Baseball for the Thinking Fan

Login | Register | Feedback

btf_logo
You are here > Home > Hall of Merit > Discussion
Hall of Merit
— A Look at Baseball's All-Time Best

Tuesday, August 17, 2004

John Henry Lloyd

Was he the “black Honus Wagner” or was Wagner the “white Pop Lloyd?”

John (You Can Call Me Grandma) Murphy Posted: August 17, 2004 at 09:22 PM | 104 comment(s) Login to Bookmark
  Related News:

Reader Comments and Retorts

Go to end of page

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

Page 1 of 2 pages  1 2 > 
   1. DavidFoss Posted: August 17, 2004 at 09:28 PM (#802485)
OK... similar projects that induct a token level of 4-5 Negro Leaguers always include Pop Lloyd. He may be a shoo-in, but I'm still interested in seeing just how good he was.
   2. The definitely immoral Eric Enders Posted: August 17, 2004 at 10:12 PM (#802536)
The really interesting question on Lloyd is not whether he is a deserving HOMer, but whether he was better than Honus Wagner.

"You could put Wagner and Lloyd in a bag together, and whichever one you pulled out you couldn't go wrong." --Connie Mack.

Babe Ruth reportedly said in an interview with Graham McNamee that Lloyd was the best baseball player he'd ever seen.

Consensus of Negro League historians would probably make Lloyd the fourth-best Negro League player, after Paige, Gibson, and Charleston (those 3 not necessarily in order).

Holway, in Blackball Stars, has him batting .327 (34 for 106) in 29 career games against major leaguers. But then, bizarrely, in the Complete Book, Holway has him batting just .241 in such games (41 for 170). Where that extra 7-for-64 came from is anyone's guess, but I suspect some sort of clerical error on Holway's part. (Unfortunately we don't know whether that error, if it was indeed made, was made in the older book or the newer one.) Ultimately the discrepancy should have zero bearing on his candidacy, which is quite strong enough already, but it would be nice to know.

Remarkably, Lloyd won the 1928 Eastern Colored League batting title with an average of .564 -- at age 44. That batting average stands as the Negro League single-season record.

Lloyd almost certainly has more career value than Wagner, though I'd suspect Wagner had a better peak. Lloyd was active from age 22 through age 48 -- 27 years -- and was a superstar for almost all of that time. (He did gradually move rightward on the defensive spectrum, moving to 2B in 1923 and 1B in 1928.)

Sadly, he ended up working his retirement years as a janitor in the Atlantic City school system.

Less sadly, in his retirement he was also instrumental in founding and running Atlantic City's Little League program, which is one reason the stadium there is still named after him.
   3. jonesy Posted: August 18, 2004 at 11:28 AM (#803779)
Eric,

I'm just finishing up Neil Lanctot's NLB - The Rise and Ruin of a Baseball Institution. I was taken with his repeated and strong comments about the poor record/stat keeping as recent as the late 1930s (not that we didn't know this), but was particularly enlightened when he said that Cum Posey (one of the few managment people that did send in boxscores) was known to have been fudging numbers to make his team look better.

Long ago I came to the conclusion that Holway is the least reliable of the SABR Negro League reseachers (Lester, Clark, Knorr, Riley) when it comes to actual numbers. I'll stop short of saying more.

My pet peeve is why only count the numbers accumulated versus major leaguers. If documentation is located, and there are many boxscores present, of Negro League teams playing versus white semi-pro teams, then why shouldn't that data be included? The bigger the sample provides the clearer picture, doesn't it?
   4. The definitely immoral Eric Enders Posted: August 18, 2004 at 11:56 AM (#803785)
The bigger the sample provides the clearer picture, doesn't it?

In this case, no, I don't think it does. I'd much rather know what Lloyd did in 29 games against Chief Bender, Addie Joss, and Eddie Plank than have a larger sample polluted by 100 Joe Highschoolers. I think the numbers against major leaguers, small sample size though they may be, tell us a lot about the level of play in the Negro Leagues vis a vis the majors, and also about the relative skills of the individual players. Ideally, I like to see both -- the record against major leaguers, followed by the record in all games.

You're correct that Holway is easily the least reliable of the Negro League researchers, and I say this as someone who has met John and found him to be a friendly person. I admire much of his work, but his research is often careless and sloppy, and his conclusions often erroneous. Yet his work is valuable because he's done so much more research than anyone else in this field. The sheer volume is amazing although the quality is uneven. (Part of the reason for the uneven quality is that for many years, Holway was literally the only person doing Negro League research, and this was before SABR, before the acceptance of semi-academic standards in baseball scholarship. Holway is Harry Wright, whereas Dick Clark and Larry Lester are Cobb and Ruth.)

Larry Lester, Dick Clark, and Ted Knorr, in addition to being first-class human beings, have always done extremely accurate and reliable work. Jim Riley's work is slightly less accurate than theirs, but like Holway, he makes up for it in volume.

Cum Posey may indeed have been fudging numbers, but all the stats we're using have been compiled by historians going back through old box scores. So Posey's nefariousness hasn't worked where history is concerned.

I suppose it's already well known around here, but MLB and the Hall of Fame recently commissioned an academic study of the Negro Leagues that's being led by Larry Lester, Dick Clark, and Prof. Larry Hogan. One of the things they're doing is compiling an exhaustive statistical record (for league games only) by combing virtually every still-existing black newspaper archive for box scores. Their ambition is to create the definitive statistical record of the Negro Leagues, and I have no doubt that they will succeed in doing so. Plans are to have day-by-day records available for every player in the league's history. It's bad timing for Hall of Merit purposes, as their study is projected to be finished and published within the next couple of years.
   5. The definitely immoral Eric Enders Posted: August 18, 2004 at 12:03 PM (#803788)
Wait-- jonesy, is Lanctot saying that Posey was fudging stat totals, or individual box scores? The former strikes me as entirely possible, the latter as far-fetched, given that thousands of witnesses were generally present. That would be very hard to pull off. Especially since he couldn't change R and RBI totals without changing the score of the game, too.
   6. karlmagnus Posted: August 18, 2004 at 12:43 PM (#803802)
Do we have I9's for Pop Lloyd? For us dummies they make life much simpler (e.g. they told me Poles wasn't quite there, even though I had believed that he was.) Was Lloyd better than Torriente?
   7. andrew siegel Posted: August 18, 2004 at 02:06 PM (#803870)
I eyeballed WS totals and WARP-1 for Lloyd (completely unscientifically but based on his I9 stats and my experience with reading zillions of season lines, then discounted 10% based on high I9 batting stats).

On WS, I have career totals:
Cobb 722, Lloyd 649, Speaker 630, Collins 574

On WARP-1:
Lloyd 247, Cobb 238, Collins 230, Speaker 220

On the other hand, I also counted number of 40/30/20 WS seasons and # of 15/10/5 WARP1 seasons:

On WS:
Cobb 8/12/20, Speaker 3/11/19, Collins 3/10/16, Lloyd 0/9/20

On WARP-1:
Collins 6/13/18, Cobb 4/12/21, Speaker 1/14/19, Lloyd 1/13/25!

Giving some credit to both metrics and to eyeballing, I have them:

Career: (1) Cobb, (2) Lloyd, (3) Speaker, (4) Collins.

Peak: (1) Cobb, (2) Collins, (3) Speaker, (4) Lloyd.

Putting that together, Cobb is a close but clear first and the other three are just about dead even. And that doesn't even include Smokey Joe Williams. It's going to be very close.
   8. andrew siegel Posted: August 18, 2004 at 02:15 PM (#803880)
FWIW, Wagner 242 2/14/19 on WARP; 655 5/12/15 on WS. Albeit in shorter seasons.
   9. John (You Can Call Me Grandma) Murphy Posted: August 18, 2004 at 02:25 PM (#803903)
Putting that together, Cobb is a close but clear first and the other three are just about dead even.

I don't know about that, Andrew. It's much more difficult for the infield guys to amass as many WS (or WARP) as the outfield guys. I'm looking to have Collins over Cobb and Speaker (Lloyd may be over all of them).
   10. andrew siegel Posted: August 18, 2004 at 02:36 PM (#803924)
John--

Cobb vs. the three others:

(1) vs. Speaker--That one is easy--contemporaries, same league, same position, Cobb better on every metric.

(2) vs. Collins--This one is more debatable. In fact, if you take WARP as gospel, Collins edges out Cobb based on a better peak. But WARP tends to overrate 2B, particularly defensively skilled 2B. And Cobb has fully 25% more WS than Collins with a peak advantage. Call their peaks even if you want and split the difference between WS and WARP on career, Cobb has Collins by 15% on career value. That's too much for him to overcome on position adjustment in my book.

(3) vs. Lloyd--The two are very even on career value--I wouldn't mind if you gave Lloyd a slight edge (though WS has Cobb 10% better). But my guesstimating has Lloyd's peak noticeably worse than the other three. That makes sense in that (with the exception of one season)Wagner's peak was similar to there's and Lloyd wasn't quite the hitter or the fielder that Wagner was. Plus at least some of Lloyd's career totals were put up in seasons where he wouldn't have played in the majors if he was white (no one played in the majors above 41 or so, while many played in the Negro Leagues until their mid-late 40's).

Before this exercise, I had them: (1) Cobb, (2) Speaker, (3) Collins, (4) Williams, (5) Lloyd

Right now, I'm leaning towards: (1) Cobb, (2) Collins, (3) Lloyd, (4) Speaker, (5) Williams.
   11. John (You Can Call Me Grandma) Murphy Posted: August 18, 2004 at 02:46 PM (#803946)
Right now, I'm leaning towards: (1) Cobb, (2) Collins, (3) Lloyd, (4) Speaker, (5) Williams.

That's totally reasonable, Andrew.

The other guy that I'm looking at is Williams. He appears to have been a monster pitcher. Could be anywhere between #1 or #5 for me.
   12. Michael Bass Posted: August 18, 2004 at 02:47 PM (#803951)
My current ranking has them

1) Cobb 2) Williams 3) Lloyd 4) Collins 5) Speaker

Of course, I'm still looking for more on Williams and Lloyd, but that's a basic feel.

