Data on Rube Foster I just posted on the 1925 ballot discussion thread:

Most interestingly, in May 2004 Robert Charles Cottrell is releasing The Best Pitcher in Baseball: The Life of Rube Foster, Negro League Giant. It'll be interesting to see if that book has any more details.

The Cottrell biography came into print in 2001; it's been my primary source on Foster's career, in fact, and yes, it has more details.

I think the evidence will show that he was a player/manager those years who chose to spot start himself in the big games. Hard to assess the value of that, but I count them as analogous to seasons in relief (like Smoltz' first two years as a reliever, but not as good). Still, he has ten to twelve good to great seasons, highlighted by 6 dominant seasons.

The evidence, drawn from Cottrell's biography, Holway's _Complete Book of Baseball's Negro Leagues_, and Riley's _Biographical Encyclopedia_, is contradictory, and I'm not at all sure how to interpret it, so I'll just lay out what I have found, for now.

First, here are some relevant passages on Foster's play 1911-14 from Cottrell.

On Foster in 1911: "Foster, for his part, was now pitching less frequently than in the past, although his outings drew large crowds and considerable publicity. Left-hander Pat Dougherty and righties Frank Wickware and Bill Lindsay led the American Giants' pitching staff."

On Foster in 1912: "In 1912, the American Giants repeated as champions of the Chicago semipro circuit, winning 112 of 132 contests. . . . Scattered box scores and news accounts suggest that Foster remained, at least on occasion, a superb pitcher, with such incomplete records displaying a 6-0-1 record for the year."

Game summaries show him beating the Indianapolis ABCs (a black team), the Plutos (no indication of what sort of team this was), the West Baden Sprudels (a major black team under the management of C. I. Taylor, who would soon move to the ABCs), and the St. Louis Giants (a major black team). His tie came against Jose Mendez, pitching for the Cuban Stars.

On Foster in 1913: "his traceable mark for the regular season stood at 6-1." No indication whether he pitched on not in championship series against New York Lincoln Giants, which the Lincolns won. Smokey Joe Williams dominated the series, winning 7 of the 15 scheduled games.

On Foser in 1914: "Foster produced a 6-3 mark in games whose news accounts have been uncovered" (the team went 126-16). These nine games include a win against St. Joseph's (an independent Michigan team), a loss in relief to the Cuban Stars, a 1-0 shutout win over the Cuban Stars, a 2-0 shutout victory over the ABCs, victories over Benton Harbor and the French Lick Plutos, a victory over the Cuban Stars, a victory over "a group of top City League players" (I'm guessing this game, as an exhibition is not counted in Foster's record), a loss to the ABCs, and a loss to the Cuban Stars.

When the American Giants played the Brooklyn Royal Giants in a championship series in September, Foster did not pitch: the Chicago pitchers in the American Giants' four game sweep were Lee Wade, Frank Wickware, and Horace Jenkins, who pitched twice.

Passages from Riley

On Foster in 1910: "Incomplete statistics show Foster contributing a 13-2 record" (to a team record of 128-6).

On Foster in later years: "That year, 1911, Foster renamed the team the Chicago American Giants. . . . Foster himself was phasing out of his active career, with incomplete data showing records of 5-4 in 1914 adn 1-0 in 1917, two of his best teams of the decade. Enormously popular at the time, he still took the mound on special occasions to enhance attendance."

Statistics from Holway

1910: Foster 3-0, 2.17 RA (Frank Wickware leads team with 6-0 mark, posts 2.88 RA)

There have been some discussions about the methodology of the i9 numbers. You can view the i9 "Projection Notes" here:



In particular, here's a quote from that page:

"Note that, from the beginning, our goal is not to claim we somehow know what these players would have done. If, at the end of the day, we can look at the projections and say that we have gotten the shape of their careers correct, that, yes, Pete Hill and HR Johnson were great players, regardless of whether we are correct that Hill hit .322 in 1907, that, at the close of day, Johnson was a legitimate MVP candidate a few times in his career and a borderline HoF candidate, that will be success."

Would anybody know how negro league players fared against white major leaguers? Is there any won-loss records?

He's 5 years away from eligibility and his personal thread hasn't been opened yet, but I'm already curious about him.

Oscar Charleston.

In the NBJHBA, Bill James says rather emphatically that Charleston deserves to be right there with Ty Cobb and Willie Mays when we discuss the greatest centerfielders - make that greatest players - ever. How strong is the evidence that supports this point of view?

(Reconstruction of post #20)

(Caution: long post)

Adventures in Demographics, Part II.

HOMers by birth Census:
1840 01 Start
1850 12 White, Wright, O'Rourke, McVey, Spalding, Barnes,
xxxx xx Sutton, Hines, Anson, Radbourn, Bennett, Richardson
1860 12 Galvin, Stovey, Connor, Keefe, Gore, Kelly,
xxxx xx Brouthers, McPhee, Ewing, Glasscock, Ward, Clarkson
1870 14 Hamilton, Delahanty, Young, Burkett, Nichols, Collins,
xxxx xx Dahlen, Davis, Rusie, Kelley, Clarke, Keeler, Lajoie, Wagner
1880 03 Flick, Mathewson, Walsh

HOMers by birth State:
NY 13 Start, White, Wright, Barnes, Sutton, Radbourn, Kelly,
xx xx Brouthers, McPhee, Collins, Dahlen, Davis, Keeler
PA 06 Bennett, Stovey, Ward, Mathewson, Walsh, Wagner
OH 04 Ewing, Delahanty, Young, Flick
MA 03 Keefe, Clarkson, Kelley (all from Cambridge)
IA 03 McVey, Anson, Clarke
CT 02 O'Rourke, Connor
NJ 02 Richardson, Hamilton
WV 02 Glasscock, Burkett
IL 01 Spalding
DC 01 Hines
MO 01 Galvin
ME 01 Gore
WI 01 Nichols
IN 01 Rusie
RI 01 Lajoie

Methodology underlying the following numbers. What I've done is take the ballplayer populations discussed below and allocate them based on their birthyear to a census, e.g. 1856-1865 to the Census of 1860. And then based on their birthstate, I've used the U.S.Census data found here to assign to them an "expectation" that they would be 'black' in an attempt to get a rough idea of how many black players would be playing in the major leagues during this period if racism were absent, but that whatever other geographic forces present that shaped the distribution of white players were also applied equally to black players.

Based on the distribution of current HOMers by birthstate and birthyear, we currently have an expectation of 1.06 black HOMers. In its most simplistic form, the generalization is that blacks live in the South and ML ballplayers are from the North, in particular from New York and Pennsylvania, which have black populations in the 1%-2% range during the Censuses of 1850-1880. (Wagner was the tail end of the 1870 census; Mathewson was born in 1880.)

This is about to change radically.

Since I'm not about to predict which players are about to become HOMers, I will shift to using the total population of Major League regulars (which I've previously defined to be players who played more than half of the league leaders in either Games Played, Games Started, or Innings Pitched at either fielding, pitching, or hitting), weighted by number of seasons as a regular, as the basis for the following numbers.

Using the same methodology described above, following are the "expected" percentages of black players for each census decade.

1840: 02.5%
1850: 05.1%
1860: 03.8%
1870: 04.9%
1880: 05.0%
1890: 08.9%-09.9%
1900: 12.5%
1910: 14.0%-15.3%
1920: 10.0%-11.6%

Since I do not have census data for those born in the Caribbean, the ranges above are based on assuming 0% or 100% black for the players from those countries. The real number lies somewhere inbetween, but the number of such players is small enough so it doesn't change the results radically.

The 1890 Census begins the arrival of the Southern white players like Cobb, Speaker, and Jackson (to pick the most prominent early arrivals). As such, it would also mark (in a less racist society) the arrival of the Southern black players, such as Pop Lloyd and Joe Williams. (They are technically in the 1880 census, but the line falls just between Lloyd, 1884, and Cobb, 1886.)

The cumulative "expectation" between 1840-1920 is in the low 9's (8.8%-9.4%), but it's unevenly distributed and heavily weighted in favor of the 1900/1910 Censuses, which also are the heyday of the Negro Leagues, and against the 1840-1880 Censuses, which are the years we've been studying to date.

Distributions of Negro League HOFers by Census:

1880: R.Foster, P.Lloyd, J.Williams
1890: J.Rogan
1900: O.Charleston, J.Johnson, T.Stearnes, C.P.Bell, B.Foster, M.Dihigo
1910: S.Paige, B.Leonard, H.Smith, W.Wells, J.Gibson, R.Dandridge
1920: L.Day, M.Irvin, J.Robinson, R.Campanella, L.Doby

Leon Day (1916) just misses the 1910 census. The rest of the players from the 1920 census are those who had the opportunity to play significant time in the major leagues.

Distribution of Negro League HOFers by birthstate:

TX: R.Foster, Williams, B.Foster, Smith, Wells
AL: Paige, Irvin
GA: Gibson, Robinson
VA: Dandridge, Day
FL: Lloyd
OK: Rogan
MD: Johnson
TN: Stearnes
MS: Bell
Cuba: Dihigo
NC: Leonard
SC: Doby
IN: Charleston
PA: Campanella

During the period from 1870-1900, the South contained about 1/3rd of the national population, and about 90% of the black population. (From an earlier post.) Note that only 10% of the Negro League HOFers were born in the pre-Civil War "free" states, Charleston from Indianapolis and Campanella from Philadelphia (2 of 20 not including Dihigo).

