He wasn't really Barry Allen, but close enough for a mere mortal. :-)

Satchel Paige often regaled audiences with the story that when he and Bell roomed together, Bell was so fast that he could turn out the light and be in bed before the room got dark. In truth, he did it one night, but only because there was a short in the wires.

Even though I have read that account numerous times, Craig, it still always manages to bring a smile to my face.

I love that story too. When we were young my brother tried to do that for hours because he thought he could be as fast as Coll Papa.

Getting tom ore serious matters, I like Cool Papa as a fan, but is he better than Lou Brock?

Getting tom ore serious matters, I like Cool Papa as a fan, but is he better than Lou Brock?

That's also my fear, Mark. I will say that I won't place him on my ballot unless I can see an MLE for him. I'd rather be safe than sorry with him.

1928 Cool Papa Bell
NNL champion St. Louis Stars

Batting
*-led league
G-83 (team 83)
AB-353*
H-113 (2nd)
D-21
T-7
HR-4
R-84 (2nd)
W-31
HP-1
SH-8
SB-16 (4th)
AVE-.320 (NNL .278)
OBA-.377 (NNL .333)
SLG-.453 (NNL .384)

St. Louis raw park factor for runs (home/away ratio) was 125 that year. The team as a whole hit .313/.364/.502, scoring 6.86 runs/game (NNL 5.03).

The St. Louis starting lineup:

1. Bell cf .........320/377/453
2. B. Russell rf ...303/372/452
3. Wells ss ........325/425/699
4. Suttles 1b ......361/407/687
5. Redus lf ........330/390/612
6. Creacy 3b .......327/378/508
7a. H. Williams c...299/309/407
7b. Palm c..........313/374/580
8. J.H. Russell 2b..276/312/355

Bell's 16 steals led the team (Wells had 14); with only 49 steals, the Stars were way behind the league leaders (KC, with 123 in fewer games):

Kansas City 123 (74 games)
Chicago 87 (83)
Birmingham 61 (82)
St. Louis 49 (83)
Detroit 48 (80)
Cleveland 47 (64)
Memphis 44 (80)
Cuban Stars 17 (47)

1928 Cool Papa Bell

Fielding-cf
*-led league
G-83*
DI-739.3*
PO-210*
A-16*
E-9*
DP-2
RF-2.75 (NNL cf 2.48)
FPCT-.962 (NNL cf .963)

Center fielders accounted for 42.7% of OF putouts league-wide; St. Louis center fielders (all Bell--he played every inning) accounted for 48.4% of the team's OF putouts, easily the highest figure in the league.

Sorry--Wells hit .365 in 1928, not .325.

Also, Bell was 5th in the league in stolen bases, not 4th. The leaders:

28 Eddie Dwight, KC
20 Newt Joseph, KC
19 Jelly Gardner, Chi
18 Newt Allen, KC
16 Cool Papa Bell, StL

Cool Papa Bell Data

Teams

22-31 St. Louis Stars, 32 Det/KC/Homestead, 33-38 Pgh Crawfords, 37 Santa Domingo, 38-41 Mexico, 42 Chi Am Giants, 43-46 Homestead Grays

Seasonal Data from Holway

1922 No data
1923 3-1 as pitcher for Stl Stars
1924 .316 for Stl; 9 sb (4th); cf
1925 .348 for Stl; 29 2b (1st), 23 sb (1st); cf; all-star
10-33 in playoff vs. KC
1926 .355 for Stl; 17 hr (5th), 23 hr/550 (5th), 24 2b (4th), 23 sb (1st); cf, all-star
1927 .318 for Stl; 18 2b (4th), 13 sb (2nd); cf
1928 .332 for Stl; 6 3b (5th); cf
11-27 in playoff vs. Chi Am Giants
51-157 in Cuban Play
1929 .309 for Stl; 25 2b (2nd), 28 sb (2nd); cf, all-star
3-17 in World Series vs. Homestead
6-22 vs. major-league competition
1930 .362 for Stl; 17 2b (5th), 15 sb (4th); lf, all-star
1931 .330 for Stl; cf, all-star
1-2 vs. major-league competition
1932 .318 for Det/Pgh,/Hom, . ba 4th; cf, ut, all-star
396 for KC; 5 3b (3rd), 4 sb (2nd); cf, all-star (all-star in both East-West and South in 32)
0-3 vs. major-league competition
1933 .339 for Pgh; 15 2b (4th), 9 3b (3rd), 7 sb (3rd); cf
2-6 in playoff vs. Nashville
1934 .364 for Pgh; 5 in ba, 7 sb (1st); cf
1935 .320 for Pgh; 10 2b (4th), 8 3b (2nd), 9 sb (1st); lf
3-24 in playoff vs. NY Cubans
5-16 vs. major-league competition
1936 .301 for Pgh; cf
8-19 vs. major-league competition
3-13 vs. major-league competition
1937 .120 for Pgh; ut (spent most of season in Santa Domingo)
21-66 in Santa Domingo
1938 no data for NeL
.356 in Mexican League – In Mexico through 1941
1939 In Mexico
1940 In Mexico, 167-382, .437 (1st), 12 hr (1st), 17 hr/550 (4th)
41-168 in Cuban Play
1941 In Mexico, 132-421, .314, 12 hr (4th), 15 hr/550 (5th)
1942 .373 for Chi Am Giants; cf
1943 .297 for Homestead; 3 sb (4th); lf
8-26 in World Series vs. Birmingham
1944 .274 for Homestead; 9 2b (3rd), 4 3b (4th); lf
6-23 in World Series vs. Birmingham
1945 .253 for Homestead; 5 2b (4th); lf
4-13 in World Series vs. Buckeyes
1946 .447 for Homestead; lf, all-star

