1947, the NLB talent level absolutely cratered. There were guys making the NLB all-star teams in the 1950s that were equivalent to milb talent and could not crack MLB rosters. So do we count pre-integration NLB players differently from post-integration players?

Race is such a loaded topic in the country so it's hard to imagine having a numbers-oriented conversation without it degenerating into name-calling.

Mixing NLB stats with MLB stats doesn't make any statistical sense because they were created by two completely segregated player pools. It has the same mathematical rigor as adding in Japanese stats.

But obviously, it's being done to redress a social injustice so ultimately none of the other arguments even matter.

This is the most intelligent response in this thread. Not sure how sincere your last sentence is, but I don’t think you redress social injustice with wishful thinking and virtue signaling

[LansdowneSt]

Originally Posted by David Watts

I just wish there was an easy way to erase all history prior to 1901. A way that doesn't require me to set up unwanted minor leagues. Start prior to when I want to and delete things. Just let me start in 1901 with all stats set to zero. One of the main reasons I play random debut so much is I don't have to look at all that pre1900 garbage.

Disclaimer: I understand it's not garbage to everyone. I've just been frustrated for a long time that it's so hard to erase all history.

I really wish we had the option when setting up a league to choose whether we want players that missed the season we are creating to return. If we could simply choose to have them stay missing, erasing history would work like a charm(just as it did prior to OOTP17), At one point, I remember Markus saying he would look into making that an option. The fact that it never took effect, makes me believe it's not a simple change. That option, if it ever added, would basically make OOTP the perfect game. I love using historical players, but I want the league to be my own.

+1 The Import League History toggle should have a drop down saying "back to" and you could pick any year be it 1871, 1876, 1901, 1947, 1961... whatever you want. I hate that the band-aid to tie the 19th c. team stats requires those stats to be associated with unrelated teams - in the way Louisville (iirc) turns into the Red Sox so all those Louisville stats look like Red Sox stats in my play merely because I imported history.

[Sweed]

Originally Posted by sprague

I agree with a lot of this.

Sample size is the main phrase here.
For example I would have no problem to equalize negro season stats with MLB, if 154 games would be brought together. So if say that took a negro player to combine the 1921-1925 stats together I would be pretty much ok with the results, because they played a similar number of games, and there was a similarity of the skills of the star players.


But we too know it is a big difference from a player hitting .356 in 140 games and a player hitting .450 in 24 games.

When sample size can be properly harmonized I would be fine with the complete integration...

I was going to post something similar. Small sample size just feels wrong to me. That would apply if Tony Gwynn or George Brett were injured and only played 54 games, and they took their BA from that year as the new single season record. I'm fully aware the cause is different but the sample size is still the sample size.

I know technically it wouldn't be one season, but a MLB qualifying number for stats, even if they required three seasons to accumulate, seems like a better way to compare and determine where something fits into the single season record book. My preference would be they go forward or backward from the small sample size to get to the required ABs, IP, or whatever stat and use that instead of limiting it to the one season small sample. I'm sure there are problems that would come from expanding into surrounding seasons to get a larger sample that I haven't thought of. It's an imperfect situation.


I don't doubt for a moment that two NLB teams playing against each other was the same caliber as a MLB game. Stars in NLB would be stars in MLB, and MLB stars would not dominate the NLB but would "just" be stars there too. From what I read here it seems the data collectors are being meticulous in getting relevant stats to the best of their abilities. I support these stats being incorporated into the MLB record book. Though I'd like to see the sample size issue addressed I'm not going to lose any sleep over it.

[OutS|der]

The way I see it the stats are just a product of the era, the only issue is when they use small sample size like the new ERA leader of 28 games otherwise you can argue about the talent levels till the cows come home.

All past totals have to be taken with a grain of salt due to how baseball has evolved over the years. Due to various reasons stats were inflated due to subpar talent, using speed, steroids, juiced balls, or when pitchers threw low 90's and could doctor the ball.

[Pelican]

Yes, and one stunning thing for me is that in the early "dead ball" years, they would use one baseball for the whole game. They weren't made as well as now, and they would get soft and lumpy, dirty and ripped. Obviously easy to get lots of movement, even without the then-legal spitball. It literally was a "dead" ball by the end of the game.

And yes, the farther back you go, the more pronounced the differences are. I don't know how anyone can play random debut, for that reason. It really becomes an apples and oranges comparison. It's barely the same game.

[Reed]

Originally Posted by Pelican

Yes, and one stunning thing for me is that in the early "dead ball" years, they would use one baseball for the whole game. They weren't made as well as now, and they would get soft and lumpy, dirty and ripped. Obviously easy to get lots of movement, even without the then-legal spitball. It literally was a "dead" ball by the end of the game.

And yes, the farther back you go, the more pronounced the differences are. I don't know how anyone can play random debut, for that reason. It really becomes an apples and oranges comparison. It's barely the same game.

Heck, present day baseball is barely the same game as it was in the 1970s and 1980s IMO.

It is interesting that the BAs in the dead ball era is the same as the BAs in the 1960s or the past few years.
Main difference now it is all HRs, KOs, and bullpen usage. (And DH, extra inning rule, shot clock, intentional walks, playoffs, interleague play, bigger bases, bases being moved some to improve the running game, Robo umps soon, etc.). Did I say it’s barely the same game as the 1970s-1985??? It’s more like a new game.

Yes random is 90+% fictional. In my random I use 1984 as my stat/modifier year since it seems like a good balanced stat year from 1901-present. I play with injuries off, players miss seasons and retire as in real life on. That way the "stars" from the various years become stars in my league although I get a surprise every now and then. If someone leads my league in a category and I am not familiar with them then I try to find out more about them. That way the player becomes a person and not just a stat.

One example. Pete Dowling was a very good young pitcher in dead ball era. He was an alcoholic so during the off season the team gave him a contract that he signed that stipulated he couldn't drink during the season. That spring he realized he couldn't go without drinking so he quit, went to California and played for a semi pro team. While there he is credited with saving 3 men from drowning when their fishing boat turned over. A few months later he was drunk and showed up at the train station just after the team departed to go to a neighboring town for a game. While drunk he decides to walk the train tracks to the neighboring town. He died when a train hits him while he was walking. These stories make the game come alive for me.
If I just played dead ball era or modern era the game would be pretty boring for me. Obviously I was a big baseball fan especially from 1970-1985 or so. For me, that is when baseball was baseball at its best.

[pilight]

Nm