I think it's fair to say that, statistically, it wouldn't be surprising for the Negro League Players to be unevenly distributed by decade, because the parallel distribution of great southern white players is also unevenly distributed by decade. There may have been great baseball being played in the South during the 19th century, black and white, but it would appear to be largely unnoticed and undocumented.

If somebody has suggestions for improving the above methodology, I will welcome them.

does anyone know where I can find out the complete rosters for the east-west all star game

does anyone know where I can find out the complete rosters for the east-west all star game

I'm about 99.99% sure they're not on-line anywhere. There are a few books that have the east-west PARTICIPANTS, but I don't remember ever seeing the ROSTERS (including those who may have been selected but did not play in a certain year).

A nice article in today's Washington Post on Ted "Double Duty" Radcliffe.

Now available from McFarland Publ, list $29.95:
the paperback edition (2005) of William F. McNeil, Cool Papas and Double Duties: The All-Time Greats of the Negro Leagues (2001)

Essentially, this is a a 20-page history and interpretation of the recognition of Negro League players followed by about 60 pages each for selection of a ballot, biographies, and ballots completed by players and historians.

What I would investigate before purchase:
- is the hardcover edition cheap and easy to get?
- is any of the data on all-time ballots available in an electronic database, probably from a HOMeboy?

Personally, I think McNeil's "Baseball's Other All-stars" is a better book to own. Had lots of good info on Winter Leagues AND Japanese Leagues too....

Do any of you have McNeil's book on the California Winter League? I'm wondering if it (or any other source) provides complete or mostly complete statistics. Information from Winter League statistics might help Chris to calibrate his MLEs.

One voter's opinion on the # of Negro Leaguers (without 5 years of MLB experience) who will be elected to his PHOM.

Total - 31

P (8) - Joe Williams, Satchel Paige, Bill Foster, Bullet Rogan, Rube Foster, Dick Redding, Ray Brown, Leon Day

C (3) - Josh Gibson, Louis Santop, Biz Mackey

1B (3) - Buck Leonard, Mule Suttles, Ben Taylor

2B (2) - Frank Grant, Bill Monroe

3B (3) - Jud Wilson, John Beckwith, Ray Dandridge

SS (4) - Pop Lloyd, Willie Wells, Dick Lundy, Grant Johnson

OF (8) - Oscar Charleston, Turkey Stearnes, Cool Papa Bell, Pete Hill, Willard Brown, Christobal Torriente, Martin Dihigo, Spots Poles


Just Missing - 17

P Just Missing (8) - Jose Mendez, Nip Winters, Hilton Smith, Bill Byrd, Andy Cooper, Chet Brewer, John Donaldson, Harry Buckner

C Just Missing (2) - Bruce Petway, Frank Duncan

1B Just Missing (0)

2B Just Missing (3) - Newt Allen, George Scales, Sol White

3B Just Missing (0)

SS Just Missing (1) - Dobie Moore

OF Just Missing (3) - Alejandro Oms, Rap Dixon, Bill Wright

From 1909-1959 (Frank Grant's first eligible year to Satchel Paige's first eligible) there were 92 HOM spots. Arguably, some would fill the 19th century backlog, but some slots after 1959 will fill the Negro League time period backlog.

Therefore, it looks like I will be electing about 1/3 Negro Leaguers during their eligibility period. That is above the 10-15% advocated by the Demographic Hypothesis and at the low end of the 30-50% Star Player Hypothesis.

This seems like a good place for this question:

What happened to the thread that was collecting the MLEs and estimated win shares for Negro-Leaguers. It's dropped out of the recent threads list, and I can't find it on the home page. If it's there, what's it called?

Negro League Win Shares Estimates

Question: Could somewhat with a book on Negro League players look up James Beckham? I'm not looking at him as a potential HoMer, but I have a chance to meet the man so I'd like to see if there's any info out there on him. I couldn't find any on the web.

Carl:

He's not listed in the Riley or Holway books, either.

FYI

Neil Lanctot, Negro League Baseball: The Rise and Fall of a Black Institution (U Penn, 2004) is this year's winner of the Seymour Medal, awarded annually by SABR to the best new book of baseball history or biography.

The citation calls it "a scholarly study of the Negro Leagues from the 1930s into the 1960s."
Seymour Medal 2005, SABR press release

hsc list of Seymour Medal winners with reviews (2005, not yet)

To Marc

pls e-mail me off-list at judco12000@yahoo.com

I have a basketball question for you.

thanks

Ray

Mandak League: Haven for Former Negro League Ballplayers, 1950-1957 (Hardcover)
by Barry Swanton (Unknown)
List Price:$29.95

Available at amazon, of course. Haven't seen it, not recommending, just thought some might want to know it's out there.

Looking at the draft website Project Protoball: Supporting Research on the Origins of Baseball, whose focus is (writings on) base-and-ball games from antiquity to 1860.

One article cited in the "Fat Chronology" will interest many of you and will provide some ea-early background for the demographic and geographic matters discussed in this thread. Jerry Malloy, "Early Black Baseball/Charles Douglass"

Maybe publisher Cliff Blau (cblau) knows when the article was completed by Malloy (deceased 2001?).

Mandak League: Haven for Former Negro League Ballplayers, 1950-1957 (Hardcover)
by Barry Swanton (Unknown)
List Price:$29.95

Available at amazon, of course. Haven't seen it, not recommending, just thought some might want to know it's out there.

There was something, or two somethings, on the Mandak League at the 2005 SABR Convention in Toronto. At least one in the book collection of articles written before the convention. My copy is not where it should be.

here are the stats the HoF put out for the Negroe Leaugers on the ballot hitting stats and pitching stats

wow, that's awesome.
earliest stats are from 1920, so guys like Redding and Santop already were 30 yrs old. It only captures one top Redding season, 1921.

gotta run, but looking forward to who seems to move up and who down...

I can see why, based on those stats, they chose Cooper over Redding. Redding didn't do anything after age 32. That certainly doesn't help his status as a career candidate.

based this I would have elected to the HoM fromt the players on the ballot only Jud Wilson (conferemed my opinion of him) , Biz Mackey (conferemed my opinion of him) and Mule Suttles(thought he was better then his stats showed) for play in the Negroe Leauges during this time with Suttles being a border line candite (that means totaly discounting pre Negro Leuges and Cuban leauges, barnstorming games ext.) I was quite disapointed with Ray Brown though and am still baffeled as to what maid them pick Cooper.

Willard Brown credited with 67 HR in 1806 AB. Contrast with Beckwith (77 in 1647), Oms (31 in 1042 tail end career), Scales (66 in 2215), Suttles (133 in 2727), Taylor (21 in 1695), Torriente (46 in 2040), Wilson (71 in 2954).

The guy I'm most curious about here is Torriente. Sure looks like he walked a lot compared to what I expected (12% of all BB+AB).

It's worth noting that our guesstimated most walktastic guy, Wilson, didn't walk all that much. 11% of all AB+BB.

So those are good walk rates, but nothing wild. Leads me to ask two or three things: 1) could walk rates may have been lower in the NgLs in general? 2) If so, does that mean anything to the MLE process (since we don't yet have lg walk/obp rates) 3) could this also in a funny way vindicate the argument that Brown's walk rate was not as big an issue as some suspected (logic being that a player whose natural tendency to hit for contact playing in a league that doesn't draw walks may doubly dampen his walk rate, making it appear more of a liability than it may be...but ironically, probably increasing his value to those within that particuarl league structure/envrionment)

Ray Brown looks fine to me. .700 winning pct, almost almost 2:1 K/BB, only two really bad seasons, 1936 and 1945, plenty of low-three, sub-three ERAs. Looks like a hoss to me.

I would, however, like to know what the league ERA is. They inconveniently did not provide that info (he said, looking the fit horse in the mouth).

Anyone know what the asterisks and itals are for?

Cooper's rates and peripherals are fabulous. Fewer hits than innings, 2.4 K/BB, .700+ winning pct. It's tough to say anything about his ERAs sans lg ERAs. He did allow about a half run more RA per nine than Brown, though.

Anyway, plenty of food for thought, and also an opportunity for Chris and I to plop the league averages into our MLEs!

Ray Brown looks fine to me. .700 winning pct, almost almost 2:1 K/BB, only two really bad seasons, 1936 and 1945, plenty of low-three, sub-three ERAs. Looks like a hoss to me.

I would, however, like to know what the league ERA is. They inconveniently did not provide that info

thats why I don't think he's a mistake but he looks worse than I thought his era numbers seem worse then Brown and Brewer (though that might easly be to changing avrages which of course they left out)

You could estimate league RA based on league average and slugging.

DL,

I can figure using RC to get most of the way there, but how would you estimate the TOB portion without walks?

Would you take the MLB average BB/AB and then take it down a percentage point or two, then figure the walks into the TOB part of RC?

Estimate the number of walks based on the players and cross-check with the number of walks the pitchers were giving up per inning. That should give you a range for number of walks per game with the best hitters on one side of the distribution and the best pitchers on the other side of the distribution. I'd use that as a decent confidence interval.

OK, to take the idea I just posited a little further....

The NgLs batted .277 and slugged .377 in 1923.

Plugging these into RC .10 runs per AB and 3.73 runs per 27 outs. That's without walks.

The majors that year hit .284/.391, or .111 runs per AB sans walks.