Career, according to Holway
1561-4754, .328
73 hr., 8/550 ab
52-152 vs. major-league competition
Mean avg. 320 for 19 seasons with statistics

Fielding, Career, according to Holway
CF 1923, 1925-29, 1931-34, 36, 42
LF 1930, 1935, 1943-46
P 1924
No fielding info for 1937-41

Career data from MacMillan 10th edition
943 g, 3707 ab, 1248 hits, 194 2b, 64 3b, 63 hr, 143 sb, .337 ba, .475 slg

he looks to me like a poor man's ty cobb

Yet again we must wait for Chris Cobb's wizardry, but this guy looks an anti-Beckwith, one whom history has overrated. Nothing like Ty, more like a poor man's Lou Brock.

Yet again we must wait for Chris Cobb's wizardry, but this guy looks an anti-Beckwith, one whom history has overrated. Nothing like Ty, more like a poor man's Lou Brock.

So far, I would agree with Karl. I haven't run the numbers yet, but a thumb-nail conversion suggests that Bell's MLE batting numbers would have been quite similar to Brock's, in a higher offense era.

More fielding value, though. Looks like in his prime he was a top-notch centerfielder.

On the subject of conversions for Bell, what would y'all like to see with respect to his Santa Domingo and Mexican League seasons?

I have numbers for some of these seasons, but I don't know how they ought to be converted.

Santa Domingo will probably be impossible to calibrate, but there's probably better data for Mexico. Given the number of NeL stars who played there from 1938 on, we should probably start discussing how to evaluate play in that league.

...a thumb-nail conversion suggests that Bell's MLE batting numbers would have been quite similar to Brock's, in a higher offense era.

More fielding value, though. Looks like in his prime he was a top-notch centerfielder.

It looks like the more interesting question is how he compares with HoMer Max Carey.

Yet again we must wait for Chris Cobb's wizardry, but this guy looks an anti-Beckwith, one whom history has overrated. Nothing like Ty, more like a poor man's Lou Brock.

Lou Brock in a high offensive era would have had been on base more so his baserunning ability would have been used more.

besides I think Bell was much better then Brock regardless of their eras

1923 Cool Papa Bell (rookie season, as a pitcher)
NNL St. Louis Stars
(from Patrick Rock's NNL yearbook)

Batting:
G-49
AB-108
H-26
D-5
T-1
HR-2
R-24
RBI-17
W-7
SB-1
AVE-.241
OBA-.284
SLG-.361

Pitching:
Bell was actually the team's ace pitcher (as a rookie), which is not apparent in any other published records of his career.

W-11 (led team; 2nd place had 5 wins)
L-7
ERA-4.53 (best of team's pitchers over 40 ip)
G-25
GS-14 (led team)
CG-9 (led team)
IP-154 (led team; 2nd place had 119 ip)
H-180
HR-9
W-34
SO-66

Cuban League record for James (Cool Papa) Bell:

Season  Team         AB   R   H  2B  3B  HR  RBI  SB   Avg
1928-29 Cienfuegos  157  44  51  10   7   5   --  17  .325
1929-30 Cienfuegos  220  52  57  14   4   3   --  --  .259
1930-f  Cienfuegos   16   2   4   1   1   1   --  --  .250
1930-w  Cienfuegos   38   9  13   0   1   0   --   2  .342
1940-41 Almendares  138  24  41   5   3   2   16  11  .297 

Notes:

1928-29 -- Led league in runs, home runs, and stolen bases. On New Year’s Day hit 3 inside-the-park home runs at spacious Aida Park in Cienfuegos.

1929-30 -- Led league in at-bats, runs, and doubles. Cienfuegos won the pennant race; Figueredo’s text says Bell won the MVP award, though the list of MVP winners shows Bell’s teammate Willie Wells (.322, 43 runs) for that year.

1930-f -- Regular season ended after 5 games because of dispute about new stadium (La Tropical) lease.

1930-w -- A short special season (Unico) was played in place of incomplete regular season; Cienfuegos played 9 games.

The feature of his Cuban League record that seems remarkable to me is his high rate of runs scored relative to hits. It suggests a) he must have been drawing quite a few walks, and b) it seems unlikely that he was thrown out much on the bases. Might he have been closer to Henderson than to Brock?

I doubt that he was as good as Henderson, who is one of the top 50 players ever. However, I am getting the mipression that he may have been better than Brock. More walks (Lou was never very patient) and better defense. As it stands right now he may just be a HOMer, but I think he will have to wait until we are engraving three plaques per year.

Wilson, Beckwith, and Suttles are/were all better IMO.