Cobb is a clear #1 to me. I feel like Speaker is probably a definate #5. The other 3 are in a jumble, hard to go wrong any way you take them.
   13. The definitely immoral Eric Enders Posted: August 18, 2004 at 02:48 PM (#803954)
Plus at least some of Lloyd's career totals were put up in seasons where he wouldn't have played in the majors if he was white (no one played in the majors above 41 or so, while many played in the Negro Leagues until their mid-late 40's).

I disagree emphatically with this statement.

It's kind of irrelevant to judge the Pop Lloyds of the world based on what run-of-the-mill white players did. Indeed, the very reason we're discussing him in the company of Cobb, Speaker, and Collins is because he wasn't like most players. He was unique. So I don't find it very hard to believe that he would have played in the major leagues in his 40s, and played well. His most comparable white contemporary, Wagner, played until he was 43. Cobb was still a good player at 41, and could have played longer if he'd wanted to. Same with Eddie Collins, who had a .468 OBP in his last season. Those guys chose to retire because, financially, they could. That was an option not available to Lloyd.

C'mon, the man led his league with a .564 batting average at age 44. I think it's safe to assume he was still good enough to play in the majors.
   14. Chris Cobb Posted: August 18, 2004 at 02:56 PM (#803975)
Eric,

What's the evidence for how many games Lloyd was playing per year past the age of 40?

I ask because my sense is that for the very great players, it's the wear-and-tear of playing everyday that pushes them out of the game, not the deterioration of skills. That's why rare pitchers and dhs last longer.

I don't doubt that Lloyd was still a good player, but he was player-manager for teams whose opposition was of widely varying quality. Was Lloyd out there every day, even against the semi-pro teams, or did he save himself for bigger games?
   15. Michael Bass Posted: August 18, 2004 at 02:59 PM (#803981)
I think I agree with Eric on this point.

At the tail end of a Negro League player's career, I agree we should toss most sub-average and maybe even NL-average level play, as they probably wouldn't have been active in the majors at that point.

But if Lloyd was still piling up the numbers at age 41+, at a rate good enough to still be a force in the majors, he certainly will get credit from me for that. Like Eric points out, many of the best players of Lloyd's generation showed an ability to play into their 40s; the fact that they hung it up while still producing shouldn't detract from his record.
   16. Guapo Posted: August 18, 2004 at 03:04 PM (#803995)
When you're trying to rank six no-doubt-about-it, they belong in the inner sanctum, HOMers, you look for every little thing that may help distinguish them.

My intellectually honest vote would probably be a six-way-tie for first. However, we all acknowledge that all of these guys are getting in eventually, and we're just ranking them as a means of starting a healthy debate, right?

Well, this is bound to be controversial, so I might as well throw it out there.

The Constitution says:

"A player’s “personality” is to be considered only to the extent that it affected the outcomes of the player’s games (e.g., via his positive or negative effect on his teammates)."

(you see where I'm going with this, don't you?)

Ordinarily, I don't give this sentence much thought. But with Cobb... is it true that the Tigers tried to trade him for Elmer Flick, because he was such a bastard, and Cleveland turned them down because *they* didn't want to have to deal with him? We all know the stories- Cobb fighting with teammates, teammates refusing to talk to Cobb, nobody showing up at his funeral. Are these all legends brewed up by vindictive sportswriters? If Cobb really was the #### they say he was, did it negatively affect his team's won-loss record in ways that don't show up in the numbers? I don't know the answer to that question. But it's worth asking.

If I was starting a team in real-life, and I had my choice of Cobb, Collins, Lloyd, Speaker, Torriente, and Williams... Cobb would probably be my last choice. If you had your choice of any of six all-time greats, why would would you choose the one who might potentially make your team miserable? On the ballot he won't rank sixth, and I may change my mind before the election. But right now I'm leaning toward Collins and Lloyd in the first two spots.
   17. karlmagnus Posted: August 18, 2004 at 03:21 PM (#804027)
Could someone please post Lloyd's I9 here, or tell me where to get it. Andrew Siegel said "based on his I9", but I haven't seen it.
   18. DavidFoss Posted: August 18, 2004 at 03:25 PM (#804032)
I'm going to go with

1. Cobb
2. Lloyd
3. Williams
4. Collins
5. Speaker

Cobb may not have been the most pleasant of individuals.

Cobb may not have been as good defensively as Speaker...

Cobb may not have been the best infielder in Negro League history...

Cobb may not have been a fireballing Negro League pitcher on the level of Walter Johnson and Pete Alexander...

Cobb may not have been a Gold Glove second baseman on many championship teams...

... but he could have done all those things if he wanted to. :-)
   19. John (You Can Call Me Grandma) Murphy Posted: August 18, 2004 at 03:26 PM (#804033)
Like Eric points out, many of the best players of Lloyd's generation showed an ability to play into their 40s; the fact that they hung it up while still producing shouldn't detract from his record.

Yes and no. If it can be shown that a player was above average (as Lloyd was), I'll give him the credit. However, if he was just hanging on because of the lower competition, then I would say no to any credit (or very little, at the very least).

Where you draw the line is not very easy to figure.
   20. The definitely immoral Eric Enders Posted: August 18, 2004 at 03:27 PM (#804039)
With Cobb, it's documented that the Tigers lost at least one game (the Aloysius Travers fiasco) due to his "personality." So if it truly were a six-way tie, then you'd have an excuse to knock Cobb down to sixth. :)
   21. John (You Can Call Me Grandma) Murphy Posted: August 18, 2004 at 03:29 PM (#804043)
... but he could have done all those things if he wanted to. :-)

With reduced WS (or WARP) numbers if he had played short or second though. His body would have been more battered than it was in CF.
   22. andrew siegel Posted: August 18, 2004 at 03:31 PM (#804049)
I have to disagree with Eric. Normally, when I'm evaluating Negro League candidates I don't get too caught up in there statistics but just try to get a sense of their skill set, consitency, and relative career length and plug them in accordingly. With Lloyd, I have had to mock up some numbers b/c/ my eyeball assessment of his career is dead-even with my assessments of a few white players eligible that same season. When I do statistical comparisons between players, I maake lots of adjustments (for league quality, for season length, for war years, for years trapped in minors by stupid management, for years playing somewhere other than the big leagues by choice, etc.). My goal is to come up with numbers that can be directly compared, in other words to see how guys would have compared to each other if they were in the same league at the same time. If the great white players of a particular era left the game around 40-42 while still capable of playing at the major league level and the great black players of that era stated in the Negro Leagues until 44-48, I'm not just going to assume that that was because the black players were uniformly better or more durable. Instead, I'm going to look for systemic reasons (the white players could afford financially to quit, the Negro Leagues' patchwork schedule left more room for part-time play by still effective elders, the lower replacement level in the Negro Leagues had the same effect, white managers were prejudiced against using older players, white press and fans wouldn't tolerate older superstars playing as shadows of former selves) and correct for those systemic biases in my system.
   23. andrew siegel Posted: August 18, 2004 at 03:32 PM (#804055)
www.i9s.org
   24. The definitely immoral Eric Enders Posted: August 18, 2004 at 03:40 PM (#804062)
What's the evidence for how many games Lloyd was playing per year past the age of 40?

That's a good question, and I don't really have a good answer.

In 1928, the year Lloyd batted .564 at age 44, he had enough playing time to lead the league in both stolen bases (according to Holway) and home runs (according to Riley). He also appeared in the Top 5 in triples once while in his 40s. Those are the only 3 instances I could find where he appears on a leaderboard for a counting stat during his 40s. (Although he was consistently near the top in batting average throughout his early 40s.)

BTW, the only counting stats for which Holway has annual leaderboards are 2B, 3B, HR, and SB. So after the age of 40, either Lloyd wasn't playing quite as much as other players, or he wasn't getting many extra base hits.
   25. karlmagnus Posted: August 18, 2004 at 03:43 PM (#804064)
Thank you. Very helpful on Lloyd, who will slot in just below Collins and above Speaker. They were sort of unhelpful on Williams, though. Why don't they do pitcher W/L, which is the only pitcher stat I understand? Somebody produced a Williams W/L with a 4 in it on the W side, but I've forgotten the L and anyway it's not benchmarked against other NL pitchers such as Foster (who may have been a mistake, looking forward at what's coming.)
   26. The definitely immoral Eric Enders Posted: August 18, 2004 at 03:51 PM (#804081)
Right now my order would be this:
1. Cobb
2. Lloyd
3. Williams
4. Collins
5. Speaker

Although all of those spots are subject to change.
   27. Dag Nabbit is part of the zombie horde Posted: August 18, 2004 at 04:17 PM (#804142)
Ordinarily, I don't give this sentence much thought. But with Cobb... is it true that the Tigers tried to trade him for Elmer Flick, because he was such a ####, and Cleveland turned them down because *they* didn't want to have to deal with him?

Been reading through the Baseball Timeline book & it lists the offer to trade Cobb for Flick as occuring (IIRC) in 1906 (maybe 1907), before Cobb had won his first batting title. It also lists it happening right after Cobb got in a fight with the black groundskeeper & the groundskeeper's wife that turned into a fight with the catcher, Schmidt.


But this is getting off topic. Right now, I have Lloyd tentatively slated for #2, & I don't know enough abot him to know where I'm placing Williams. I ain't really that concered about my placement of them because it's just a matter of time until they all get in.
   28. TomH Posted: August 18, 2004 at 04:26 PM (#804164)
Sanity Check time!

Lloyd and Williams are by consenus two of the best pre-integration black stars ever. But most would put them behind the threesome of Paige, Gibson, and Charleston. Perhaps "most" are incorrect, but perhaps not. Let's say they rank #4 and #5, which assumes (in their favor) no strong arguments for other worthy candidates.

If I make up a typical all-time team of white players prior to 1947, I get names like Cobb Speaker DiMaggio Ruth Ott Wagner Hornsby Collins Lajoie Gehrig Young Johnson Mathewson Alexander; let's say for argument sake that's the top 14.

If we place 5 black stars in among these, that would be 26% (5/19) of the best. Pretty generous, no? So where do the 4th and 5th best black stars slot in? Surely not in the top half of the above group; that would be saying that 50% of the best players of that era were black-- an assessment that is to me very difficult to support.