In 1923, the NL and AL walked 7563 times in 85224 ABs, which is 8.9% of all ABs or .088 walks per AB. So the majors are at .134 runs per ab. They made 61006 outs (AB-H) and were caught stealing about 1250: .745 outs per AB. So the bigs scored .187 runs per out, *27 outs = 4.86

Actuals: NL 4.85/ AL 4.98.

So a little low, but not unreasonable. So what if we take the NgLs down to a 7% lg average walk rate? (or about 1.88% lower than MLB, but I was rounding)

(on base factor)
.277 h/ab
.07 bb/ab
---------
.347 tob

(total bases factor)
.377 tb

(opportunity factor)
1 ab + .07 bb = 1.07 PA

OB*TB/OPP= .122 Runs

(Outs factor)
.723 batting outs/ab
.01 CS outs/ab (per MLB rate)
---------------------------
.733 outs

RC/OUTS = .122/.733 = .166 R/out

RC PER GAME = .166 * 27 = 4.49 R/G

4.49 / 4.86 = 92.4% of MLB scoring.

Does that strike everyone as reasonable?

Now if 4.49 is your lg avg RA, I wouldn't even bother with ERA+ for NgL pitchers. Just go straight to RA+. Andy Cooper's RA in 1923 was 4.86, or a 92 RA+. Jose Mendez was 4.51, 100 RA+. Redding is 5.74 which is probably near replacment (it's 27% 'neath the league). William Bell (a rookie), was worse.

If this works for everyone, I'll start putting together lists of league batting totals and RA totals.

I think it's best to deal with runs and runs allowed averages (aka "TRA"), as the ERAs in those statistics are almost entirely estimated. Nor did the project collect fielding statistics, except for raw totals of errors (which were not broken down by position). So I'm not sure what these ERAs are based on.

Anyone know what the asterisks and itals are for?

I think the italics at least (maybe both) are for years where they're using statistics that were published at the time, rather than independently compiled numbers. Notice that Biz Mackey for 1947 is given no games played--this is a pretty good sign these numbers were taken from something published in 1947 that didn't happen to have a games column. Also in 1945 Mackey played 36 games, but is listed with zero walks, zero sac hits, zero stolen bases, and zero errors. One or more of those columns should probably be blank, rather than filled with zeroes, as I'll bet the source simply didn't list that data.

Something else: the "league" averages apparently combine east and west, so there aren't any separate averages for the NNL and ECL, for example.

Where these numbers differ significantly from what we have seen, and from the MLEs that have been generated, what should we regard as authoritative. Is this better data than what we've seen before?

Among the position players, at least, George Scales would have to be the big revelation, no?

I think we pretty much already knew that Scales had some flashy offensive numbers (href="http://www.baseballthinkfactory.org/files/hall_of_merit/discussion/george_scales">see his thread</a>). He played in a couple of very good hitting parks, so some air was taken out in the MLEs. If you compare them, the new numbers are not really significantly different from what we've seen before.

Here's the link: George Scales thread.

Sunny,

The big difference to me is in torriente's walks, which might raise him up a bit from the Clementesque comparisons that have been made of him.

A second diminishng aspect of Scales that appears in his thread is how his bat appears to suffer when the competition becomes fiercer in the no-league depression era.

Gary,

It looks like some of the league averages combine East and West and some don't. The ECL has its own league averages for 1923-1928, the 1929 ANL appears to be blended into the NNL league average. Then E/W have the same league average until the rise of the new NNL and its NAL counterpart, whereupon there seems to be differentiation again.

Cancel my last post!!! Gary, I'm sorry, you were correct that there's no differentiation in league averages between East and West. My bad all the way; sorry everyone for any confusion.

So I put together what numbers I could regarding NgL walks from the data we've got. And what we've got is 43 hitters and 21 pitchers, sort of. The hitting stats include pitchers who didn't otherwise play a position (only the current HOF pitchers, so Paige and Williams for instance). The pitching stats include some innings by Charleston, Wells, and other guys in the Hall for something other than hurling. Of course, these are all All-Star kind of players as well. They don't all draw walks, but many of them do not, surprisingly.

What I've done is aggregate them by season and dump them straight into a chart. I have not made any attempt to normalize or weight the totals to account for any undue influence by one player within one year. The tables include the AB or INN, the BB or BB allowed, the BB/AB or BB/9, and the NL and AL BB/AB and BB/9. It is worth noting that the AL and NL differed pretty widely during the period, with the AL being more walktastic than the NL.



First off we just need more data. We need full league totals for the NgLs to make a truly robust comparison. That said, if these tables suggest anything, it's that the top hitters in the black game were drawing walks at much slower rate than comparably awesome players in the majors, especially the AL. The NgLers barely edge the AL's overall league average in most years or else finish between the NL and AL averages (including pitchers).

I'd prefer more data to give an exact answer to the question of the moment, but I do feel comfortable suggesting that the league average in the NgLs was probably a couple percentage points or more lower than the MLB walk rates. Nearer six percent, as a league, than ten, you'd have to say.

The data on the pitchers is much sketchier since it includes some mystery-pitcher innings from non-moundsmen and because there's not many innings in most seasons. What little we can say, however, is that the top Negro League pitchers were walking fewer men than the MLB guys. Which should probably be a no-brainer given the lower walk rates exhibited by the top hitters, but I figured I'd better say it anyway.

Can anyone suggest where to with this data now? Especially in terms of estimating league walk rates and generating RA+ for pitchers?

If this helps, these are NeL figures I have (for all players):

Walks/Plate Appearances
1916 west 681/6915, .098
1921 NNL+ 1965/25468, .077
1928 east 842/11437, .074
1928 west 1444/20516, .070

Walks/9 Innings Pitched

1916 west 3.35
1921 NNL+ 2.67
1928 east 2.62
1928 west 2.47

Sorry, that's Walks/At Bats, not Plate Appearances.

FWIW, I do think that during the 1920s and 30s, at least, there were somewhat fewer walks in the NeLs than in the majors; but the difference might not be quite as radical as it appears, especially in the 1930s. As you move into the 30s, box scores get sketchier and sketchier; you're a lot more likely to have box scores with no walk or strikeout info, as well as no at bats, which makes it impossible to assign walks to individual batters. I think this results in something of an undercount in walks (as well as pitchers' strikeouts) in the 1930s.

Btw, in 1916 NeL games between the top teams in the west, the walk rate was actually higher than in the majors overall:

league-BB/AB-BB/IP*9
NeL----.098---3.35
NL-----.073---2.41
AL-----.099---3.24

Old Paradigm: Walks are something that pitchers do.

New Paradigm: Walks are something that hitters do.

Still I would wonder, especially in a small sample, whether there were particular pitchers with big workloads who were especially generous or stingy with walks at particular times. Not to say that I would do the labor necessary to find out myself, but I am curious. Same of course with hitters, as Doc mentions....

One thing I'd like to do is see if the SAC numbers in the NgLs are significantly larger than in MLB. If the one-run strategies were in greater use in the NgLs, then we'd at least have some sense of how come the R/G was lower.

In my copious free time!!!!

ok, here's some preliminary results.

Using the sample of hitters available to us in the HOF pdfs, I figured out the relative AVg and SLG of the hitters in the sample for each year. And in each year, i used the group's sample relative averages, and used them as the basis for a group of MlB players who had the same relative avg/slg compared to their league. i recorded their bb/ab, compared to the big leagues, then used that ratio as the basis of an estimate for the NgL walk rates....



Then I figured out a very basic RC the league based on the lg avgs and bb/ab




Then finally, I used the yearly pitching stats on the HOF pdfs to figure individual pitchers RA+



Ray Brown comes out not so great. Most of the other guys like up with expectation. Could be a park effect. Dunno.

Now I have to submit my ballot and make memorial day potato salad.

Then I figured out a very basic RC the league based on the lg avgs and bb/ab

I remember noticing with Gary A's data that the basic RC formula severely underestimates NeL runs scored. Chris Cobb and Gary A responded that the difference reflected the errors. The 1920s NeL were similar to 1890s NL in number of errors. See # 173 to 179 of the Major League equivalencies thread.

These are league fielding percentages I have:

1916 western NeL: .949
1921 NNL+ : .948
1928 eastern NeL: .956
1928 NNL: .955
1929 ANL (east): .949 (from Pittsburgh Courier published stats, 9/28/29)

Note to self: Revisit Hilton Smith.

Is this affecting anyone that Dick Redding looks like a marginal player after age 32? I've heard a lot of people refer to Redding as a "career" candidate.

Brent, thanks for the redirect to the errors discussion. Here's a quick chart comparing RC vs R/G in the seasons Gary and Chris provided.


YEAR LG R/G RC/27 DIFF
------------------------
1889 NL 5.84 4.07 1.77
1901 AL 5.35 4.31 1.04
1900 NL 5.21 4.22 .99
1921 NNL 5.10 4.16 .94
1902 AL 4.89 4.31 .58
1928 NNL 5.03 4.53 .50
1911 NL 4.42 4.02 .40
1936 NL 4.71 4.55 .16
1927 NL 4.58 4.62 -.04
1926 NL 4.54 4.60 -.06
1931 NL 4.48 4.54 -.06

Despite the fact that I've chosen not to list the fielding pcts., the influence of errors over time is easily visible since league error rates have been going downward since the beginning of bsaeball. Thus the difference between estimated and actual R/G narrows considerably and rapidly.

test

So to make a quick theoretialy jump from the very small sample of data above....