Maybe the upside is Tim Raines and the down side is Willie Wilson? Throw Willie Davis and Kenny Lofton into the mix?

The one thing I don't doubt: he was, indeed, very fast, and he used that speed to play baseball.

I'm also wondering if he's got any of the Harry Stovey effect working for him: speed showing up as power, and scoring more runs than the raw offensive numbers justify.

I like the Kenny Lofton comparison, but with a longer career. Raines had some great OBP's, so unless we are sure that Bell got on base at a near-Wilson clip (for an era high in BA) I don't want to make that comparison either. Then again, I seem to be the guy that underrates the NeL player in all of these threads.

I know this underates his speed, but I see George Van Haltren as a comp--CF's who initially had some success as pitchers, hit for B+ average, had moderate power, walked some but not a ton, ran the bases very well, played for a long time at a high level. Right now I've got them back to back at 10 and 11.

1926 St. Louis Batting Averages:

1b Suttles, Mule 0.498
c Murray, Mitch 0.396
ut Ross, William 0.379
ss Wells, Willie 0.371
of Bell, Cool Papa 0.355
ut Bobo, Willie 0.347
3b Creacy, Dewey 0.337
of Redus, Frog 0.323
of Russell, Branch 0.319
2b Russel, Johnny 0.313

TEAM .364

There is one very odd thing about Cool Papa Bell's career that should be remembered. He converted from a right-handed hitter to a switch hitter in 1927. If he hadn't done this, his statistics would be somewhat, to say the least, different.

Apparently, Candy Jim Taylor, the St. Louis Stars' manager was the one who wanted Bell to become a switch-hitter. His reasoning was that the superhumanly fast Cool Papa would be better able to beat out bunts and infield choppers left-handed.

This switch killed Bell's power. A natural right-handed batter, Cool Papa (since most pitcher's are right-handed) batted left-handed after converting in most of his at bats. When he first converted, Bell stated that he would often switch back to righty once he had gotten in the hole left-handed.

The Negro Leagues, because of roster size limitations, had an absolute fetish for versatility, sometimes to the detriment of the player's actual careers. If Bell had played in the Majors, I don't think that he would have ever been converted to a switch-hitter.

The only Major League player that I can ever remember doing something like this was the unusually named UL Washington. It killed his career.

But there are, at least, three major Negro League stars who seem to have converted to switch-hitting (Bell, Dihigo, Mackey) to the detriment of their hitting ability. Of course, in every case, it was a right-handed batter converting the majority of his at bats into worse left-handed swings (I have never heard of any left-handed batter being converted into a righty).

The switch cost both Mackey and Bell most of their power and (with his conversion into a starting pitcher added on top) kept Dihigo from being the greatest hitter of the 1930s, black or white.

(There were probably a bunch of lesser Negro League players would did this too; but, of course, as with everything in the Negro Leagues, it is hard to tell.)

The effect can be plainly seen in Bell's statistics (from Mac 10 with Gary A's 1928; 1925-26, 174 games, 1927-28, 176 games, pro-rated to 154):

YEARS-AB-H-2B-3B-HR-BA-SA-SB
25-26 154 648 232 47 12 23 .358 .574 42
26-27 154 660 211 34 09 08 .320 .435 25

Two notes on these stats:

1) Bell's 1925 and 1926 slugging is, of course, inflated by the extreme right-handed hitting St. Louis park effect.

2) Stolen bases were the most haphazardly kept (i.e. recorded in box scores) of all Negro League statistics and these figures probably mean very little other than Bell would have stolen at least this many and possibly more.

After converting to switch-hitting, Bell struggled for a few years (1927 to 1929) before starting to improve (in 1930, from Mac 10, he hit .352 and slugged .542, though, of course, 1930 was a great hitters year). From 1931 to 1936, Bell hit .322 (739-238). In context, this was much better than his 1927-1929 statistics.

By the end of his career, Bell was a high average hitter with some power. His 1938 to 1941 Mexican statistics (287 games), at ages 35 to 38, pro-rated to 154 games:

G-AB-R-H-2B-3B-HR-RBI-BA-SA-SB-BB-SO
154 638 166 234 39 20 14 90 .367 .556 34 84 43

In the Majors of that time, Bell would have been hitting around .320-.330 and slugging around .450-.475 with about equal numbers of walks and strikeouts (one each per every 10 at bats).

As for comps, Lou Brock really doesn't fit, unless perhaps Bell never converted to switch-hitting. Brock had, especially in the context of his time, pretty good power somewhat limited by his problems with the strike zone. Bell seems to have had pretty good plate discipline.

On top of that, Brock was a left-fielder who was defensively challenged and made up for it with his speed. Bell, as Gary A's stats seem to show and by reputation, was a great centerfielder. Brock's best Negro League comp is Sam Jethroe.

Bell's best comp is probably Max Carey though I think Bell was significantly better than Carey.

Max Carey 1921-1925 in the lively ball era.

YEAR-AGE-BA-SA-SB
1921 31 .309 .430 37
1922 32 .329 .459 51
1923 33 .308 .452 51
1924 34 .297 .412 49
1925 35 .343 .491 46

(Carey's career pretty much fell off the cliff at age 36 in 1926).