Ty Cobb was thought by most in 1939 to be the best player ever. And the people then knew what he was like. Tris Speaker was thought by some to be Cobb's equal. Eddie Collins is judged by notable experts to be the best second baseman ever. These guys are not among the lowest of the group named above.

I would strongly support including Pop Lloyd on an all-time team as a backup shortstop. Smokey Joe Williams is possibly one of the best 10 pitchers ever to toe a mound. But in the 1934 competition, I'm sorry, they aren't gonna get a medal from me.
   29. The definitely immoral Eric Enders Posted: August 18, 2004 at 05:01 PM (#804253)
I think quotas are useful in many ways, like making sure no position or type of player is dramatically underrepresented or overrepresented. But I can't see using a hard and fast quota like TomH does in #28.

This isn't some game where we're just dumping the Negro Leaguers into predetermined slots on the scale. When it all comes down to it, we're comparing players to each other. Maybe there are 6 black players among the 20 best pre-1947; maybe there are 2. In any case, the players should be judged on their own individual merits and not simply dropped into some mathematically predetermined racial slot.

Tris Speaker was thought by some to be Cobb's equal. Eddie Collins is judged by notable experts to be the best second baseman ever. These guys are not among the lowest of the group named above.

The people who thought Speaker was Cobb's equal were wrong.

And if Collins and Speaker are, as you say, not among the lowest of the group that you named, which players on your list of 14 would you say are inferior to them? Just curious.
   30. TomH Posted: August 18, 2004 at 05:42 PM (#804315)
Eric, I agree with you...I wasn't arguing for hard quotas. But maybe flexible quotas? And 5 black stars among the best 12 seems a bit outside my personal 'flex' zone. Maybe others disagree.

I'd have DiMaggio, Ott, Lajoie, Matty, and Alexander at the bottom of the Super 14 named above. Which is nuthin to be ashamed of. Collins, Speaker, Young and Hornsby in the next group up.
   31. Daryn Posted: August 18, 2004 at 06:27 PM (#804388)
Tom,

It would surprise me if ONLY 5 of the top 19 players of any given time were black. Anything lower than 50/50 doesn't cause me to do a "sanity check" (not to say that anything above 50/50 would either -- I'm not really into quotas). But maybe that is just me. I'll have Williams and Lloyd 2 and 3, and Torriente will probably be 7, behind Welch.
   32. jonesy Posted: August 18, 2004 at 06:39 PM (#804412)
Eric,

Here are some excerpts from Lanctot; pages 140-142.

"This inneffective....sporadically generating incomplete and ultimately meaningless statistical information...

"In 1939, NNL officials took tentative steps to address the problem.

"While this system was an improvement over recent seasons, it was undermined by the usual bickering among league members.

"In late May, Effa Manley informed the Courier that the standings were wrong this week, blaming the Grays' failure to comply with the new method and questioning whether Posey preferred to 'have the results handled by him in case the race is close and he might be able to do a little juggling."

"Despite their obvious limitations...

"Posey , for example, was unable to issue final averages in 1940, explaining that several clubs had failed to submit all their score sheets.

"Nevertheless, league officials continues to complie and publish occasional statistics based on what little data they had available, often resulting in criticism from skeptical writers.

"...claimed Josh Gibson batted an eye-popping yet unlikely .542."


More to come.
   33. jonesy Posted: August 18, 2004 at 06:53 PM (#804441)
"Despite the hiring of the Howe and Elias bureaus, the two two leagues' statistical woes never entirely disappeared. The absence of official scorers remained the fundamental obstacle to accurate averages, forcing the two agencies to rely on data submitted by occasionally indifferent league teams. Recognizing the uneven compliance throughout the NAL, Tom Baird viewed the use or the Howe agency as a mistake, predicting, 'how incorrect it would be (on) account of (the) way we have of scoring our games.' In the NNL, the Homestead Grays entrusted a player to keep score, yet as Buck Leonard explaianed, 'maybe he didn't know how...
Or in the middle of the game, he'd have to go in...and some other player would have to finish the box score.'

"By 1946, the ongoing problems were apparent to Effa Manley, who informed the Elias Bureau that statistics were 'still not perfect,' although she admitted that 'the mistakes are due to the team owneers and not your office."
   34. The definitely immoral Eric Enders Posted: August 18, 2004 at 07:05 PM (#804464)
OK, it sounds like what I suspected: The cumulative stats published by the league were corrupt and not to be trusted. But historians already knew that; all the numbers we're throwing around have been compiled from individual box scores, so we're safe from the contaminated data. (Although we still long for more of the uncontaminated kind.)
   35. jimd Posted: August 18, 2004 at 07:26 PM (#804505)
Daryn: It would surprise me if ONLY 5 of the top 19 players of any given time were black.

Why?
   36. John (You Can Call Me Grandma) Murphy Posted: August 18, 2004 at 07:59 PM (#804562)
Regarding the percentage of African-American players, I have no idea what the "ideal" number should be. The only thing I care about is the validity of the i9 stats and Chris and whoever else's conversions of those numbers. If that gives us a 75% or even 95% number, I say "damn the torpedoes, full speed ahead!"
   37. KJOK Posted: August 18, 2004 at 11:54 PM (#804985)
Remarkably, Lloyd won the 1928 Eastern Colored League batting title with an average of .564 -- at age 44. That batting average stands as the Negro League single-season record

I posted this before, but this is 100% false.

With 40 of his team's 46 games of box scores now accounted for, here are his 1928 stats:

G-40
AB-154
H-55
D-4
T-1
HR-6
BB-8
HBP-1
Sac-5
SB-4
AVE-.357
OBP-.393
SLG-.513

There's NO way in those missing 6 games that he could have hit well enough to go from. .357 to .564.
   38. KJOK Posted: August 19, 2004 at 12:10 AM (#805033)
1928 Lloyd Fielding:

2B
G-26
Inn-222
PO-74
A-69
E-4
DP-11

1B
G-11
Inn-100
PO-135
A-11
E-0
DP-4

3B
G-3
Inn-25
PO-8
A-4
E-0
DP-0
   39. jimd Posted: August 19, 2004 at 12:28 AM (#805124)
The only thing I care about is the validity of the i9 stats and Chris and whoever else's conversions of those numbers. If that gives us a 75% or even 95% number,

Suppose it says 5%? or 0%?

There are sanity checks for these numbers. A conversion which says that these leagues were a lot stronger than the white "majors" is as demographically dubious as one which says that all the players were below replacement level.
   40. The definitely immoral Eric Enders Posted: August 19, 2004 at 12:52 AM (#805226)
KJOK,
Where did you get those 1928 numbers from? They disagree with Riley, Holway, and basically every other published source. Not that I don't believe you; but if we're going to throw all their research out the window I'd like to know where the new stats are coming from.
   41. John (You Can Call Me Grandma) Murphy Posted: August 19, 2004 at 12:53 AM (#805231)
Suppose it says 5%? or 0%?

That would also be fine with me, Jim. I don't believe in quotas, so I'm not looking to elect a certain number of African-Americans so I can feel good about myself.

A conversion which says that these leagues were a lot stronger than the white "majors" is as demographically dubious as one which says that all the players were below replacement level.

If the conversions are stating that the Negro Leagues as a whole were as good or even better than MLB, then that would make me pause big time.
   42. KJOK Posted: August 19, 2004 at 01:03 AM (#805277)
Where did you get those 1928 numbers from?
As my posts says, they're compilations from the actual newspaper box scores. Boxscore sources include:

Chicago Defender
Chicago Tribune
Cleveland Plain Dealer
Cleveland Gazette
Detroit Free Press
Kansas City Star
Kansas City Times
Kansas City Call
Kansas City American
Memphis Commercial-Appeal
St. Louis Globe Democrat
and others...
   43. The definitely immoral Eric Enders Posted: August 19, 2004 at 01:16 AM (#805337)
I guess what I'm asking is, who compiled them? Was it you? If so, I certainly admire your dedication.
   44. KJOK Posted: August 19, 2004 at 03:34 AM (#805721)
I can't take credit for compiling all these. I do some compiling, but it's using ProQuest or Paper of Record from the comfort of my computer desk. The compiler who has spent many hours looking at sometimes hard to read microfiche may or may not reveal himself here, but he is a Negro Leagues expert who is very meticulous in compiling these records, double checking, making sure league batting totals balance to league pitching totals, etc. I trust these numbers are around 99.99% correct.
   45. jonesy Posted: August 19, 2004 at 11:59 AM (#805953)
Lanctot, page 190:

"Meanwhile, the occasional attempts to mirror the major league all-star game by allowing fans to select players through voting proved largely unsuccessful, as leagues already unable to compile accurate statistics were hardly in a position to tabulate thousands of ballots. Nevertheless, black newspapers dutifully published the supposed results of the voting, although Rollo Wilson noted that 'I know several baseball bugs here and there but I never yet have talked with one who had sent in a listing of his favorite players.' Ed Harris was equally dubious, suggesting in 1941 that 'now everyone who has his ear cocked to the windward knows that Cum Posey sits in his den and makes up these statistics.'

Now I am making a jump here to 1939 (page 84 of Lanctot):

"The improved compliance manifested itself in the adoption of a new plan requiring all results of league gams to be wired to Gottleib or Manley, and all scoresheets to be sent to Posey for statistical compliation. Some owners continued to be lax in reporting statistics and scores, but black newspapers received information more consistently in 1939 than they had in previous years."

Together with KJOK's recent post, I continue to have less than full confidence in the currently accepted numbers. Eric, you're right about that new grant/project to compile Negro League stats. They can't come soon enough.
   46. John (You Can Call Me Grandma) Murphy Posted: August 19, 2004 at 01:50 PM (#806006)
I think we're overreacting to some extent about these Negro League numbers. It wasn't as if there weren't superlatives thrown around to Johnson, Foster, Hill, Williams, Lloyd and the rest. It pays to be cautious about a Jules Thomas or Ben Taylor, but the ones that we have elected already had great reps outside of their stats.
   47. karlmagnus Posted: August 19, 2004 at 02:09 PM (#806035)
So did Hal Chase. If the stats are as shaky as they seem to be, we can only rank NL players within their own group, and attempt to pick a reasonable number -- say 12-15, in line with their share of the overall population. Not a quota, but recognizing that if shaky stats give you twice as many NL HOMers as their share in the population, you're probably analyzing them wrong and rounding up too much.
   48. John (You Can Call Me Grandma) Murphy Posted: August 19, 2004 at 02:22 PM (#806052)
So did Hal Chase.