The NNL may have improved over time (and presumably the ECL too), particularly in the fielding dept. I'm guessing that given glove technology's rapid advances in MLB (and its resultant part in the reduction of errors in MLB) that the NgLs ultimately followed a similar error-reduction path (though perhaps not exactly the same since league quality went up and down depending on the economic conditions and field-maintenance may have been impacted by finances).

So what I'd like to figure out how to do is to know the slope of the line from 1920 to 1948 in terms of NgL errors. Then we can take that information and estimate R/G a little better. If fielding was improving by ten pct points each decade, that's a pretty substantial improvement. The 1921 vs 1928 NNL R/G vs RC certainly suggest it. Does anyone have data that can help us draw that line a little more definitively?

I was going to ask a question about Mendez' poor showing in the 1952 Pitts Courier poll (and how this reelates to pre-1930 leaguers in general), but I can't find my hard copy right now, nor where we posted it in one of our threads. If someone else has it, please link or print.

If the question is Why?, then I think Sunny mostly answered it on the ballot thread:
-he was older than the voters
-his salad days were in a foreign country during a time when US players had only recently begun playing there
-by the time he made it to KC, his glory days had mostly faded anyway.

And you might also add that his salad days took place in a very different kind of league environment that was at least somewhat unfamiliar to the voters.

Parroting what sunny said, it might be akin to a VC player in 2050 voting on Ichiro or Matsui.

Or a VC guy from 2000 voting on black players from the integration era.

field-maintenance may have been impacted by finances

It isn't obvious how to score the "errors" caused by poor quality of the playing field, many of which are noticeable. Is it reasonable to suppose that black and white leagues scored equally on average? Are there anecdotes about systematic differences anywhere, eg between black and white pro leagues or between the US and Cuba?

In the aughts or so, there were many concerns about scoring differences even within league and evidently some frustration that leagues cannot easily make scoring more uniform.

If the question is Why?, then I think Sunny mostly answered it on the ballot thread:
-he was older than the voters

How did Mendez fare versus his contemps in the poll? What others from his day (and before) did well in the poll?

Over on his blog (http://agatetype.typepad.com/agate_type/), FOHOM Gary A. has been calculating park factors for the NNL in the years 1920-1924 & 1928 by comparing home and away games, then balancing them for # of games at home versus away (they differ in the NgLs a lot of the time). Even these five seasons give us a good start for looking at the park effects on our previous RA+ estimates. In the chart below I've taken his PFs and applied them to the estimated RA+ table in post #319 above to give a sense of who is being affected.


OK, couple things here
1) The PFs for CP Bell's three big pitching seasons: 110, 133, 104.

2) Most of these guys are gaining RA+. While this might seem counterintuitive, I think it's not. The park illusions were probably pretty dramatic in the Negro Leagues. IIRC, Schorling park, for instance, is the same park that drained the life from the stats of the Hitless Wonder White Sox. On the other hand, the St Louis Stars' home park was an absolutely band box. It's responsible for that 133 I noted for Cool Bell. In addition, these guys only represent but a small fraction of the parks, teams, and seasons Gary A has numbers for. I'd thought previously that the RA+ numbers for this group seemed very low given their stature in the leagues, it appears that PFs are part of the answer why they were coming out so low before.

3) IIRC, Dihigo's Cuban Stars didn't even have a regular home park, so I guess their PF should simply be 100 for all seasons. I dunno for sure.

4) This is kind of the tip of this ice berg. With just six seasons from one league to work with, we're only beginning to see the transformative effects of parks on pitchers' estimated RA+. I don't know what will happen, but it will be interesting to see which backloggers, if any, rise up from this process to make themselves more distinguishable from the mass of candidates.

5) Is Bullet Rogan as good as Willie Foster? I think he might be better with more park information (and more team-based information too). Certainly seems better when considering his hitting.

And let the record show that Hilton Smith is still at 126, best of the still eligibles by a significant margin.

I like Hilton Smith also but you have to concede that Dick Redding's best years aren't there.

IIRC, Dihigo's Cuban Stars didn't even have a regular home park, so I guess their PF should simply be 100 for all seasons. I dunno for sure.

Probably not, because they didn't play even numbers of games in differen parks.

Also, and this is HUGE, you need to adjust both for park and LEVEL OF COMPETITION.

A team might have played in a 'pitcher's park', but they may have also played against easier pitching, somewhat negating the park effect....

Here by the way is the detail on the generations of black baseball players in the HoF, by decade of birth.

• 1840s-'50s: avg. 0 per decade
• 1860s-'70s: 1.5
• 1880s-1910s: 7.25
• 1920s: 2 (Campy and Doby, "the lost generation")
• 1930s-'40s: 8.5
• 1950s-'60s: 2 so far (per decade)
• 1970s-present: 0 so far

Born in '20s, not in HoF, nor are any of them in the HoM:

Jim Gilliam, Elston Howard, Sam Jones, Minnie Minoso, Don Newcombe, Vic
Power, Al Smith, Artie Wilson

You can probably add a few more. When was Clarkson, e.g?

Thanks to my friend Alan Holst who compiled the following detail.
----------------------------------------------------------------

born 1980-1989

HALL OF FAME 0

NOT IN HALL OF FAME
Albert Pujols, Francisco Rodriguez, Dontrelle Willis

born 1970-1979

HALL OF FAME 0

NOT IN HALL OF FAME
Carlos Delgado, Vladimir Guerrero, Derek Jeter, Andruw Jones, Pedro
Martinez, David Ortiz, Manny Ramirez, Alex Rodriguez, Johan Santana, Alfonso
Soriano, Miguel Tejada

born 1960-1969

HALL OF FAME 1
Kirby Puckett

NOT IN HALL OF FAME
Roberto Alomar, Moises Alou, Albert Belle, Barry Bonds, Joe Carter, Juan
Gonzalez, Dwight Gooden, Ken Griffey Jr., Tony Gwynn, Barry Larkin, Fred
McGriff, Kevin Mitchell, Terry Pendleton, Mariano Rivera, Gary Sheffield,
Sammy Sosa, Darryl Strawberry, Frank Thomas, Bernie Williams

born 1950-1959

HALL OF FAME 3
Eddie Murray, Ozzie Smith, Dave Winfield

NOT IN HALL OF FAME
Harold Baines, George Bell, Andre Dawson, Julio Franco, Ken Griffey, Pedro
Guerrero, Rickey Henderson, Willie Hernandez, Bill Madlock, Willie McGee,
Dave Parker, Willie Randolph, Tim Raines, Jim Rice, Lee Smith, Lou Whitaker,
Frank White

born 1940-1949

HALL OF FAME 6
Rod Carew, Reggie Jackson, Ferguson Jenkins, Joe Morgan, Tony Perez, Willie
Stargell

NOT IN HALL OF FAME
Dick Allen, Don Baylor, Vida Blue, Bobby Bonds, Bert Campaneris, Jose Cruz,
George Foster, Hal McRae, Tony Oliva, Al Oliver, Ken Singleton, Reggie
Smith, Luis Tiant, Jimmy Wynn

born 1930-1939

HALL OF FAME 11
Hank Aaron, Ernie Banks, Lou Brock, Orlando Cepeda, Roberto Clemente, Bob
Gibson, Juan Marichal, Willie Mays, Willie McCovey, Frank Robinson, Billy
Williams

NOT IN HALL OF FAME
Felipe Alou, Matty Alou, Earl Battey, Leo Cardenas, Rico Carty, Curt Flood,
Vada Pinson, Bill White, Maury Wills

born 1920-1929

HALL OF FAME 2
Roy Campanella, Larry Doby

NOT IN HALL OF FAME
Jim Gilliam, Elston Howard, Sam Jones, Minnie Minoso, Don Newcombe, Vic
Power, Al Smith, Artie Wilson

born 1910-1919

HALL OF FAME 6
Willard Brown, Ray Dandridge, Leon Day, Josh Gibson, Monte Irvin, Jackie
Robinson

NOT IN HALL OF FAME
Sam Bankhead, Dave Barnhill, Gene Benson, Pee Wee Butts, Jimmie Crutchfield,
Piper Davis, Luke Easter, Sammy Hughes, Sam Jethroe, Slim Jones, Max
Manning, Buck O‚Neil, Quincy Trouppe, Bill Wright

born 1900-1909

HALL OF FAME 10
Cool Papa Bell, Ray Brown, Martin Dihigo, Bill Foster, Buck Leonard, Satchel
Paige, Hilton Smith, Turkey Stearnes, Mule Suttles, Willie Wells

NOT IN HALL OF FAME
Newt Allen, John Beckwith, Chet Brewer, Bill Byrd, Perucho Cepeda, Rap
Dixon, Vic Harris, Alex Radcliffe, Double Duty Radcliffe, George Scales,
Chino Smith, Luis Tiant

born 1890-1899

HALL OF FAME 7
Oscar Charleston, Andy Cooper, Judy Johnson, Biz Mackey, Louis Santop,
Cristobal Torriente, Jud Wilson

NOT IN HALL OF FAME
Bill Bell, Dave Brown, Dizzy Dismukes, John Donaldson, Fats Jenkins, Dick
Lundy, Dave Malarcher, Oliver Marcelle, Dobie Moore, Alejandro Oms, Roy
Parnell, Dick Redding, Nip Winters

born 1880-1889

HALL OF FAME 6
Pete Hill, Pop Lloyd, Jose Mendez, Bullet Rogan, Ben Taylor, Joe Williams

NOT IN HALL OF FAME
Pelayo Chacon, Bingo DeMoss, Bruce Petway, Spot Poles

born 1870-1879

HALL OF FAME 1
Rube Foster

NOT IN HALL OF FAME
Walter Ball, Pat Dougherty, Home Run Johnson, Bill Monroe

born 1860-1869

HALL OF FAME 2
Frank Grant, Sol White

NOT IN HALL OF FAME
Charlie Grant, George Stovey

born 1850-1859

HALL OF FAME 0

NOT IN HALL OF FAME
Moses Fleetwood Walker

born 1840-1849

HALL OF FAME 0

NOT HALL OF FAME
Bud Fowler

the RA+ numbers for this group seemed very low given their stature in the leagues, it appears that PFs are part of the answer why they were coming out so low before.
. . .
4) This is kind of the tip of this ice berg. With just six seasons from one league to work with, we're only beginning to see the transformative effects of parks on pitchers' estimated RA+.