Of course, Carey was known for his incredible stolen base percentage and Bell's number of caught stealings are unknown and probably always will be; but it is pretty much apparent that Bell would have been stealing 50 to 60 bases in his best years like Carey.

Considering the length of his career, it is quite likely that Bell would have stolen about the same number of bases as Cobb did (892) if he had been in the Majors. It is quite likely that the career record would have been Bell's and not Cobb's when Brock broke it. In the context of Bell's era, that would have been impressive.

As for direct era comparisons, the best comp for Bell would probably be Doc Cramer, though comparing Cramer to Cool Papa is kind of like comparing a show pony to Man O'War.

In the Majors, Bell would have played from about 1924 to 1945 and (especially considering he batted first, was durable, and would have been getting 650 or so at bats a year) cleared 3000 and probably 3500 hits with ease.

And, though Bell gets no credit for it, I think that, in the Majors, Bell would have (operating under the idea that, if it isn't broke, don't fix it) never become a switch-hitter and had an even better career.

Rereading the thread and noticing my grammatical mistakes, I noticed that I forgot to include my contemporary comp, which is, of course, Tim Raines.

If Bell had played in the 1980s, I think he would have had a better BA than Raines but probably less walks resulting in about the same OBP. With the switch-hitting, Bell would have probably had less power; but, without it, probably a little more (always remembering that there is a speed component to slugging).

Bell would have stolen significantly more bases in the Raines era and probably cleared 1000 easily. Of course, Bell was a centerfielder and would have had a little longer career than Raines; but, all in all, I think they are quite comparable.

Was Bell the greatest centerfielder of his time? I would think career-wise he has to be a definite contender for the honor. I'm not sure he can beat the Averills, Bergers, or Simmons on peak, though.

Fairly clearly not as good as Simmons or indeed Raines -- let's not get carried away here; this is an overblown reputation to match Beckwith's undiscovered ability. Medium average singles hitter with some speed -- Willie McGee, say.

Fairly clearly not as good as Simmons or indeed Raines -- let's not get carried away here

I'm not getting carried away, karlmagnus. I'm just asking a question. If you have seen my prelim, you will notice that Bell is not on it. For the time being, I'm not bullish on him.

But he did, I believe, play the most games in CF for any of his contemporaries (Simmons mostly played in LF, BTW), was a great defensive and base stealing star, and could hit some (though it appears that he wasn't Ty Cobb as some chroniclers have suggested, that's for sure :-). I don't see that combination as inner-circle material, but it may be enough for a spot on my ballot eventually.

Karl Magnus, as always, reveals his anti-Negro League bias and lack of knowledge with his comparison of Bell to McGee: medium average singles hitter with SOME speed.

Saying Bell had some speed is like saying Ruth had some power. Bell was incredibly fast, almost superhumanly fast, and every player and sportswriter, white or black, who saw the man play commented upon it.

However McGee is, in some ways, a good comp. McGee apparently had a good personality, played in St. Louis like Bell where he was beloved, and even had the same monster year followed by everything-evens-out year pattern (1985-86) that Bell had (1940-41) at one point in his career.

But McGee, although he played until 40, aged badly both durability and performance wise and had poor strike zone judgment (career high 38 walks, 33 BB and 91 SO per 162 games) throughout his career.

And, most importantly, McGee had nowhere near Bell's speed. McGee stole 352 bases during his career with a high of 56 in 1985, his career year. But, in the early 1980s and on that St. Louis rug, that is not really that impressive.

Bell probably would have stolen 50 to 60 bases in the Majors at his peak in the early 1930s; but, in context, that is like stealing a 100 bases or so in the early 1980s. From 1920 to 1960, stealing 60 bases would have lead either Major League probably way more than 90 percent of the time.

Another modern Willie perhaps makes an even better comp for Bell, that being Willie Wilson. Like Bell, Wilson played an extraordinary center field and was superhumanly fast. I think that Wilson was the fastest player I have ever seen with my own eyes.

But Bell, who was pretty much beloved by his peers and consistently durable until he was quite old (baseball-wise) in his career, was obviously a much better player than Wilson.

Wilson had extreme character issues, being consistently portrayed by his peers and sportswriters as one of the biggest a-holes in the world (not to mention his drug habit).

But, most importantly, Wilson never really worked to improve or expand his game. Wilson played until he was 38 or so, simply coasting by on his natural god-given speed-based gifts.

Wilson stole 83 and 79 bases in 1979 and 1980 at the ages of 23 and 24; but, after that never really was the dominant base stealer of his time, apparently because he didn't really like making the effort to steal bases.

Also, like McGee and unlike Bell, Wilson had poor strike zone judgment with a career high of 39 BB and 32 BB and 86 SO per 162 games.

If Wilson had Bell's personality, he would have stolen over 100 bases some year easily and lasted into his 40s in the Majors.

A comparison of the modern careers of McGee, Wilson, and Raines is interesting.