Hal Chase was a great defensive player, but he was also a crook. That's why his stats don't match his rep. If he had been honest, I'm confident you would not be using him as a counterpoint at this time.

If the stats are as shaky as they seem to be, we can only rank NL players within their own group, and attempt to pick a reasonable number -- say 12-15, in line with their share of the overall population.

What was the percentage of African-American HOF stars from the sixties? There was a ton of them that easily exceeded the percentage of African-Americans in the country at the time. The number of black stars doesn't necessarily have to correlate with the percentage of black people in the country.
   49. karlmagnus Posted: August 19, 2004 at 02:33 PM (#806064)
Yes it does have to correlate, at least approximately, unless there's a very clear socio-economic reason why not. In the case of the Negro leagues, the reasoning's the other way -- they offered much less money and security than the majors, so African-Americans with decent alternative prospects may have spurned them, like Leever in the 1890s.

In the 1960s, the players for which chose their career paths in the 1950s, African American opportunities had suddenly expanded in the highly lucrative world of baseball, whereas they hadn't particularly expanded in society as a whole. Not surprising then that there was a surge of black ballplayers. It didn't last -- the growth in ballplayers today is primarily in Latin American Hispanics.

I am more prepared to believe that enthusiasts for the Negro leagues round the very limited stats in favor of their heroes than that some completely inexplicable historical factor led to a tsunami of African American baseball greatness outside the major league structure. Don't forget, there are no Negro league "experts" attempting to belittle their achievements; the expertise is all in one direction. Cap Anson never joined SABR!
   50. Michael Bass Posted: August 19, 2004 at 02:38 PM (#806074)
That's a pretty flawed argument. To the extent that Negro League ball was less lucrative than major league ball, that is offset by the general opportunities at the time for African-Americans being much less than their white counterparts.

I see no evidence that NL pay as compared to non-baseball opportunities for African Americans was any less lucrative than ML pay as compared to non-baseball opportunities for the non-banned players.
   51. karlmagnus Posted: August 19, 2004 at 02:53 PM (#806094)
No, but for there to be more African American HOMers relative to population, negro league baseball would have to be relatively more attractive than the majors, and I certainly see no evidence of that.

It is a simple statistical theorem; if you are picking 120 balls out of a huge urn, 10% of the balls in which are blue, then there is only a 2.9% chance that more than 18 of them are blue (Excel's BINOMDIST function gives it.)
   52. karlmagnus Posted: August 19, 2004 at 03:13 PM (#806118)
That's because boxing is not something that middle class kids want to get into. As a middle class child, I very much wanted to be an international cricketer, presumably the equivalent of a major league baseball player. This proved to be impossible, of course, owing to my total lack of cricketing talent or athletic ability in general. At no time did I or anyone I knew have an ambition to be a boxer. Kids from white collar or decent blue collar backgrounds freequently grow up wanting to be Nomar or Barry Bonds; they do not grow up wanting to be Mike Tyson.
   53. John (You Can Call Me Grandma) Murphy Posted: August 19, 2004 at 03:20 PM (#806131)
karlmagnus, I had that post on for maybe a minute before I deleted it because I saw a flaw in my argument, but you still jumped on it. :-)
   54. Chris Cobb Posted: August 19, 2004 at 03:22 PM (#806136)
If the stats are as shaky as they seem to be, we can only rank NL players within their own group, and attempt to pick a reasonable number -- say 12-15, in line with their share of the overall population. Not a quota, but recognizing that if shaky stats give you twice as many NL HOMers as their share in the population, you're probably analyzing them wrong and rounding up too much.

This doesn't make sense. If the stats are as shaky as they seem to be, they are just as likely to be misleading in their representation of Negro-League players in comparison to another as they are in their representation of Negro-League players in comparison to major-league players.

Our rankings of the Negro-League players are going to be subject to more uncertainty than our rankings of major-league players, because their statistical foundation isn't firm. So be it.

But if we're going to give the Negro-League players equal treatment, we have to look at their records, with skepticism but without prejudice, and try to decide how they compare to major-league players. Otherwise we'll have a Hall of Merit with two different standards, and that isn't a Hall of Merit.

In five years, when the new, definitive statistical encyclopedia of Negro League baseball comes out, with much more reliable data and thorough studies of competition levels in black baseball, we may look back and see that we were wrong in some cases. In five years, when fielding analysis of 19th-century baseball on the basis of existing statistics has become much more advanced, we may look back at our nineteenth-century picks and see that we were wrong in some cases. That's the hazard we run in making choices based on limited information.

But the HoM will nevertheless have set a standard for historical assessment of merit that can be built upon later. If we don't set that standard, we will have failed to accomplish a basic goal of the project.

Don't forget, there are no Negro league "experts" attempting to belittle their achievements; the expertise is all in one direction.

This leaves out the work of people, discussed above, who are seeking to reconstruct the historical record as accurately as possible. Their expertise is not directed to advocacy.
   55. karlmagnus Posted: August 19, 2004 at 03:24 PM (#806139)
I wish I had the ability to liquidate posts with foolish arguments in them :-))
   56. John (You Can Call Me Grandma) Murphy Posted: August 19, 2004 at 03:33 PM (#806158)
I wish I had the ability to liquidate posts with foolish arguments in them :-))

Hasty, not foolish. My brain usually sees clearly, but not always quickly. :-)
   57. karlmagnus Posted: August 19, 2004 at 03:45 PM (#806181)
Mine sees quickly, but quite often foolishly :-))
   58. jonesy Posted: August 19, 2004 at 03:51 PM (#806197)
I've been looking for this since yesterday.

I have two boxscores from August of 1925. The games were between the Rockland (Mass.) All-Stars, a semi-pro team, and the (1) the Philadelphia Giants and (2) the Philadelphia Red Caps. The Giants and Red Caps appear to be two squads of one team traveling together.

The games was in Rockland, Massachusetts. Former Boston Red Sox pitcher Buck O'Brien -- then 43-years old -- defeated the traveling teams 3-0 and 4-3. The Rockland team was a very strong semi-pro team and had several players with pro experience. The shortstop had played briefly in the International League.

The Giants team/lineup appeared intact -- as listed in (SABR's)Clark and Lester's The Negro Leagues Book -- Burlin White, Dick Seay, Pender Ricks, Leonard Pierce, Babe Lewis, Ralph Jefferson, Joe Lockhart, etc.

The coverage in the local newspaper was complimentary to both teams, giving proper due when earned.

Now, I realize that the 1925 Philadelphia Giants are nowhere near the top of the heap in the overall history of the Negro Leagues, but I feel strongly that this type of material (Negro League versus strong white semi-pro squads), when located, needs to be included in the overall statistical picture (to eradicate small sample size).

Just my opinion, for what its worth.
   59. Daryn Posted: August 19, 2004 at 05:39 PM (#806501)
Isn't possible blacks would be over represented because they are better at baseball than whites. There are only three possibilities -- as a group blacks and whites are equal, whites are better or blacks are better. If blacks are/were better, then regardless of census counts or socioeconomic motivations, they may still rightly be overrepresented in our HoM. I'm not much of an historian, but I don't think the overepresentation of blacks in modern basketball is solely due to socio-economic reasons. They might actually be better than whites.

I know the above borders on rascism and Al Campanis would switch it around to say blacks are inferior, but I have to say I've never been convinced that it is necessarily a fallacy to say certain races are better at certain things than others. Just ask Dusty Baker ;).

All that being said, I trust Chris' method becuase it is based on the sample of players who played in both blackball and integrated ball -- that has to be the best way of comparing the two leagues. That method seems to show blackball was inferior to whiteball, but I simply would not have been surprised if the comparison showed the opposite.
   60. andrew siegel Posted: August 19, 2004 at 05:52 PM (#806547)
Back to the discussion of Lloyd, I was looking at his stats and his defensive testimonials again and it struck me that one possible comp for offensive and defensive talent (though not for distribution of offensive and defensive skills) is ARod. So, imagine that ARod plays 18 more years at an elite level and loses only 2 or 3 of them to injuries or off years.

(1) That's an unbelievable package, among the top 5 ever on career value.
(2) It still doesn't move him to the head of the pack against a Ruth, Wagner, Bonds, Williams, Cobb, etc. Put bluntly: I'd take 18-21 years of Bonds over 25-28 of ARod. Discuss.

I'm still hemming and hawing about these guys, but right now I think James has Lloyd in the right range (20-30). That might only be good enough for 4th on my ballot.
   61. The definitely immoral Eric Enders Posted: August 19, 2004 at 06:13 PM (#806618)
I think 25 years of A-Rod would rank well above 20th or 30th all time.
   62. John (You Can Call Me Grandma) Murphy Posted: August 19, 2004 at 06:14 PM (#806625)
I think 25 years of A-Rod would rank well above 20th or 30th all time.

He would be at the very least top five.
   63. karlmagnus Posted: August 19, 2004 at 06:16 PM (#806633)
A-Rod will not be a major league shortstop in 2020, trust me; his body will have slowed down too much. Nor would Lloyd have been an ML shortstop at 46. The fact that these guys went on 5 years longer than their ML equivalents is another indication that NL competition levels were poor -- it's like Iron Man McGinnity pitching in the minors at 57. We shouldn't give them extra counting stats for a period when they wouldn't have been able to survive on an ML roster.
   64. The definitely immoral Eric Enders Posted: August 19, 2004 at 07:29 PM (#807026)
Nor would Lloyd have been an ML shortstop at 46

He wasn't a Negro League shortstop at 46, either. As stated earlier in the thread, he moved to 2B in 1923 and 1B in 1928.
   65. andrew siegel Posted: August 19, 2004 at 08:30 PM (#807296)
I fear that you guys overrate ARod a bit. He's an A offensive player, not an A+, or an A++ like Barry. He was a very good SS but not a consistent gold glover, let alone an Ozzie. Even if he hadn't gone to the Yankees, he was destined to play a third of his career at less demanding defensive positions due to the strength of his bat, the demands of the position, and his body type. In other words, he's a 32-35 WS player (with a few seasons slightly better than that and a bunch of seasons in the 20's mixed in). I think you need to timeline a lot or be a pure career value voter to put such a guy in the top 5 All-Time even if he plays 25 years.