From a FOHOM standpoint I can't see the mountain. Bell is not an important pitcher. Donaldson and Mendez are not important league pitchers; it's better to retire with a few innings at 95 and 101 than at 89 and 98 but only molehill better. The transformation appears to be limited to the relative standing of two men who waltzed in, Rogan and Foster up three and down six.

--
born 1840-1849

HALL OF FAME 0

NOT HALL OF FAME
Bud Fowler

A 1909 sketch of his baseball career will interest several of you. It is reprinted as the main part of "Bud Fowler (John Jackson)" by Hugh MacDougall, Cooperstown Village Historian, 19c Notes #2005.2, p3.

born 1840-1849

HALL OF FAME 0

NOT HALL OF FAME
Bud Fowler

I'm pretty sure Bud Fowler was born March 13, 1858.

Just to make this a bit more granular. Dates of birth.

1911--W. Brown, Josh Gibson--HoM, HoF
1913--Dandridge--HoF

1913--Trouppe
1915--Clarkson, Easter
1916--L. Day--HoF, played in NeLs at age 19 before W. Brown ever did

1919--Irvin and J. Robinson--HoM, HoF

1920--A. Wilson

1921--Campanella--HoM, HoF
1923--Doby--HoM, HoF

I've said generally black players born in the '10s had the unusual experience of lacking playing opportunities BECAUSE OF INTEGRATION. More accurately, it would appear to be about 1913-1920, depending on the particular trajectory of the player's career. Leon Day was a phenom and got in on the heyday of the NeLs. A. Wilson, whom Gadfly says was actually born in 1916 (and if that's true though, frankly, it doesn't jive with a playing career from 1944-1962), was not a phenom and didn't. Of course, Easter was only a year younger than the alt-Artie and played into the '60s.

Anyway, when you read the table above and see 6 black HoFers born in the '10s, keep in mind all but one were born by 1913 and after 1919, only one from 1914-1918. Yes, that's a narrow little window, and yes, it could be a coincidence. But it could be something else.

Along with "the lost generation," the other gap among black players is at 2B. Here are James' top guys with comments (his and mine).

1. Bingo DeMoss--active 1912-1930, hit about .215 in the NeLs as far as I know from his thread. "Fast." Great small ball player.

2. Newt Allen--active 1923-1944, hit about .300, "Graceful, slick-fielding, quick." Clearly a better hitter than DeMoss but also consider the environment.

3. George Scales--born 1900, active 1921-1948, hit about .313. MLEs figure to .278-.292 (Doc did two versions due to discrepancies in the records) with OPS+ 109-118. Hit with power; Trouppe said he was the best 2B he ever saw. NeL observers favored the slick fielding model, obviously.

4. Sammy Hughes--active 1930-1946, hit .296. James says he's a "complete Ryne Sandberg-type 2B."

5. Bill Monroe
6. Frank Grant

7-10+ are a bunch of guys we didn't even give a thread to.

Then there's Marvin Williams, b. 1920, active 1943-1961, with MLEs of .282 and OPS+ 111. He played only briefly in the NeLs and James' list is the best of the NeLs, so no Marvin. But with the exception of Grant (and maybe Monroe and maybe Sol White, none of whom actually played in the NeLs anyway), isn't Marvin the best 2B that came out of the NeLs? Though I wonder if Allen and Hughes didn't deserve a closer look.

But after Luke Easter, Marvin looks like the best player of "the lost generation," and being possibly the best black 2B in history other than Jackie Robinson (who is actually a year older) seems like a big thing.

And we know that from the end of WWII through the integration of the Boston Red Sox in 1959 was the toughest time for a black player to make a mark since before 1920. That exactly prescribes Marvin's (as well as Artie Wilson's and Bus Clarkson's) career. The only HoMers or HoFers who made the MLs who were as old as any one of the three are Jackie and Irvin, two guys and two guys only. The very best black player born within 3 years of Jackie and Irvin would be Marvin Williams, IMO, and he couldn't play in the MLs???

Sunny,

Bobby Estallela born 1911. Marv Williams born 1920.

Of course, part of what's going on here is that Dandridge and Day are mistake selections, so really let's talk HOMers only:
1911: 2
1919: 2
1921: 1
1923: 1

So a 12 year period where only 6 guys are HOMers. Is that reasonable? Is it reasonable that there's zero guys between 1911 and 1919?

How's that compare to the overall HOM?
1911: Greenberg, Medwick, Brown, Gibson
1912: Vaughan
1913: Mize
1914: Dimaggio
1915: Gordon
1916: Slaughter
1918: Doerr, Feller, Reese, Williams
1919: Irvin, Robinson
1920: Lemon, Musial, Wynn
1921: Campy, Newhouser, Spahn
1922: Wilhelm
1923: Doby
Total: 23, 6 black, 17 white

Is about 1/4 of electees from the period reasonable? Well, let's note that there were two major leagues and two negro leagues with generally 28-32 teams operating at any moment en toto. And let's say that you don't think that all the NgL teams were at a truly MLB level. Remembering that black players were scattered about the hemisphere and contracting the NgLs a bit to shore up their total quality, call it 24-28 teams of MLB quality baseball, of which 1/3 to 2/5 are black teams. Even this very basic analysis suggests we're a bit shy in this period on black players. Perhaps by one, perhaps by several. Depends on your point of view.

Yeah, and of course it comes down to cases. Is Marvin Williams better than Nellie Fox? Clarkson better than Sewell? Artie Wilson better than Rizutto?I don't know yet. I am in the middle of a major re-eval incorporating Joe's PA numbers for pitchers as well as a new look at the "lost generation." And that's enough players right there to force a re-eval of pretty much everybody. So I don't know where these guys are going to end up, but I do know that they are a lot better than I had thought at the time when they first became eligible.

I'd say 6 out of 29 is about 2 too many demographically speaking, based on population percentage. White kids born in 1911-23 were zapped by the Depression, so equally likely to choose MLB rather than something more lucrative -- only after 1925 or so did the postwar boom factor in.

We have had this discussion ad infinitum, but there are two sides to it.

And just for the record, the other side is that HoMers (and well, ML caliber athletes generally) are all outliers anyway, and would not in any way be expected to reflect demographics. I mean, how many blacks should we have in the boxing or basketball HoM? Based on population percentage, I mean?

The answer of course is none, because you would never even think to base membership in a boxing or basketball HoM on population percentage. So why baseball?

I have pointed this out before.

There are no HOM pitchers born 1909-1917,
that is, between Ray Brown (1908) and Bob Feller (1918):

1908 3 Ferrell RBrown Wells
1909 3 Ott Herman Hack
1910
1911 3 Greenberg Medwick Gibson
1912 1 Vaughan
1913 1 Mize
1914 1 DiMaggio
1915 2 Gordon WBrown
1916 1 Slaughter
1917 1 Boudreau
1918 4 Doerr Reese TWilliams Feller

Given that one would expect pitchers to be 30% of the HOM population, this is a more significant dearth (becaue it is longer in duration) than the shortage of Negro Lague players cited above.

BTW, Dr. Chaleeko, does bbref.com have a bad birthdate for Willard Brown?
That's the source for my 1915 bdate. Your list has him as 1911.

Here's some other names to add to the list Sunnday put together to give a sense of the range of players potentially effected by the integration period.
1911: Pullman Porter
1912: Bob Griffith, Booker McDaniels
1913:
1914: Silvio Garcia, Theolic Smith, Bill Wright, Dave Barnhill
1915: Gene Bremer, Roberto Ortiz
1916: Johnny Wright
1917: Piper Davis, Claro Duany, Bob Thurman
1918: Max Manning
1919: Pee Wee Butts
1920: Dan Bankhead, Sam Hairston, Jim LaMarque
1921: George Crowe
1922: Sam Jethroe, Bonnie Serrell, Wilmer Fields, Connie Johnson
1923: Rene Gonzalez, Alonzo Perry

Coming close: Bobby Avila (1924), Joe Black (1924), Minnie Minoso (1925), Bob Boyd (1925), Suitcase Simpson (1925), Hank Thompson (1925), and Toothpick Sam Jones (1925).

BTW, Dr. Chaleeko, does bbref.com have a bad birthdate for Willard Brown?
That's the source for my 1915 bdate. Your list has him as 1911.