Willie McGee (b 1958, career 1982-1999)
2201 G, 2254 H, 100 AOPS, .295 BA, .396 SA, 352 SB
Willie Wilson (b 1955, career 1976-1994)
2154 G, 2207 H, 093 AOPS, .285 BA, .376 SA, 668 SB
Tim Raines (b 1959, career 1979-2002)
2502 G, 2605 H, 124 AOPS, .294 BA, .425 SA, 808 SB

Bill James credits McGee with 224 WS (high 36), Wilson with 237 (high 31), and Raines with 390 (high 36). Interestingly, you can see right in the Win Shares the defensive superiority of Willie Wilson over Willie McGee.

So, basically, you have, in Cool Papa Bell, a player who was as fast on the bases and as gifted in center field as Willie Wilson with the good personality and belovedness of Willie McGee (or Mookie Wilson, another partial comp) but who is also durable (lasts as a regular deep into his 30s), motivated, and has much better strike zone judgment than either WW or WM (or MW).

In other words, you have Tim Raines with more defensive value and more speed and probably a better BA but a little less walks and probably power over a probably longer career that does not have the drug-related early career problems or lupus-related late career problems of Raines.

And, of course, this doesn't address the fact that, pre-integration, it was a lot easier to shine and pile up WS in the Majors than after integration.

I think Bell would have cleared 400 WS if he had played in the Majors, with a good portion of it being defensive value.

Was Cool Papa Bell the greatest CF of his time? Certainly not if Turkey Stearnes was "of his time." It seems pretty clear we've already made that decision.

Was Cool Papa Bell the greatest CF of his time? Certainly not if Turkey Stearnes was "of his time." It seems pretty clear we've already made that decision.

You're absolutely right, Marc. I forgot about the Turkey.

Karl Magnus, as always, reveals his anti-Negro League bias and lack of knowledge with his comparison of Bell to McGee: medium average singles hitter with SOME speed.

There's room for more than one "gadfly" here. :-)

Once again, we're just trying to get at the truth here. I can only put fifteen names on my ballot. Comparison to 70s-80s speedsters doesn't necessarily help me much. I don't think there was much need to manufacture runs in the 20s-30s.

A Cuban note on Bell:

In 1928-29, Bell played his first season in Cuba for the Cienfuegos club in Cuba. Cool Papa batted .325 and slugged .573 while leading the league with 17 SB in 157 ABs. The actual number of games that Bell played for Cienfuegos is unknown.

Cienfuegos played 53 games to decision and Bell obviously did not play the whole season. Of course, Negro League players, for their first Cuban League season, were usually contracted by Cuban League clubs after the season had started. If they played well, they were asked back.

But this supports the idea that Bell was stealing 1 base per 10 at bats during his prime.

2) In 1929-30, Bell returned to Cienfuegos and lead the league with 220 at bats; but his SB total is unknown. Cienfuegos played 51 games to decision. In other words, Bell would have batted 664 times in a 154 game schedule (He probably played every game).

If Bell had stolen bases at the same rate per at bat in 29-30 as he did in 28-29, he would have finished with 72 stolen bases.

There is, of course, sample size issues (etc) with this calculation, but it is simply intended to give a feel for how many bases Bell was stealing in his prime.

In his whole Cuban career, Bell stole 30 bases in 333 at bats, though 11 of these came at age 37.

Remember that high-run environmentrs reduce the value of a stolen base attempt in two ways; (1) by decreasing the value of a base and (2) by increasing the cost of an out. Given the run environemnt of the major leagues in the 1920s and 1930s, Bell would have had to steal a lot of bases at a clip somehwere around 80% before they would have added much value to his career. His speed might help in other ways (defense, doubles and triples, taking extra bases) but the impact of the steals would have been marginal.

We seem to be all over the map with our comparisons. So far we've had George Van Haltren, Lou Brock, Willie Wilson, Willie McGee, Tim Raines and Rickey Henderson thrown out as possible comps.

Here's what I think we know about Bell:

1.) He was really, really fast. One of the fastest men to ever play professional baseball, if not the fastest.

2.) He was a great defensive center fielder.

3.) He played for a long time.

4.) He had plus batting average, minus (a little) power.

5.) His offensive development was, shall we say, uneven? Some years he looks average offensively and other years he looks very good indeed.

The key question, it seems to me, is plate discipline. There's some interesting data cited above to suggest that he was good at getting on base, based on his ratio of runs scored to hits.

Based on what we know, I think we can throw a few of the comps out the window:

1.) Rickey Henderson. No evidence at all to suggest that he had Rickey's power or his insane plate discipline.

2.) Willie Wilson. Speed probably comparable along with defensive prowess, but plate discipline is a mess and I doubt he had as much power as Bell.

3.) Willie McGee. As a hitter for average might be comparable. Power might be similar (though I think Bell probably had a touch more). Defense definitely better.

4.) Lou Brock. I throw this one out the window right off the bat, because Brock was a mediocre corner outfielder whereas Bell was an excellent center fielder. That's night and day in terms of assessing a player's quality.

So that leaves us with Van Haltren and Raines. Looking closer at those guys...

1.) Tim Raines. Bell probably didn't have as much power. On base will be tough to match. Almost certainly not as good a hitter as Raines was, based on what we know. On the other hand, once again we've got a corner outfielder in Raines, and a so-so one at that, versus an excellent center fielder. If Bell can come within 10 percent of Raines's on base percentage, I'd say that Bell might have been as good a player.