Maybe ARod is a bad example b/c/ it is so ahistorical, but I see Lloyd as a very similar player: an A offensive player, not an A+; a very good defensive SS who would have won an occassional major league GG, but not a perenial, let alone an Ozzie; and a guy who played a very long time but spent a third of his career at other positions. On peak, he's either the second or third greatest SS of All-Time (he and ARod are close; George Wright, Jennings, Vaughn, and maybe Banks are hot on their tails). In terms of career value, he's second (miles ahead of third) after you adjust for the excess years that I think he played (he's probably first by a small margin if you don't adjust). That's a truly amazing package--you can make a case for putting him in the top 10 All-Time based on it. But I'm a voter who weighs peak and prime heavily and I don't think he was at his best on the Williams/Mantle/Cobb level, let alone the Ruth/Bonds/Wagner level. Put differently, he's the Cy Young of position players.
   66. The definitely immoral Eric Enders Posted: August 19, 2004 at 09:00 PM (#807399)
Lloyd's defensive reputation was that of a consistent Gold Glove winner. Other than that, I agree with most of what's in your post, andrew.
   67. The definitely immoral Eric Enders Posted: August 19, 2004 at 09:17 PM (#807440)
I am more prepared to believe that enthusiasts for the Negro leagues round the very limited stats in favor of their heroes

Falsifying historical records is a very serious charge, karl. In the future might I humbly suggest that we refrain from such suggestions unless we have good reason to believe they are true.
   68. KJOK Posted: August 20, 2004 at 12:36 AM (#807809)
I don't think the problem with Negro League stats generally is someone falsifying them. I think the problem is more as follows:


In the Nov. 3, 1928 Baltimore African American, there is a big writeup on the seasons of the Homestead Grays, which says:

The Grays played 148 games

Dennis Graham led them in batting with an average of .578

John Beckwith hit 53 Home Runs for the season.

Oscar Owens went 40-5


However, the Grays actually only played 21 games against other ECL teams, and with 19 of those games compiled, we find:

Graham hit .397 in those 19 "league" games
Beckwith hit 2 HR's in those 19 games
Owens went 0-2 in those 19 games

So, the problem isn't someone trying to alter the records, the problem is selecting what records to use, and/or determining the level of competetion that those records were compiled against.
   69. KJOK Posted: August 20, 2004 at 12:45 AM (#807833)
Just one slight corretion - I should have said "21 games against other ECL teams" as the Grays were not an official member of the ECL in 1928 which is why they played so many "other" games and so few games vs. ECL teams...
   70. sunnyday2 Posted: August 20, 2004 at 03:50 AM (#808717)
I've always said that it comes down to player A and player B. But OTOH very soft quotas can be helpful as context. But of course it is also true that everybody getting votes here is an outlier, so why would total population be an indicator.

Anyway, my point is that I think the discussion here is great in providing more soft/contextual evidence, as in post #58. Thanks, jonesy.

So while there is good reason for everyone to be uncertain about his ballot, I think just for myself that overall we've given a lot more of the "benefit of the doubt" to recent Negro Leaguers than we ever did for the previous class of qualitative candidates (i.e. pre-1871), though of course I have all of the same Negro Leaguers in my PHoM as are in the HoM with but one exception.

So right now I'm thinking (for 1934, who really cares about 1933, right?):

Cobb
Speaker
Collins
Lloyd
Williams

big gap

Jennings
Torriente
Pike

It's as if we had Aaron, Mays, Robinson, Mantle, ARod and Clemente all at once, and in that company Clemente would have to wait too.
   71. jonesy Posted: August 21, 2004 at 01:48 AM (#810340)
In case anyone is interested I found two more Philadelphia Colored Giants boxscores, both coming in August 1925 versus a Brockton, Massachusetts semipro all-star team.

On August 15, the Brockton squad beat the Giants by a 3-1 score. The Giants used their regular lineup. Bill Rankin was the pitcher. The Brockton team was notable for its right fielder, Haddie Gill, and its pitcher Ollie Hanson, both of whom were cup-of-coffee major leaguers; Hanson with the Cubs in 1921 and Gill with the Reds in 1923. (Gill had his own story in SABR's 1997 edition of The National Pastime.) Will Jackman was coaching 3b for the Giants. Maybe he was incapacitated for he did not play in any of the four games I have located.

The Giants took a 5-3 contest twelve days later. Pitching for Philadelphia was someone named Crudup and Leonard Pierce. The Brockton squad had one future major leaguer in the lineup, Freddy Moncewicz, a Boston College player who was playing in the Cape Cod League that season and who got into a couple of games with the Boston Red Sox in 1928.

So of the four games I located, the Giants (clearly a lesser caliber Negro League team)lost three of four to the semipro squads, albeit each time (O'Brien twice and Hanson once) to pitchers with major league experience.
   72. Michael Bass Posted: August 31, 2004 at 02:08 PM (#828505)
Did Chris ever do a WS analysis of Lloyd? I could have sworn I remember seeing it once, but now I can't find it.
   73. Chris Cobb Posted: August 31, 2004 at 04:20 PM (#828692)
Did Chris ever do a WS analysis of Lloyd? I could have sworn I remember seeing it once, but now I can't find it.

No, I didn't. Finishing my study of Smokey Joe Williams and donig a WS analysis of Jules Thomas are on my list ahead of a Lloyd study, so I doubt I will get to Lloyd prior to the 1934 election. Don't think it will affect the outcome much, anyway, which is why I haven't made it a priority . . .
   74. John (You Can Call Me Grandma) Murphy Posted: September 03, 2004 at 02:50 PM (#835224)
No, I didn't. Finishing my study of Smokey Joe Williams and donig a WS analysis of Jules Thomas are on my list ahead of a Lloyd study, so I doubt I will get to Lloyd prior to the 1934 election. Don't think it will affect the outcome much, anyway, which is why I haven't made it a priority . . .

I would hope that you reconsider, Chris, because your analysis is usually the catalyst for a player going up or down for this project. Since I have Lloyd ahead of the other four, your work could either reinforce or change my opinion before the election next week.

I know Pop is, without a doubt, going to be a member of the HoM very shortly. But if he deserves to go in '34, your work may do the trick. Thanks!
   75. karlmagnus Posted: September 03, 2004 at 03:17 PM (#835255)
John, rushing in where angels fear to tread, I have to think ranking Pop Lloyd ahead of Cobb is a stretch. Even I9 gives him only 3906 hits and a .324BA, compared with Cobb's 4191 and .367. Yes I know he was an SS, but one would have to be a truly amazing SS to make up for that difference, plus whatever fraction by which I9 overstates. I know Ruth said Lloyd was th best he ever saw, but hell would have had to freeze over before Ruth gave that accolade to Cobb, so Lloyd was a neat, New York-friendly way of neither bragging nor dissing his white competitors.

Lloyd was almost certainly a much nicer man, on the other hand Cobb was clearly a greatly superior investor; I regard that comparison as a wash :-))
   76. John (You Can Call Me Grandma) Murphy Posted: September 03, 2004 at 03:35 PM (#835275)
Yes I know he was an SS, but one would have to be a truly amazing SS to make up for that difference, plus whatever fraction by which I9 overstates.

If Lloyd is in Wagner territory, he belongs over Cobb. There have been quite a few outfielders throughout baseball history that you can point to that were in Cobb's ballpark as a player. How many were in Wagner's (or possibly Lloyd's)? The gap is huge there.

I know Ruth said Lloyd was th best he ever saw, but hell would have had to freeze over before Ruth gave that accolade to Cobb, so Lloyd was a neat, New York-friendly way of neither bragging nor dissing his white competitors.

I always take a player's accolades for another player with a grain of salt. I bet someone could point to a few other Ruth quotes where he has a different greatest player of all-time
   77. John (You Can Call Me Grandma) Murphy Posted: September 03, 2004 at 03:36 PM (#835276)
Lloyd was almost certainly a much nicer man, on the other hand Cobb was clearly a greatly superior investor; I regard that comparison as a wash :-))

LOL
   78. yest Posted: September 03, 2004 at 06:43 PM (#835527)
Lloyd will almost definatly be 4th on my ballot to the anecdotal despite what in my eyes is far less then HoM statistics (at least what I found is)for a Negroe Leaguer. I was rather disopointed I didn't find much better statistics. If it was almost any other Negroe Leaguer with the same statistics he wouldn't be anywhere close to my ballot.
   79. John (You Can Call Me Grandma) Murphy Posted: September 04, 2004 at 02:57 PM (#836843)
Yest, I'm shocked that you have Lloyd on your ballot at all! :-D
   80. Chris Cobb Posted: September 04, 2004 at 04:46 PM (#836938)
OK. So I've done a Pop Lloyd study, in the usual fashion.

This post contains all the data from Holway and Riley, which I used to assess the i9s projections for Lloyd. Analysis follows in the next post, estimated win shares for Lloyd in the third.