I just picked it off Sunny's list without double checking (hey, I trust the guy). So your reference may be correct. He's listed as 1911 in the Clark/Lester book, however.

Thanks for the additional names, Doc. I should take another look at Silvio Garcia.

Silvio...weren't we just talking about Bob Dylan somewhere.... Another one. How many ballplayers has he sung about anyway?

jimd, are there any pitchers currently under consideration who were born between 1908 and 1918?

Bucky is 1910, Dutch 1909, and the Dizzies are 1909 and 1915.

I think there is a significant difference between an almost decade-long gap between black HoMers and a similar gap between pitchers.

White pitchers (at least) had a fair shot at having a career, and at leaving a legacy that obsessive compulsives like us could one day look at and evaluate for HoM-worthiness.

Black players born before Jackie Robinson did not. And even then, the ones who got to play in the NeLs ca. 1920-1945 had a reasonable fascimile of a chance. Those who came of age 1945-1955 really did not (like the guys who came of age before 1920).

I'm not interested in filling gaps that have the appearance of being random, but I am interested in filling gaps that are social and cultural. Or to put it another way, if on the day that a player was born, he was already destined to not get a fair chance, then some adjustments are appropriate. If on the day he was born he had a chance, and if he got that chance, and just didn't end up having a HoM career, I can't get too worked up about the unfairness of that.

I'm not interested in filling gaps that have the appearance of being random, but I am interested in filling gaps that are social and cultural.

Isn't there an argument to be made that the 1908-18 pitcher gap is an effect of WWII? Except for the early-blooming Dizzy Dean, all of the pitchers currently under consideration from that birth decade had part of their prime during WWII. The three under consideration didn't spend time in the military. What of those who did, like, say, Johnny Sain (b. 1917)?

Chris, well, yes, good point. I am interested in filling any gaps in the WWII generation. They were destined at birth to lost part of their career. But I don't care too much whether they're pitchers or not. Players at all positions were equally affected with random variations.

How many ballplayers has he sung about anyway?

You'd be surprised. Some of them even predict players who weren't even born yet...:
It Ain't Me, Babe
The Lonesome Death of Jamey Carroll
Motorpsycholyons Nightmare
Song to Woody Williams
Jimmy Outlaw Blues
Gates of Edens
It's All Over Now Vida Blue
It's All Right Ma, I'm Only Riedling
Desolation Rowe
Obviously Five Relievers (Tony LaRussa's Blues)
I Dreamed I Saw Don August's Team
Dear Gaylord
The Wickman Messenger
Positively Huston Street
Mighty Jack Quinn (Quinn the Eskimo)
The Million Lollar Bash
Pucketts of Rain
Going to Apodaca

There's a man that loves baseball....

I've been doing some research into which Negro League teams actually won pennants. I'd like to run my research by the rest of the group, just to make sure I haven't made any errors and to possibly fill in a few holes.

NNL 1
1920 Chicago American Giants
1921 Chicago American Giants
1922 Chicago American Giants
1923 Kansas City Monarchs
1924 Kansas City Monarchs
1925 Kansas City Monarchs
1926 Chicago American Giants
1927 Chicago American Giants
1928 St. Louis Stars
1929 Kansas City Monarchs
1930 St. Louis Stars
1931 St. Louis Stars

ECL
1923 Hilldale Daisies
1924 Hilldale Daisies
1925 Hilldale Daisies
1926 Bacharach Giants
1927 Bacharach Giants

1929 Baltimore Black Sox

(1930 Homestead Grays
(1931 Homestead Grays

1932 Cole’s American Giants

NNL 2
1933 Cole's American Giants
1934 Philadelphia Stars
1935 Pittsburgh Crawfords
1936 Pittsburgh Crawfords
1937 Homestead Grays
1938 Homestead Grays
1939 Homestead Grays
1940 Homestead Grays
1941 Homestead Grays
1942 Homestead Grays
1943 Homestead Grays
1944 Homestead Grays
1945 Homestead Grays
1946 Newark Eagles
1947 New York Cubans
1948 Homestead Grays

NAL
1937 Cincinnati Tigers
1938 Kansas City Monarchs
1939 Cleveland Bears
1940 Kansas City Monarchs
1941 Kansas City Monarchs
1942 Kansas City Monarchs
1943 Birmingham Black Barons
1944 Birmingham Black Barons
1945 Cleveland Buckeyes
1946 Cleveland Buckeyes
1947 Cleveland Buckeyes
1948 Birmingham Black Barons
1949 Baltimore Elite Giants
1950 Indianapolis Clowns

1. Which teams, if any, would be considered Eastern champions before the formation of an official league (the ECL), particularly '20, '21 and '22?

2. There wasn't much in the way of league continuity in the east from '28 to '32. I've seen the Black Sox described as champs for '29, and the Grays for '30 and '31. What about '28 and '32? And do I have the other three correct?

3. There's a western gap as well, between the end of the first NNL and the debut of the NAL. Which teams would be considered the western champs for '32-'36?

4 Do I have the NAL right? For example, I've seen the Tigers listed both as champions and as the third-place team for 1937.

5. I've seen descriptions of the American Giants as champions in '32 but was that in the west or the east?

And here's why I'm asking. We often use the phrase around here that "a pennant is a pennant is a pennant." So I've been checking out the crossover between pennant-winning teams and Hall of Merit players. After all, the best teams are often going to have the best players. And it's unlikely that a team could win multiple championships without having a HOMer or two in the crowd.

Now, I realize you can go too far with that theory, especially if you look at it from the opposite side (starting with the player, rather than the team). Being a part of four championship teams doesn't make Paul O'Neill and Tino Martinez HOMers. And it appears that this line of reasoning was partly responsible for the induction of Judy Johnson, who played for several different dynasties. So I'm not for a second pushing for every player who wins a bunch of pennants to get into either Hall.

Yet I do think there is some validity in looking at it from the team angle. The Chicago American Giants won three pennants from 1920-22 and from that team, we've selected Rube Foster and Cristobal Torriente. The Monarchs multiple winners gave us Jose Mendez, Joe Rogan and a bit of John Beckwith. The St. Louis Stars gave us Cool Papa Bell, Mule Suttles and Willie Wells. Looking at the NNL of the '20s, each of the multiple championship teams gave us multiple HoMers.

The same is true over in the east. The Hilldale Daisies had Mackey and Santop and one year of Lloyd. The Homestead Grays had Charleston, Gibson and Williams. Yet sandwiched between the two we find another championship team: The Bacharach Giants. They won back-to-back pennants in '27 and '28 and yet not a single HoMer represents that team. It strikes me as a little odd that five of the multiple champions of '20-31 have multiple HoMers while the sixth has none.

Moving into the '30s and '40s, we've done a great job of representing the eastern teams and the second NNL. The American Giants have Bill Foster and Turkey Stearnes. The Crawfords Bell and Gibson and Paige. The nine-time champion Grays have Brown and Gibson and Leonard, as well as Cool Papa Bell and Jud Wilson before they're done. And even the one-time champs often have a HoMer such as Wilson for the Stars and Irvin for the Eagles.

But things don't seem so even-handed over in the west and the NAL. The Monarchs are represented well with Paige and occasionally Turkey Stearnes or Willard Brown. Yet the other NAL franchises seem to be ignored. The Black Barons won three pennants yet they have no HoMers to show for it. The Buckeyes won three in a row, yet they too have been neglected.

Maybe that's right. You can certainly win a pennant without a Hall of Fame player, can't you? Except I can't name too many that have done so, especially not a team that won two or three in a row. Ignoring the solo winners, there are three franchises that had multiple pennants yet no HoMers. It seems odd to me. It makes me wonder: are we underestimating Dick Lundy, Oliver Marcelle and the "million dollar infield" of the Bacharach Giants (which moved almost in toto to the Baltimore Black Sox)? I find it interesting that two of the franchises are also a part of the underrepresented 1940s and they happen to include some of the lost generation players that sunnyday has been shedding light on. Artie Wilson played for the Black Barons (though I confess I still think Bus Clarkson is the better candidate). Sam Jethroe and Quincy Trouppe played for the Buckeyes. Have we been fair to the NAL?

Okay, I may have gotten a bit rambly there. At least help satiate my curiosity as to which teams were considered champions.

Very quick answers to some of your questions:

1. Whenever there isn't a league (the east, 1920-22), of course it's hard to say who the champions were. I can post what records I have for eastern teams in those years, though it might take some time to put them together. One complication is that, at least in 1921-22, Hilldale and the Bacharach Giants were affiliated with the NNL (though they didn't compete for the championship), and did not play the other important eastern teams (Lincoln Giants, Cuban Stars, Brooklyn Royal Giants).

2. The ECL disintegrated in late May / early June 1928. The league only had five teams: Baltimore Black Sox, Bacharach Giants, Lincoln Giants, Cuban Stars, and Philadelphia Tigers. For what it's worth (not very much), Baltimore probably had the best record when the league failed, at 5-2. There was no press consensus about who the "champion" of the east was that year, though I've always thought that Hilldale had the strongest claim to the title, for various reasons. That's very disputable, though.

The 1929 American Negro League pennant was won by Baltimore, as far as I know. There was no league in 1930 or 1931, so no champions. In 1932 Cum Posey organized an East-West League--since it folded mid-season, I'm not sure if there was a formal champion, but the Detroit Wolves seem to have had the best record.