2.) George Van Haltren. Here we have an actual center fielder, but probably not as good as Bell was. Hitting wise, we might have a close comp. Van Haltren had some average, some on base percentage, and a little power, but not an excess of any of the three, so he might be a fit. Bell might have been Van Haltren, with better speed and a bit better defense. That's basically where I have him slotted on my ballot now (Bell 15th, Van Haltren in the 20s.)

Personally, I don't think Bell was likely Tim Raines, I just don't buy the power. I don't often trust my (ample) gut, but it is telling me that Bell's power was much more foot speed than bat speed, and as such would be dampened more in the majors than in the Negro Leagues.

More pertinent to my ballot is whether Bell was a better player than Spots Poles. They're cut from the same mold: great-fielding CFs with steals and average. I see Poles as a suped-up Max Carey, and that's how come he's on my ballot where Carey never was.

Is Bell better than Poles?
Is he better than Carey?

In 1928, at least, I'm pretty sure that the St. Louis Stars were just not attempting to steal very often. I don't think there was much of a reporting problem in the St. Louis papers; the Globe-Democrat was always pretty good about reporting steals (though I do recall a couple box scores missing bottom sections).

Anyway, you can compare the team's stolen bases to their opponents: it turns out that, despite the fact that the Stars had many more base runners than opposing teams (StL OBP was .364, to their opponents' .334), opponents outstole St. Louis 66-49.

I also figured (crudely) team stolen bases per time on base (minus extra base hits and sacrifices--didn't have counts for offensive double plays or errors, so left them out for now). The league figure is .090; St. Louis opponents come in at .084, but the St. Louis offense is at .063, not the lowest in the league (that would be the Cuban Stars at .042), but well below average.

It does make sense, given the high number of runs the Stars were scoring, that Candy Jim Taylor might have reined in the basestealers a little.

Personally, I don't think Bell was likely Tim Raines, I just don't buy the power. I don't often trust my (ample) gut, but it is telling me that Bell's power was much more foot speed than bat speed, and as such would be dampened more in the majors than in the Negro Leagues.

Why would Bell's foot speed have been less help to him in getting extra base hits in the majors, and more importantly, why are his assets being judged strictly on a "how would it have helped him in the major leagues" criteria?

If Bell at his peak in 1928 was "reined in" by the manager, then he certainly wasn't Henderson, Brock or Carey. I stick to my analogy of Willie McGee, by all means plus a little extra fielding but possibly minus some plate discipline and SB/CS percentage.

Well, I'll fling a bit more fuel on this fire with Bell's MLEs, projected on the same methods and using the same conversion factors as the other recent NeL candidates (though I haven't posted updated MLEs for Turkey Stearnes using the offense-level adjustments for the 1930s, which would show him even better than the previously posted MLEs did).

Cool Papa Bell MLE Projections

Year Team  G   PA    BB  Hits TB    BA   OBP  SA
(1922 StL 74   296   17   86  118 .308 .347 .422)
(1923     84   336   21   74  104 .234 .282 .330)
1924     138   580   48  143  182 .269 .330 .343
1925     150   645   52  174  251 .293 .350 .423
1926     152   684   55  185  273 .295 .351 .434
1927     154   693   57  171  212 .269 .330 .334
1928     154   693   57  171  227 .269 .329 .357
1929     154   693   63  164  216 .260 .328 .343
1930     148   666   57  187  262 .308 .368 .430
1931     149   626   70  160  203 .287 .367 .365
1932 **  145   609   55  189  241 .341 .400 .435
1933 Pgh 154   647   72  167  219 .290 .369 .381
1934     151   634   72  187  213 .332 .407 .378
1935     154   647   71  173  221 .300 .377 .383
1936     150   630   74  169  196 .303 .385 .353
[1937 Santa Domingo]
[1938-41 Mexico]
1942 Chi 140   588   66  177  207 .339 .413 .396
1943 Home120   480   52  134  186 .313 .388 .434
1944     118   472   48  130  160 .308 .378 .378
1945      92   323   30   75   93 .257 .325 .319
1946      82   283   23   90   96 .347 .400 .368
NeL     2505 10592 1022 2846 3657 .297 .365 .382
+ 37-41 3230 13637 1371 3710 4665

**Det/Pgh/Hom/KC

Notes

1) These projections are to a major-league norm through 1929 and to the National League subsequently.

2) I’ve projected Bell for 1922 and 1923, but I do not think he would have been in the majors in those seasons, so I have not included them in any of the career totals.

3) Bell’s plate discipline looked very similar to Max Carey’s in the years for which we have BB data for Bell, so I’ve projected his BB and OBP on that basis.

4)I’ve projected Bell in 37-41 as overall, the same level of play as 1936 & 1942. It’s clear from Bell’s Mexico numbers and his numbers from 1942-43 that he wasn’t in decline during these years. He might have enjoyed a late peak, but I have no clear idea of how to convert the Mexican League, so I decided it would be better to base an estimate for these years on Bell’s surrounding NeL seasons. If he did enjoy a late peak here, it would raise his career ba, obp, and sa somewhat.