John Henry Lloyd data

From Holway

1906 .108 for Phi X-Giants (appx. 1/11), listed as playing second base
1907 no batting data, SS for Philadelphia Giants
12-43 (.279) in Cuban play
1908 no seasonal batting data, SS for Phil Giants
10-20 in World Series vs. Leland Giants
1909 .409 (9-22) for Phil Giants, SS
8-33 (.242) vs. major-league competition in Cuba
1910 11-34 (.324) for Chicago Leland Giants, Holway all-star SS
16-40 (.400) vs. Cuban Competition for Leland Giants
11-31 (.356) vs. major-league competition in Cuba
2-13 (.154) in Cuban play
1911 18-50 (.360) for NY Lincoln Giants, Holway all-star SS
0-5 vs. Walter Johnson, Honus Wagner & a major/minor-league squad
1912 .162 for NY Lincoln Giants, SS & manager
2-4 vs. major-league competition
46-135 in Cuban play
1913 40-123 (.325) for NY Lincoln Giants, SS & manager, Holway all-star
4-14 (.286) vs. major-league competition
1914 .323 for Chi Am Giants, 8 2b, 6 3b (tied for lead), Holway all-star SS
0-3 vs. major-league competition
1915 28-79 (.354) for NY Lincoln Stars, .182 for Chi Am Giants, Holway all-star SS
24-61 (.393) in Cuban play
1916 51-110 (.373) for Chi Am Giants (led lg.), Holway all-star SS
3-16 vs. ABCS in playoff
1917 .193 for Chi Am Giants, Holway all-star SS (only one player in west hits .300)
11-29 (.359) in World Series vs. NY Lincoln Stars
1-3 vs. major-league competition
1918 .215 for Brooklyn Royal Giants, SS
1919 .310 for AC Bacharach Giants and Bkn Royal Giants, SS
2-13 vs. major-league competition
1920 11-33 (.333) for Brooklyn Royal Giants, SS & manager, Holway all-star
4-15 in playoff vs. Hilldale
2-13 vs. major-league competition [possible error repeating 1919 stats??]
1921 .331 for Columbus Buckeyes, SS & manager
1922 .310 for AC Bacharach Giants, SS
4-22 (.182) in World Series vs. Chi Am Giants
45-131 (.344) in Cuban play
1923 .349 for Hilldales (2nd in lg.), leads lg with 30 2b, SS & manager
1924 .333 for AC Bacharach Giants, 2B & manager
73-196 (.372) in Cuban play
1925 .333 for AC Bacharach Giants, 2B & manager
46-126 (.371) in Cuban Play
1926 .346 for NY Lincoln Giants, 2B & manager, Holway 2B all star
1927 .350 for NY Lincoln Giants, 2B & manager
36-102 (.353) in Cuban Play
1928 .563** (not reliable avg.) for NY Lincoln Giants, 2B & manager, Holway 1B all star
1929 .362 for NY Lincoln Giants, 1B & manager, Holway 1B all star
1930.434 for NY Lincoln Giants, 1B & manager, Holway 1b all star
12-32 (.375) vs. Homestead Grays in playoff
1931.182 for NY Harlem Bombers, 1B & manager
1932 No data, played for AC Bacharach Giants

Career Totals from Holway
970-2881, .337
41-170, .241 vs. major-league competition
.320 mean avg. 06, 08-31 (25 seasons)
284-807 in Cuban play, .352 (8 seasons)
30-119, .252 in listed ab vs. major-league competition

From Riley

1907 .250 for Phil Giants
1910 .417 for Leland Giants vs. all competition
1911-1913 .475, .376, .363 for NY Lincoln Giants
1915 batted .390 in 10-game series vs. Chi Am Giants
1917 missed some playing time when spiked by Frank Warfield
1921 .336 (vs. league competition, presumably) for Columbus Buckeyes
1922 .387 for NY Bacharachs (according to Riley, they relocated to AC after this season)
1923 .418 for Hilldale (this & 1922 probably vs. all competition)
1924 .444 for Bacharachs (all competition)
1925.330 for Bacharachs
1926-7 .349 & .375 for NY Lincoln Giants
1928 .564 (must be vs. all competition)
1929 .362
1930 .312

Lifetime .368 avg. in black baseball
Lifetime .321 avg. in Cuba, in twelve seasons
   81. Chris Cobb Posted: September 04, 2004 at 04:59 PM (#836947)
Analysis of John Henry Lloyd data

Holway’s numbers, using my standard conversion factor, point to an MLE avg. of .280 to .293 for Lloyd’s full career

Here’s Lloyd’s career i9s line

12043 ab, 3905 hits, 699 2b, 250 3b, 89 hr, 1161 bb, 694 sb, .324 ba, .384 obp, .446 slg

Given the career avg. Holway’s data indicates, my view of the i9s career numbers is that they are significantly overestimated, more than the usual 5%, since they have Lloyd hitting .324 MLE vs. 293 from Holway. I just don’t see any evidence to support such a number: I suspect they took the Riley career average (.368) and treated it as an average from Negro-league competition, when it is clearly an average that includes a significant amount of play vs. “all competition.” Either that, or they are giving Lloyd (and Smokey Joe Williams) some sort of bonus for being legendary figures. . . I’ve reviewed their projections season by season, and I think that they overestimate much more substantially for Lloyd’s post-1920 career than for his career in the teens. Unless there is strong evidence that the lively ball didn’t have nearly as much effect on the Negro Leagues in the 1920s as it did in the majors, I can’t see giving Lloyd batting averages in many of his 1920s seasons that were _higher_ than his documented averages in the negro leagues, which i9s consistently does. Again, I think they are treating averages achieved vs. all competition as negro-league averages.

Based on the season-by-season review of i9s, to provide estimated MLE batting win shares for Lloyd’s career, I will modify the i9s projections as follows:

1) drop 1906 and 1931 from Lloyd’s MLE career
2) prorate the i9s data by .9 1907-1910 and 1917-1920
3) prorate the i9s data by .95 1911-1916 (Lloyd’s peak)
4) prorate the i9s data by .85 1921-30
5) use KJOK’s data for the 1928 season, where i9s, influenced by the legendary .563 batting avg., has Lloyd hitting .394.

This gives Lloyd a basic career batting line of 11668 ab, 3411 hits, .292 ba, which is in line with MLE translations based on Holway’s data, which I believe give us the most reliable level of competition as a base for translation. Lloyd is a no-doubt HoMer, but he’s definitely not Eddie Collins’ equal; my estimates put him more in line with Nap Lajoie. Absolute most conservative interpretation of the data puts him above George Davis and Bill Dahlen, and they were first-ballot electees.

Here's the method by which I arrive at the data you'll see in the next post.

To estimate MLE batting win shares, I prorated the i9s data based on comparisons to the actual data as explained above, then found the closest matches for Lloyd among major-league players that season and prorated their bWS for Lloyd according to plate appearances. I alternated between AL and NL seasons, using NL in even years and AL in odd years.

To estimate MLE fielding win shares, I’ve used Honus Wagner as a model, prorating his rates by 5%. This makes Lloyd an A shortstop, rather than an A+ in win share letter grades, but it’s still giving him credit as a top-notch defensive player. As a second baseman in his later years, I’ve modeled him on Eddie Collins late in his career. For first base, I’ve made him a B-level defender (he was 44 when he started splitting time between 1st & 2nd in 1928!)
   82. Chris Cobb Posted: September 04, 2004 at 05:05 PM (#836951)
John Henry Lloyd estimated MLE career win shares

Season bws + fws = total
1907 7.1 + 4.6 = 11.7
1908 17.4 + 6.3 = 23.7
1909 20.5 + 8.0 = 28.5
1910 18.6 + 9.0 = 27.6
1911 23.6 + 9.4 = 33.0
1912 15.6 + 8.9 = 24.5
1913 24.0 + 8.3 = 32.3
1914 23.5 + 9.8 = 33.3
1915 15.6 + 7.4 = 23.0
1916 27.9 + 9.3 = 37.2
1917 12.3 + 6.8 = 19.1
1918 20.7 + 8.2 = 28.9
1919 6.4 + 4.3 = 10.7
1920 22.6 + 7.6 = 30.2
1921 12.5 + 7.7 = 20.2
1922 10.8 + 5.8 = 16.6
1923 14.0 + 5.6 = 19.6
1924 10.5 + 3.9 = 14.4
1925 7.6 + 2.7 = 10.3
---- highly conservative view ends career here
1926 8.0 + 2.5 = 10.5
1927 9.6 + 2.7 = 12.3
1928 10.5 + 2.4 = 12.9
1929 3.0 + 0.9 = 3.9
1930 4.8 + 1.1 = 5.9

Through 1920, 363.7 ws
Through 1925 444.8 ws
Through 1930 490.3 ws

347.1 bws
143.2 fws

All totals are scaled to 154-game seasons

A few notes on Lloyd's fielding win shares:

His fielding win shares at SS total 127, which would be 3rd all time behind Ozzie Smith and Bill Dahlen.

His 143.2 career fielding win shares would be, I think, second all time. Dahlen edges him with 143.3, and both are ahead of Rabbit Maranville, (142.9), Honus Wagner (141.8), Ozzie Smith (139.8). and Cal Ripken (136.4), who are the only players I have been able to find with more than 130 fielding WS.

Based on my analysis, I will have Lloyd settled quite firmly in the 4th slot on my ballot.
   83. John (You Can Call Me Grandma) Murphy Posted: September 04, 2004 at 05:34 PM (#836978)
Thanks, Chris!

Well, that changes thing now. After your analysis, I'll have to move Lloyd into the fourth slot also.

That means my top five will now be Collins, Cobb, Williams, Lloyd and Speaker.

Lloyd was a great, great player, but he still wasn't Wagner.
   84. Michael Bass Posted: September 04, 2004 at 06:37 PM (#837126)
Thanks so much, Chris! I've got him in 5th place on my ballot, and this kind of confirms my view. As compared the the titans ahead of him, just doesn't quite have the same sort of peak. Nothing to be ashamed of, just a notch below them.
   85. yest Posted: September 05, 2004 at 02:14 AM (#837726)
Yest, I'm shocked that you have Lloyd on your ballot at all! :-D
why because he a Negroe Leaguer, a shortstop or both?
   86. John (You Can Call Me Grandma) Murphy Posted: September 05, 2004 at 02:24 AM (#837730)
why because he a Negroe Leaguer, a shortstop or both?

Definitely because he's a shortstop. Sorry for the confusion.

BTW, I made those additions of yours in the Plaque Room.
   87. David Concepcion de la Desviacion Estandar (Dan R) Posted: November 20, 2007 at 03:31 PM (#2621107)
Are there any seasonal MLE's for Lloyd? AB, H, TB, BB, etc?
   88. Brent Posted: July 11, 2008 at 04:01 AM (#2852607)
After taking another look at this thread, I realized that a lot of new data have become available since we first looked at Lloyd four years ago. I’ve decided to see what these data have to say about his MLEs.