3. About 1932-36 in the west, I don't know. The second, Gus Greenlee-led NNL considered itself national, and had midwestern teams in those years (the American Giants, most notably). The Monarchs were almost certainly the best non-league team in the region then, though they spent much of their time barnstorming all over the country.

4. I think the Monarchs were NAL champions in 1937, 1939-42, with the Memphis Red Sox champions in 1938.

5. The American Giants played in the Negro Southern League in 1932.

One general comment: Holway's W/L records do not tell who was the official champion of any given year, since they count every game against top black competition, not just games against league opponents. I can't say for sure, but I don't think the Cleveland Bears or Cincinnati Tigers were ever considered NAL champions at the time.

Thanks, Gary.

1. I look forward to seeing the 1920-22 records.

2. I knew about the league collapse in 1928. As for who would be considered the champion of the non-league, that's admittedly speculative. However, I've heard the Black Sox often described as having only won one championship in '29. Now that could be because 1928 doesn't count. Or it could be that the Homestead Grays are considered the champs for that season.

4. Can you cite a source for the NAL 1937-1942? I admit that I was surprised to see Cincinnati and Cleveland listed as champs in '37 and '39. However, I didn't get that from Holway. That's how it's listed in the Major League Sports Almanac. It sounded dubious to me but it was the best information I had at the time. Also, the Almanac lists the Chicago American Giants as runners-up for '37, Memphis for '38 and Kansas City for '39.

5. Thanks, that's right. I even have the American Giants listed as southern champs.

Clark/Lester's Negro Leagues Book has this about the NAL:

1937: First half champion-Kansas City; second half-Chicago; NO PLAYOFFS.
1938: First half-Memphis; second half-Atlanta; Memphis won playoffs, 2 games to 0. (KC had best record overall, but didn't win either half.)
1939: First half-Kansas City; second half-St. Louis; KC won playoffs, 3 games to 2.
1940: Kansas City won both halves, no playoffs.
1941 and 1942: Kansas City league champions, no split season or playoffs.

Also, in 1939 the Homestead Grays won both halves of the NNL, but a special postseason tournament was won by the Baltimore Elite Giants, which put the pennant into dispute. I'm not exactly sure what the official league position was.

The Negro Leagues Book lists official standings, as they were published in newspapers at the time.

Actually, it would be good to look up the Chicago Defender, which is online via SABR's ProQuest sub for these years, and see what it says, particularly about disputed pennants. I don't have time to do it at the moment, but might get to it later this weekend.

Once again, thanks for the help, Gary. Your records for the early NAL certainly sound better than what I originally had. As I said, it struck me as pretty dubious that either the Tigers or the Bears were considered pennant winners. Thanks to your listings, I was able to look around and verify a lot of what you posted such as a '38 pennant for the Memphis Red Sox (though Atlanta disputed the championship as the series was cut short). I'll leave the '28 eastern pennants and the '37 western ones as disputed for now. And you're right that the NNL tried to be a national league not an eastern one- they had Chicago, Nasnville, Columbus, Indianapolis and Detroit (all traditionally western cities) in the original line-up. Western teams slowly moved east or dropped out over the years and by '36, the most western club was in Pittsburgh.

So here's what I have now. There's still a couple of blanks but it's certainly in better shape than it was yesterday.

West
NNL 1
1920 Chicago American Giants
1921 Chicago American Giants
1922 Chicago American Giants
1923 Kansas City Monarchs
1924 Kansas City Monarchs
1925 Kansas City Monarchs
1926 Chicago American Giants
1927 Chicago American Giants
1928 St. Louis Stars
1929 Kansas City Monarchs
1930 St. Louis Stars
1931 St. Louis Stars
NSL
1932 Cole's American Giants
1933-1935- no western champs, NNL 2 is sole major league containing teams from east and west
1936- no western champs (?)
NAL
1937 Chicago American Giants/Kansas City Monarchs
1938 Memphis Red Sox
1939 Kansas City Monarchs
1940 Kansas City Monarchs
1941 Kansas City Monarchs
1942 Kansas City Monarchs
1943 Birmingham Black Barons
1944 Birmingham Black Barons
1945 Cleveland Buckeyes
1946 Cleveland Buckeyes
1947 Cleveland Buckeyes
1948 Birmingham Black Barons
1949 Baltimore Elite Giants
1950 Indianapolis Clowns

East
ECL
1923 Hilldale Daisies
1924 Hilldale Daisies
1925 Hilldale Daisies
1926 Bacharach Giants
1927 Bacharach Giants
1928 Baltimore Black Sox/Hilldale Daisies
ANL
1929 Baltimore Black Sox
non-league championships
1930 Homestead Grays
1931 Homestead Grays
1932 (?)
NNL 2
1933 Cole's American Giants
1934 Philadelphia Stars
1935 Pittsburgh Crawfords
1936 Pittsburgh Crawfords
1937 Homestead Grays
1938 Homestead Grays
1939 Homestead Grays
1940 Homestead Grays
1941 Homestead Grays
1942 Homestead Grays
1943 Homestead Grays
1944 Homestead Grays
1945 Homestead Grays
1946 Newark Eagles
1947 New York Cubans
1948 Homestead Grays
1949-1950- no separate eastern champs as NNL is absorbed into NAL

Answering another question: I've now found a source claiming that the Baltimore Black Sox were the East/West League champs in 1932.

One more thing: the Monarchs won the 1946 NAL pennant, losing the world series to Newark 4 games to 3 (from Clark/Lester).

I've put together some data on the Bacharach Giants and Hilldales for 1921 and 1922. They did not play the Lincoln Giants, Brooklyn Royal Giants, or Cuban Stars during these seasons, but did play many games against NNL teams, as they were associate members of the league.

1921 vs. NNL   W  L  T  (runs scored-runs allowed) 
Hilldale Club    12  7  1  (116-77)
Bacharach Giants 24 22  2  (268-209)

1921 Hilldale vs. Bacharach
Bacharach Gts     9  8  0  (86-87)
Hilldale Club     8  9  0  (87-86)

1922 vs. NNL
Bacharach Giants 17 28  1  (218-241)
Hilldale Club     6 12  0  (64-95)

1922 Hilldale vs. Bacharach
Bacharach Giants  6  4  0  (61-67)
Hilldale Club     4  6  0  (67-61)

In 1921, the Bacharachs toured the midwest twice, while also entertaining four NNL clubs (Chicago American Giants, Columbus, Detroit, and Indianapolis) in the east, both in Atlantic City and at neutral sites. Hilldale, on the other hand, didn't tour the midwest, so all its games against NNL teams were at home or at neutral sites on the east coast.

In 1922, however, both teams toured the midwest, while only Indianapolis and Cleveland went east.

This is what I found in the Chicago Defender about Negro League pennants for a few "uncertain" seasons:

1937: Kansas City defeated Chicago 4 games to 1 for the first NAL title (Defender, Sept. 25)

1938: Memphis led Atlanta 2 games to 0 when NAL president Robert Jackson cancelled the series out of exasperation with the two clubs, who were bickering over the scheduling of the games. (Defender, Oct. 1)

1939: Kansas City defeated St. Louis 3 games to 2 for the NAL pennant (Defender, Sept. 9). The NNL situation was pretty complicated. At season’s end, the top four of the league’s six clubs engaged in a playoff tournament for the “Jacob Ruppert Memorial Trophy,” which, the Defender said several times, represented the league championship. These were the overall final standings as reported in the Defender (Sept. 9):

Homestead 33-14
Newark 29-20
Baltimore 25-21
Philadelphia 31-32
Black Yankees 15-21
NY Cubans 5-22

(Note that these totals don’t balance.)

The Defenderalso reported the second half results (which do balance):

Baltimore 13-8
Homestead 10-7
Philadelphia 16-12
Newark 11-14
Black Yankees 6-10
Cubans 2-7

It would seem that the Grays probably won the first half.

In the playoff tournament, Homestead beat Philadelphia in the first round, while the Baltimore Elite Giants beat Newark. That meant the final matched the winners of the first half with the winners of the second half. The Elite Giants won, making them champions of the NNL (Defender, Sept 9, 23, 30).

1946: The Kansas City Monarchs were NAL champions, winning both halves (Defender, Sept. 14), then losing to the Newark Eagles in the world series (Defender, Oct. 5).

Great work, Gary. That just leaves us with the ECL collapse of 1928 and whether or not Baltimore's 5-2 record counts as a pennant or whether Hilldale's independent record trumps it. Oh, and the possibility that the Detroit Wolves won the East/West League instead of Baltimore.

I haven't done an exhaustive search, but on looking through the 1932 Chicago Defender I found in the July 16 issue an article noting that the Homestead Grays had been awarded the first-half pennant of the East-West League, the source being league president Cum Posey (also the Grays' owner).

Great work, Gary. That just leaves us with the ECL collapse of 1928 and whether or not Baltimore's 5-2 record counts as a pennant or whether Hilldale's independent record trumps it.

There's probably not anyone more of an expert on 1928 ECL than Gary, so if he can't name an ECL winner, probaby no one can, but for what it's worth, here are the records of the 'Big 6' Eastern teams against each other per Gary's research:

Thanks for posting that. Here's how I see it:

Hilldale, even though they were outscored, had the best record among these teams--plus they played all the other teams, played more games against them than anybody else, and divided their schedule evenly between home and road games (30-29). They had a winning record against everybody but the Lincoln Giants (5-6) and Homestead Grays (4-4). Ironically, they weren't even a member of the ECL in 1928, having dropped out before the season started.