5) The drop in Bell’s production due to learning to switch-hit that gadfly pointed out is obvious in the projections. I have to think that he was further affected by being turned from a hitter who could take advantage of the StL park to a hitter who, by and large, could not. I agree with gadfly that there’s no way he would have been converted to a switch-hitter in the majors.

6) Bell’s numbers in Chicago in 1942 may be a little high, as I think that was a weak league that year, but they also suggest that Bell would have been advantaged by playing in large ballparks. Like Stearnes, his speed appeared to enable him to take advantage of the outfield space.

7) I look forward to seeing what Bell’s OPS+ looks like based on these numbers.

8) My view of him, at this point, based on these numbers, is as a Doc Cramer/Lloyd Waner type of hitter. He was definitely better than Cramer, who had less plate discipline and put up similar ba and sa in a higher-offense environment and who was not a base-stealer. He held his value a lot longer than Lloyd Waner did and his peak value was probably a bit higher, with his base-running skill again giving him a marginal advantage. As a fielder, I plan to treat him as likewise comparable to these two, who WS rates as A and A+, respectively.

9) I think there is no way Bell would have stolen more bases than Ty Cobb (as gadfly argues) had he played in the majors during his career. The style of play was too much against it. I could see him stealing 600-700 bases, but not 1000.

10) Overall, he was a player who was, I think, better suited to the NeL style of play than the ML style. Those who evaluate him on what he did, rather than what he would have done, might rank him a bit higher than these numbers show.

11) I think he was a player who has been overrated by history for reasons that Lloyd Waner has been overrated by history, but his career value is surely quite high.

Minus some plate discipline? And you are basing that on...?

In 1928 Cool Papa Bell was 25 years old. The reasons his manager "reined him in" for one season don't prove anything at all about his ability to steal bases over a career that spanned more than 20 years, though I know that you like to find any excuse you can to downgrade Negro Leaguers.

Thanks, Chris, for your excellent work on this.

Gadfly,

When you say that Bell would have played from 25-45, do you mean 1925-1945 or ages 25-45. Both seem unlikely. If he would have ended his Major League career in 1945 are you saying that he would have lasted longer in the majors than the NeL? And I doubt that anyone 45 years old would be a contributing regular, maybe a part time guy.

Right now I am deciding between Bell, Carey (who would be #13 if he hadn't been elected GVH (#15), and Edd Roush (a forgotten man but around #20 for me). Without further statistical information, like WS and MLE's he will be off my ballot.

And Chris, Do you want any help doing any of the calculations? Nor sure I can help with the theory stuff, but I look to be having a few days off this week and could help to calculate for some NeL guys.

9) I think there is no way Bell would have stolen more bases than Ty Cobb (as gadfly argues) had he played in the majors during his career. The style of play was too much against it. I could see him stealing 600-700 bases, but not 1000.

I agree with you, Chris. A thousand stolen bases sounds like a pipe dream for the twenties and thirties style of play.

I'm looking forward to see what his WS projections say, but it's not looking good for Bell.

BTW Chris, do you have Jud Wilson's WS projections ready yet? They will be crucial for this election, IMO.

Okay, so I just read Chris's post and it seems as if Bell palyed until 1946. Why are we looking at him now? Isn't 82 games in 1946 enough to push him back until 1952? I am not a biographical expert, but I would guess that Home is the Homestead Grays? Were they a minor league team at this point? I am confused.

And Chris' numbers have him lower than I figured. Of course they don't include SB's and Defense. I think I will credit him with being teh best basestealer of his era and a gold glove level CFer at least into the last 30's. His offense, if Chris is right, is troublesome, however.

Okay, so I just read Chris's post and it seems as if Bell palyed until 1946. Why are we looking at him now? Isn't 82 games in 1946 enough to push him back until 1952?

Because NeL stars tended to play much farther into their 40s than ML stars, their eligibility criteria are different. They become eligible after their 45th birthday or, if they have played past age 44, after they have been retired one full year. By these criteria, this is Bell's first year of eligibility.

I would guess that Home is the Homestead Grays? Were they a minor league team at this point? I am confused.

It is the Homestead Grays, though they were playing the majority of their games in Washington at this point. They were in the Negro National League, the eastern of the two Negro Leagues, and were the dominant team in that league at this time.


BTW Chris, do you have Jud Wilson's WS projections ready yet? They will be crucial for this election, IMO.

Now that I have MLE stats for Bell, win shares for Wilson is my next project. I should have them ready tomorrow. I should have WS projections for Bell shortly after that.

Now that I have MLE stats for Bell, win shares for Wilson is my next project. I should have them ready tomorrow. I should have WS projections for Bell shortly after that.

Much obliged, Chris!

Fantastic stuff, Chris. Lloyd Waner had an OPS of 746 and an OPS+ of 99, so Bell, an almost exact contemporary, should translate his OPS of 747 into about 99.2. As I said, Willie McGee, and not a HOMer or anywhere close to it. Jud Wilson, who I'd never heard of before "1947", IS a HOMer, and will be in my PHOM, and Beckwith may be, but Bell, no.

... they also suggest that Bell would have been advantaged by playing in large ballparks. Like Stearnes, his speed appeared to enable him to take advantage of the outfield space.