We now have three major new data sources that give us contextual information, such as league averages: (1) the early results of the Hall of Fame study covering post-1920 seasons; Lloyd’s record appears in Hogan’s Shades of Glory (with the league averages available in an HoF report that provides statistics for the candidates on the 2006 special NeLg ballot), (2) Cuban League statistics from Figueredo’s Cuban Baseball, and (3) Gary Ashwill's compilations of Negro League and Cuban data that are posted on agatetype.typepad.com. (In Lloyd’s case, the Cuban data are especially important—his prime years occurred before the organization of the formal Negro leagues and he frequently played winter ball in Cuba.)

Rather than doing season-by-season MLEs, I’ve split Lloyd’s career into four periods and calculated average rates for each period. I had several reasons for taking this approach: (a) we don’t have reliable data for all of his seasons, especially for some seasons before 1920; (b) the season-by-season approach would require a lot more work, especially if it uses regression; (c) when season-by-season MLEs are based on small samples using regression, I’m not convinced that they provide meaningful information on the player’s peak performance anyway—the pervasive lack of peaks for NeLg MLEs has been frequently noted in HoM discussions.

The approach I’m taking in deriving MLEs is a simplification of Chris Cobb’s method (and is the same method that I previously used for Carlos Morán’s MLEs). Here’s a summary of the method: (1) I restrict my analysis to data for which league averages are available or can be calculated. (2) I build my estimates from three rates: the walk rate (BB+AB)/AB, the batting average, and isolated power. (3) I adjust these rates for quality of play, multiplying BA by 0.9 and ISO and walk rates by .81. (No quality-of-play adjustment is made for games played against major league teams.) (4) I adjust the BA and ISO statistics to a National League context by multiplying by the ratio of the two league averages for the period. (However, because the HoF study doesn’t report league averages for walk rates or OBP, I decided not to make any contextual adjustments to walk rates. Gary’s data for 1921 and ’22 show NeLg walk rates similar to those in the NL.) (5) I report Lloyd’s MLEs as average rates for four periods: 1907–18 (ages 23 to 34), 1919–23 (ages 35 to 39), 1924–28 (ages 40 to 44), and 1929–30 (ages 45 to 46), and also as career rates.

One caveat is that I haven’t made any park adjustments. Lloyd played for some teams in pitcher’s parks (e.g., Chicago American Giants) and others in hitter’s parks (e.g., Lincoln Giants), but because he moved around quite a bit, I don’t think the absence of park adjustments is likely to seriously distort his career averages. If someone with more expertise on NeLg park factors wants to add these adjustments, however, please go ahead.

1907–18 (ages 23 to 34)

This is Lloyd’s prime, but it also is the period for which we have the least data. I use the following sources: (a) Gary’s compilation of NeLg teams for the 1916 season, when Lloyd was playing for the Chicago American Giants. Lloyd hit .324/.385/.400 in 210 AB in a context (excluding pitchers) of .258/.331/.336. (b) Data from Figueredo for five Cuban League seasons between 1908–09 to 1915–16, where Lloyd hit .330/—/.400 in 412 AB in a context (x pitchers) of .257/—/.312. (c) Although Figueredo doesn’t report walks, for one of Lloyd’s CuLg seasons (1908–09) we have walk data compiled by Gary. With 96 AB, Lloyd drew 5 walks. (d) Data from Gary on three series played in Cuba by Lloyd’s NeLg teams—the 1907 Philadelphia Giants, the 1910 Leland Giants, and the 1912 Lincoln Giants—against CuLg teams. Lloyd hit .305/.366/.356 in 174 AB in a context (x pitchers) of .233/.312/.265. (e) Data from Gary on Lloyd’s performance for CuLg teams in four series played against major league teams in 1909 and 1910. Lloyd hit .333/.378/.391 in 69 AB in a context of .231/.296/.280.

Converted to MLEs, these data suggest that in the NL Lloyd would have hit:

BA/OBP/SLG––LgBA/LgOBP/LgSLG (x pit)––BA+/OBP+/SLG+/OPS+
.306/.352/.399–.260/.326/.345––––––––––118/108/116/124

These MLEs show us that Lloyd’s strength was hitting for average. He also had mid-range power (doubles and triples), but was below average in drawing walks.

For comparison, here are rates for the 12-year primes of some of the better hitting HoM shortstops (using ages 21 to 32 for Trammell, ages 22 to 33 for Davis, Ripken and Yount, and ages 23 to 34 for Banks and Cronin):

Davis 113/111/119/130
Yount 115/110/119/129
Banks 105/102/127/128
Cronin 106/109/113/122
Ripken 107/107/114/121
Trammell 110/109/107/116

If we take Wagner as the first tier in shortstop hitting and Vaughan as the second, we see that Lloyd, during his prime, ranks solidly within the third tier. (Boudreau also belongs in this group, but needs a discount for wartime competition levels.)

I also note that of these players, Yount appears to be the most similar to Lloyd as a hitter. (Of course, during these seasons, Davis, Yount, and Banks all spent considerable time at positions other than shortstop.)

1919–23 (ages 35 to 39)

During this period, Lloyd continued to play shortstop. For 1919 I don’t have any data that meet my standards. For 1920 and 1923, I use NeLg data from Hogan. Lloyd hit .340/.358/.452 in 188 AB in a context (including pitchers) of .271/—/.366. For 1921 and 1922 I use Gary’s compilations, which provide a slightly larger sample and more contextual information. For these seasons, he hit .337/.375/.433 in 501 AB in a context of .272/.332/.372. Combining these data, I obtain the following MLEs:

BA/OBP/SLG––LgBA/LgOBP/LgSLG (x pit)––BA+/OBP+/SLG+/OPS+
.313/.343/.391–.287/.342/.392––––––––––109/100/100/100

Although his batting average was up due to the lively ball, it appears that Lloyd’s hitting overall was beginning to slip as he aged.

1924–28 (ages 40 to 44)

In 1924, Lloyd joined the Bacharach Giants as a playing manager. The 25-year old Dick Lundy was already well established at shortstop, so Lloyd moved himself to second base. About the same time, it appears that Lloyd probably also made some adjustments in his batting, becoming more patient and selective as his walk rates and overall batting performance picked up. Using data from Hogan, his rates over these seasons were .355/.417/.471 in 788 AB in a context of .274/—/.380. These convert to the following MLEs:

BA/OBP/SLG––LgBA/LgOBP/LgSLG (x pit)––BA+/OBP+/SLG+/OPS+
.331/.386/.429–.292/.351/.409––––––––––113/110/105/115

Regarding Lloyd’s fielding, I’ll mention that although he was playing second base in the States, he continued to play shortstop in the Cuban League through the winter of 1925–26, suggesting that absent Lundy as a teammate he would have been capable of playing shortstop for at least a couple more seasons. (Having sufficient data from Hogan, I didn’t use Lloyd’s 1920s CuLg data in calculating his MLEs. However, I note that they are broadly consistent with what we have here.)

1929–30 (ages 45 to 46)

In 1929, Lloyd moved full time to first base. Using data from Hogan, during 1929–30 Lloyd hit .373/.423/.488 in 287 AB in a context of .278/—/.388. The MLEs are:

BA/OBP/SLG––LgBA/LgOBP/LgSLG (x pit)––BA+/OBP+/SLG+/OPS+
.361/.405/.478–.308/.369/.454––––––––––117/110/105/115

The question about these seasons is whether Lloyd still would have been able to play in the major leagues as a 46-year old first baseman. Assuming that he was at least average defensively, the answer is yes—he was better than a replacement level major league first baseman. I agree with Chris’s decision to give him MLE credit through age 46.

Career totals

Chris assigned Lloyd 11,668 career AB over a 24-year career, which seems about right to me. I’ve allocated them as follows: 6,274 AB (523/yr) for 1907–18, 2,571 (514/yr) for 1919–23, 2,099 (420/yr) for 1924–28, and 724 (362/yr) for 1929–30. The resulting career MLE statistics are

BA/OBP/SLG––LgBA/LgOBP/LgSLG (x pit)––BA+/OBP+/SLG+/OPS+
.315/.359/.407–.275/.337/.373––––––––––115/107/109/116

These rates imply 3,679 career hits (which would have been fourth all-time) and 4,752 total bases.

In summary, based on career value, Lloyd clearly ranks as the number two shortstop behind Wagner. For prime value, it’s less clear. He could rank anywhere from second to about eighth depending on where we think his defense fits along the scale from excellent to merely very, very good. I’d probably rank him fourth in prime value, behind Ripken and Vaughan. For peak value, unfortunately, I can’t provide a ranking. However, I can understand a peak/prime-oriented voter ranking him behind Vaughan and/or Ripken.
   89. mulder & scully Posted: July 11, 2008 at 04:35 AM (#2852623)
Brent, that is great stuff. Fantastic.
   90. DL from MN Posted: July 11, 2008 at 01:51 PM (#2852741)
I have this feeling we may find some extra walks for him later like we did with Dick Lundy. It is unusual for above league average hitters to be below league average drawing walks.
   91. Brent Posted: July 12, 2008 at 03:09 AM (#2853741)
DL,

Thank you for your comment. However, I have to disagree with your premise that it's unusual for hitters with BA and SLG above league average to be below league average in drawing walks. Although there's probably a tendency (or correlation) for better hitters to draw more walks, throughout all eras there have been good hitters with high averages who did not walk much. One of my revelations when I first read sabermetrics in the 1980s Bill James Baseball Abstracts and in Palmer and Thorn was that players like Al Oliver and Steve Garvey weren't nearly as good as we all thought they were because, despite good batting averages, they didn't get on base much.

For example, I looked at the 27 players with 3,000 hits, and 7 of these players were below league average in walks (Clemente, Lajoie, Brock, Gwynn, Yount, Ripken, and Molitor). My MLEs show Lloyd drawing walks at about 75 percent of the league average rate; that's about the same rate as Lajoie and better than Clemente (70 percent)... and much better than Oliver (65 percent) or Garvey (55 percent). And some other really good hitters, like Aaron and Wagner, were only a little bit above league average in walks. The Hall of Fame has lots of players who were below average in walks. The HoM has fewer of them, since we actually pay attention to walk rates, but you can see that there are a number of HoM-quality hitters who were below average in walks.