Lincoln Giants were slightly more home-biased, playing 21 at home and 13 away, plus they didn't play Baltimore. A winning record against Hilldale (6-5) and Bacharachs (4-1), but a losing one to the Grays (3-5) and Cuban Stars (4-5-1).

Homestead didn't play the Bacharach Giants or Cuban Stars, and played the fewest games among the top six.

The Bacharachs have a better argument than I'd realized, though they didn't play the Homestead Grays, and dropped four out of five to the Lincolns. They were nearly even with Hilldale (7-8-1, outscoring them 85-83).

Baltimore had a losing record against every other team except the Cuban Stars (7-0-2).

The Cuban Stars didn't play Homestead, and had a losing record against everyone else except the Lincoln Giants.

Thanks a lot, KJOK. That does seem to confirm what Gary wrote in post #351: that Hilldale has the best claim to a pennant for 1928.

"Is Marvin Williams better than Nellie Fox? Clarkson better than Sewell? Artie Wilson better than Rizutto?"

Looking through my ballot I'm starting to lean towards Yes, Maybe and No. Clarkson is intriguing. The offensive numbers are better than Sewell but there is no information on defensive reputation (though he did play multiple positions well). If he's only a mediocre SS then Sewell was better. If he's better than a mediocre SS but not anything special then I think I'd rather have Bancroft and Sewell. If he's good enough to stay at SS for his whole career, he's on my ballot to stay. I think we've done a good job gathering offensive numbers but we have scant evidence on the defensive abilities of our remaining NgL candidates beyond the all-glove candidates like Dick Lundy and Ray Dandridge. I'd love to have some better information on Clarkson and Trouppe in particular.

I'm very slowly revisiting my NgL candidates and MLEs and whatnot. One thing that's long puzzled me is the question of how different walk rates might have been in the Negro Leagues, and how to translate that difference in MLEs. I had a day off today, so I thought I'd tackle the question. Here's what I did and what I found.

1) I used the SBE to figure out what the MLB walk rate was for 1920-1948. I defined walk rate as BB/AB since these were the two pieces of info I had from the Shades of Glory project's translations. The MLB walk rate from this period was .094.

2) I inputted the NAL/NNL/ECL/ANL/NSL/EWL player stats from Shades of Glory and the HOF .pdf released during the NgL comission's work last year. These include walks. I also cribbed league averages from the .pdf document. I had a group of 34 players totaling ~60,000 NgL ABs. They walked at a rate of 1 per 10 ABs (.1028 to be exact).

3) But I don't have NgL league OBP or walk data to directly compare. I tried this workaround. I found the relative AVG and SLG of the NgL group. These turned out to be 120 and 133. Then I used the SBE to generate of list of comparable players: 3000 PA during the period in question, a relative average (versus MLB with pitchers, to match the information supplied on the .pdf) of 110-130 and a realtive SLG of 123-143. It spat back 24 names
Averill
Cobb
Dickey
Fournier
Goslin
Hafey
Heath
Heilmann
Herman
B Johnson
Keller
Klein
Kurowski
Medwick
O'Doul
Ott
A Simmons
Slaughter
Speaker
Stephens
Terry
Trosky
Wheat
K Williams

Good group. Ott's probably the best player in this period, but there's lots of solid HOMers here. Our NgL sample is a little more top heavy (Charleston and Gibson) but also a little more bottom heavy (Dandridge, marcell, and Johnson), so I think it's a useful comp list. This bunch of players walked .112 times per AB. This rate was 1.193 times the MLB average. The total sample was approx. 132,000 ABs.

At its most reductive this information could conceivably be shown as a proportion:


If true, this suggests that the NgL walk rate was around .086, or 8.5% lower than the MLB rate.

I should note, however, that the NgL stats are not weighted and some players have very few ABs (Robinson, for instance) or have quite a few (allen is over 3000). If instead I apply every single player's walk rate to the lowest number of ABs (Jackie's 53), the walk rate rises a smidge to .1036. If the proportion holds that means the NgL walk rate was .087 per AB.

OK, so then question is, does this proportion hold? It might. As I mentioned, I've got a wide variety of players among my 34 NgLers, some walk, some don't. Most players on my MLB list did walk, some a lot. I can think of two reasons for doubt:

1) The NgLs were more uneven in terms of total quality than MLB, suggesting that the best players might have walked more often than they might against leagues of more uniform quality.

2) The style of play was more contact and speed oriented than MLB was, meaning there was greater incentive to put bat on ball. In addition, I looked at SH per AB totals, and they are 35% higher among the NgL guys than the MLB guys, possibly supressing NgL walk totals.

So on the whole, I'm led to think that this is useful information to have, though incomplete. In previous iterations of MLEs, I've used a static model of walk rate, keeping it the same from NgL to MLB. I think now that I will not, but instead will compare the player to the study period walk rate.

But more interesting, this information could open up some avenues of inquiry into league RS and R/G totals, giving the possibility of estimates that can help us compare a player to his run evironment more accurately. That would be reallllly helpful information.

that's it, just thought i'd report in on it for anyone interested.

Crossposting:

DL from MN Posted: March 19, 2007 at 10:24 AM (#2314017)

Since Holway hasn't logged in I'll quote his e-mail:

"Clarkson is a nobody. Lundy is one of the Big Three shortstops. Hit
30 points higher than Wells. Took (an older) Lloyd's job away from him.
Won three flags as a manager. Considered the smoothest fielding shortstop.
A travesty that he didnt make the Hall of Fame. I'd have named him in the
first five on the latest ballot."

andrew siegel Posted: March 19, 2007 at 10:28 AM (#2314023)

There are two HoM arguments for Lundy:

(1) We are underrating the Bancroft-Rizutto-Long excellent glove, ok bat SS's, and Lundy belongs at the top of the list of those guys.

(2) We are underrating Lundy's hitting by a rung or two and he really would have been a 105 OPS+ guy in the bigs which with his fielding and his career length would make him an obvious HoMer.

I have no particularly reason to believe either argument is true, but either or both certainly could be.

DL from MN Posted: March 19, 2007 at 03:51 PM (#2314237)

I guess this is what we have to answer:

06) Nellie Fox 348pts
34) Phil Rizzuto 109pts
39) Dave Concepcion 99pts
42) Dave Bancroft 72pts
50) Vern Stephens 51pts
62) Fred Dunlap 62pts
72) Luis Aparicio 24pts
76) Rabbit Maranville 23pts
82) Herman Long 16pts

Dick Lundy - no votes

Chris Cobb Posted: March 19, 2007 at 04:09 PM (#2314252)

One of the things that really hurt Lundy's MLEs in my original study of him was his _really_ anemic walk totals, according to the best data we had at that time. I have not re-done MLEs using the HOF data, partly due to lack of time in my life for that work and partly because the borderline HoM NeLers have had their cases depend more on MxL and minor league data.

A quick trip over to the HoM site to check Lundy's stats suggests that his walk totals during his prime may be higher than the old data indicated. I'll have to check that when I am at home and have access to my MLE spreadsheets. He's no walking machine, but a revised view of his OBP might help his OPS+ significantly. Given how badly his hitting trailed off in the mid-1930s, though, I have a hard time seeing him sustaining a major-league career of more than 13-14 seasons, and it seems unlikely that he was a better hitting/fielding package than Dave Bancroft, who is just hanging on in our balloting. We are overrating Fox, but if Lundy's comparables are the Rizzuto/Concepcion, Bancroft/Aparicio/Maranville/Long group, then it seems unlikely that our undervaluing of Lundy (if there is any), has been so great as to keep him from deserved election to the HoM.

I'll post again on this matter once I have had a chance to look at the data I used for Lundy's MLEs.

Chris Cobb Posted: March 19, 2007 at 05:51 PM (#2314331)

More on Lundy:

Lundy's walk rate presented in the HoF data is significantly higher than the rate used in his original MLE projections, which was based off of limited data from 2 seasons.

In the MLE, Lundy's walk rate was .0483 BB/PA for his career. That is a rate adjusted upward due to the low BB environment of the NeLs in comparison to the majors.

In the HoF data, Lunday career walk rate is .0818 BB/PA for his career. That rate should be adjusted upward to approximate a major-league rate. That's a very significant change, and a change that is supported by a large data set.

Even if it were not adjusted upward, that rate will move Lundy substantially up as a hitter. In the MLE formulas, BBs, hits, and total bases are interconnected, so I can't do revised estimates quickly, and I won't have time to work through the formulas until later this week, but, in the last set of MLEs I posted, Lundy showed up with a career OPS+ of 92 through 1935. My guess is that this change will move him up to around 100, depending on how expanding his numbers to full major-league seasons and regressing them affects the whole workup. It seems a safe prediction that he will look like the offensive equivalent of Dave Bancroft (career OPS+ of 98). I think it very unlikely that he will show up as the equivalent of Joe Sewell (OPS+ of 109, and OBP-heavy), or Willie Wells (who, iirc, had a career MLE OPS+ in the 110-115 range -- my data on hand don't have the career value), but I won't categorically exclude the possibility of a larger rise until I have actually put the data through the full process -- there are too many factors involved to make an eyeball estimate trustworthy, except insofar as to say that his MLEs are going to rise significantly.