Will we find extra walks for Lloyd like we did for Lundy? With Lundy, we originally had data for only two seasons, fewer than 350 AB. When the HoF data came out, we suddenly had much more data-- 16 seasons with nearly 2,300 AB.

For Lloyd, since we now have the results of the HoF study, we won't see that kind of dramatic increase in data for the 1920-30 period. Indeed, we probably already have most of the NeLg data that we will ever get for that period. I'm sure that future researchers will continue to locate some hard-to-find box scores, so that a few new games will continue to be added to the dataset. But the overall picture for these years isn't going to change. For 1907-19, on the other hand, there is still a lot of work to be done in finding and tabulating data. However, given that Lloyd's walk rates were relatively low in both the data that Gary's already collected for 1907-16 and the 1920-23 data from the Hall of Fame study, I'd be a bit surprised if new data were to dramatically change his walk profile for this period.

On the other hand, there's one facet of the MLE calculations where new information could have a fairly big effect on his walk rates. I'm referring to the quality of play adjustment that I made in converting his NeLg walk rates to major league equivalents. I multiplied by .81 (that is, reduced his NeLg walk rates by 19 percent), even though I really don't have any solid evidence on which to base that number. I used it because (a) it's very close to what Bill James used in his MLE calculations for Triple-AAA minor leaguers (note that the NeLg adjustment factors we've been using are quite similar to James's Triple-AAA adjustment factors); and (b) it's the same adjustment factor that I had used earlier in calculating MLEs for Carlos Morán, and in that case it seemed necessary to reduce his walk rates by at least that much to get plausible results.

I'm not sure how one could go about refining the estimate of that parameter. I suppose we could look at changes in walk rates as players moved between the Negro Leagues and major and minor league baseball. But I did want to point out that this adjustment has an important impact on the results and it's an area where there's a fair bit of uncertainty.
   92. Chris Cobb Posted: July 12, 2008 at 04:18 PM (#2853984)
Brent, this is great work!

Where did you get the HoF project stats for Lloyd? The last I knew, they had only released stats for the candidates for the big "sweep-in-all-the-Negro-Leaguers-we-missed" election.
   93. Gary A Posted: July 12, 2008 at 05:17 PM (#2854024)
Lawrence Hogan printed the stats for those Negro Leaguers already in the HOF (before the special election) in his *Shades of Glory* book.
   94. Paul Wendt Posted: July 12, 2008 at 07:41 PM (#2854141)
For 1907-19, on the other hand, there is still a lot of work to be done in finding and tabulating data. However, given that Lloyd's walk rates were relatively low in both the data that Gary's already collected for 1907-16 and the 1920-23 data from the Hall of Fame study, I'd be a bit surprised if new data were to dramatically change his walk profile for this period.

Unfortunately there may be plenty of games whose hard-to-find coverage does not include bases on balls for batters, even some box scores without bases on balls for pitchers.
There was no paper of record (SLife, TSN, and I haven't noted how commonly their boxes list batter walks anyway). Somewhere I read that Chicago Defender was the first black daily in 1905. How soon were there several of them? And how soon did they pick up baseball coverage in what detail?

There is some coverage of the crack colored clubs in Chicago Tribune at the turn of the century but it is extremely limited. More important than missing games (the majority), there is a standard box score for high schoool, college, bankers league, merchants league, Western League, and major league games: runs, hits, PO-A-E. The only(?) exceptions are the local major league teams. My foreboding is that even local newspapers who report the Union Giants game with some box score and prose story every time they come to town (our hope to find) adopt that standard, without even batter "at bats".
In the 1890s with only one major league team the Trib provided 11-column box scores for (all?) its games. So that model was out there, too. It will be great to find some local newspapers who also provided extra-informative box scores for local games. Keep our fingers crossed.

(I have looked up all the game reports I could find in the Milwaukee Daily News 1899-1901 iirc. MDN provided 7-column box scores for the Brewers and thus for a few games against crack colored teams. So the double standard was not unique to the Trib. My hope is that some smaller cities pridefully went the whole hog when the local team played against the next city or against a top semipro team from Chicago or against one of the colored teams. That is my hope prior to the black daily newspapers.)

Phil Dixon has completed a work on the 1905 Philadelphia Giants. If my source is correct, there were no daily black newspapers for him. That is the year before Lloyd arrived and two years before he succeeded Johnson as the regular shortstop, selected because it was the team's greatest season. Does anyone know what detail he was able to complete?
   95. Paul Wendt Posted: July 12, 2008 at 07:50 PM (#2854151)
>>There is some coverage of the crack colored clubs in Chicago Tribune at the turn of the century but it is extremely limited. More important than missing games (the majority), there is a standard box score for high schoool, college, bankers league, merchants league, Western League, and major league games: runs, hits, PO-A-E. The only(?) exceptions are the local major league teams. My foreboding is that even local newspapers who report the Union Giants game with some box score and prose story every time they come to town (our hope to find) adopt that standard, without even batter "at bats".<<

Missing games are more important than missing atbats, for anyone relying on one source. I mean the standard box score points to something more important, if I forebode correctly. ( iifc ) Even after doing a lot more work in the next decade and recovering most of the coverage from numerous (not most) small-city and town newspapers, we may not have much detail.

One advantage of standardization is uniformity ;-)
Much was lost with the general adoption of AB-R-H-RBI but several decades, I believe, anyone providing box scores at all felt obliged to provide that much --plus surname, fielding position, and in effect batting position-- for everyone who played.
   96. DL from MN Posted: July 14, 2008 at 01:36 PM (#2855717)
I think the rise in walk rate as the years get later (and presumably better documented) is why I don't necessarily believe the earliest walk rate. I agree, great work.
   97. Gary A Posted: July 15, 2008 at 09:14 PM (#2857430)
Somewhere I read that Chicago Defender was the first black daily in 1905. How soon were there several of them? And how soon did they pick up baseball coverage in what detail?

The Defender was a weekly until sometime in the 1940s or 1950s. The first black daily was the Atlanta Daily World in 1934.

As far as detailed baseball coverage in the black weeklies: I haven’t gone as far back as 1900, but by 1909 the Indianapolis Freeman was very good (though mostly limited to the Midwest and South).

I wouldn’t generalize from the Chicago Tribune, which was generally irritatingly poor on box scores, to anything else, circa 1900 or at any other time. Again, my experience is mostly later, but small towns or cities varied greatly—it depended on the newspaper. The visit of a major black team or teams was a big, big event, and could elicit quite detailed coverage (see, for example, this account of a Cuban Stars / Kokomo Red Sox game in 1916, from the Kokomo Daily Tribune).

I think the rise in walk rate as the years get later (and presumably better documented) is why I don't necessarily believe the earliest walk rate.

I assume you’re talking about Lloyd’s walk rates as used in Brent’s work above? The four eras Brent marked out are equally reliable statistically; in fact, the earliest (1907-1918) includes the best and most complete data available, Cuban box scores. Also there does not seem to be a major rise through his career, though you *would* expect an older player to become more selective. Just looking at OBP-BA, in Brent’s translation:

1907-1918: .046
1919-1923: .030
1924-1928: .055
1929-1930: .050

1921 and 1922 are certainly no less well-documented than, say, 1927 and 1928.
   98. Gary A Posted: July 16, 2008 at 01:37 AM (#2858242)
For anyone who's interested, here are what I've got for Lloyd's walks and at bats:

years: at bats, walks, W/(AB+W)
1907-16: 549, 42, .071
1920-23: 689, 33, .046
1924-28: 842, 85, .092
1929-30: 287, 25, .080

Totals: 2,367 at bats, 185 walks, .073.

To break down the 1907-16 period:
Cuba 339, 23, .064
1916 210, 19, .083

And here's a breakdown for 1920-23, which has the lowest walk rate in Brent's study. 1920 & 1923 are from the HOF project, 1921 & 1922 are from my compilations:
1920 HOF 44, 2, .044
1921 ---- 333, 23, .065
1922 ---- 168, 5, .029
1923 HOF 144, 3, .020
   99. Brent Posted: July 17, 2008 at 01:36 AM (#2861862)
Another point regarding Gary's data on walks is that he made sure that the batter walks for each game are balanced with the pitcher walks. This is an important step in ensuring accuracy, since it ensures that there is no systematic understating of walks. For example, if there were an error causing one batter to have too few walks, there would have to be an offsetting error in the opposite direction causing another batter to have too many walks to keep the data balanced. Balancing the pitcher data with the batter data helps avoid these kinds of errors and ensures that the data don't have systematic biases.

I believe that the HoF study is also supposed to be using this type of balancing, though as far as I know they haven't published much on their methodology yet.
   100. Brent Posted: August 20, 2008 at 04:52 AM (#2909749)
After a discussion with Chris Cobb on the Major League Equivalancies thread, I've decided to switch the conversion factor for walks from .81 to .95, consistent with Chris's recommendation. This new conversion factor leads to some modest changes to the Lloyd MLEs I posted here:

Career
BA/OBP/SLG––BA+/OBP+/SLG+/OPS+
.315/.364/.407––115/108/109/117

1907-18
BA/OBP/SLG––BA+/OBP+/SLG+/OPS+
.305/.355/.399––118/109/116/125

1919-23
BA/OBP/SLG––BA+/OBP+/SLG+/OPS+
.313/.347/.391––109/101/100/101

1924-28
BA/OBP/SLG––BA+/OBP+/SLG+/OPS+
.331/.394/.429––113/112/105/117

1929-30
BA/OBP/SLG––BA+/OBP+/SLG+/OPS+
.361/.412/.478–.308/.369/.454––117/112/105/117
Page 1 of 2 pages  1 2 > 

You must be Registered and Logged In to post comments.

 

 

<< Back to main

BBTF Partner

Support BBTF

donate

Thanks to
Guts
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Syndicate

Page rendered in 0.7417 seconds
49 querie(s) executed