This is what i meant about Bell's footspeed increasing his power. As I understand it, the ballparks of the NgLs were more prone to having extreme measurements than even the old-tyme parks in MLB. Like I said, it's a gut feeling, and if my intuition about that is correct, then I think Bell would have had more opportunities to use his footspeed for extra bases in the NgLs than in the big leagues.

That said, the crux of the matter for Bell is what Chris Cobb wrote: if you are thinking about him in terms of what he might have done in MLB, you won't like him so much, because his style of play was much more valuable in the NgLs.

That said, I'm sticking with Max Carey as the player-type I associate Bell with. Bell might have been a little better or a little worse, and I think he'll probably end up anywhere between 10 and 20 in my ballot rankings.

Bell seems to fall right into the outfield glut. The hard thing for me is that the glut--all of whom are fairly similar in value--stretch out on my ballot from Duffy at #5 to guys like Griffin in the 30's. Small differences make a big difference in ranking. And we are essentially throwing darts as to where he falls in the glut. Right now I'm sticking to my decision to peg him to Averill and Van Haltren near the top of the glut, but he has virtually no chance to move up and may well move down to the level of guys like Roush, Veach, and Carey--all of whom I have/would have around 20 or 25. It is worth noting that lots of guys who might be comps (Hartsel, Roy Thomas, Spot Poles) are outside my consideration set.

karl, I think that it is a bit early to totally blow off a guy with Bell's reputation, though the early returns are not encouraging.

My other comment is I can't get too excited by the fantasies of what NeLers would have done in the MLs. I mean MLEs are one thing. They represent a fairly straight line linear analysis. But when we say that, well, so and so woulda played a different position and woulda played different seasons and wouldn't have been allowed to steal so many bases, I mean this is all unknowable.

My goal with the NeLers is to elect the best 20-25-whatever of them based on what they did in the real world in which they lived and played ball. Of course, integrating those players into a ballot and a HoM that includes white MLers, well, yes, we have to have a way to compare the two. But significantly altering reality, I don't know.

IOW if Joe Blow was more valuable in the NeLs than John Doe was, I have a hard time regardless of what their particular skills were of concluding that it woulda been the other way around in the MLs. Or even if I draw that conclusion, I'm not sure I'm comfortable voting that way. I think that's a stretch.

Similarly, if Buzz Arlett was not as valuable for his minor league career as Ike Boone, I don't see how we could possibly project more value for Buzz in the bigs. Now we're playing god.

Fantastic stuff, Chris. Lloyd Waner had an OPS of 746 and an OPS+ of 99, so Bell, an almost exact contemporary, should translate his OPS of 747 into about 99.2. As I said, Willie McGee, and not a HOMer or anywhere close to it. Jud Wilson, who I'd never heard of before "1947", IS a HOMer, and will be in my PHOM, and Beckwith may be, but Bell, no.

Yeah, but in your last ballot you had Beckley second, citing that he would have had 3,200 hits, more than Ruth, or something silly like that. Well you will see in the above translation that Bell has 3,710 hits, more than anybody but three players in baseball history; 1,251 hits more than Lloyd Waner, and also 1,456 hits more than McGee. So yeah, if Willie McGee had picked up an extra 1,500 hits or so, played great defense instead of mediocre defense, then you'd have a reasonable comp. You'd also have a slam-dunk Hall of Famer.

The comparisons with Lloyd Waner only go so far. As has been noted, Bell played for longer and at his peak was better and better for longer. Also, as Chris Cobb suggests, Bell's game translates much better to his actual environment, the Negro Leagues, than it does to the National League. Why we should be judging Bell based on a league that excluded him rather than the league he actually competed in I don't know. Again, I have him 15th on my ballot, and that's where he's going to stay.

Karlmagnus, I'd be interested to know how many Negro Leaguers you think would make your personal Hall of Merit.

IOW if Joe Blow was more valuable in the NeLs than John Doe was, I have a hard time regardless of what their particular skills were of concluding that it woulda been the other way around in the MLs. Or even if I draw that conclusion, I'm not sure I'm comfortable voting that way. I think that's a stretch.

Exactly. The MLEs are a valuable tool to assessing a player's stats, but I think it is absurd to judge a Negro Leaguer based on how his game would have translated to the National League or whatever. My understanding of the rules was that the goal was to judge a player's value to histeam, how much he helped his team win games. We aren't trying to figure out how good these black players would have been, we're trying to figure out how good they actually were. Assessing how Bell's speed game would have fared in the major leagues in the 1930s is interesting, but no more relevant than trying to assess how Heinie Manush's game would have translated to the black leagues of the time.

My PHOM would contain 12 NLers or a few more, nowhere near the number the group is electing, which in many cases (Fosters, Suttles, for example) seems to me to be pure affirmative action.

I have included Bell's NL stats, in which he played till his mid 40s but not the stats in which he commuted between Mexico, Guatemala and Mars. If we can get Martian MLEs, then obviously one can include those; on the other hand I think it vanishingly unlikely he would have played in the ML till 1946, except during the heavily discounted war years.

I remain open to further eveidence from Chris and others on the man, but if his WS comes to more than Beckwith and Suttles, I shall discount all 3 accordingly. One can only rank these players relative to each other, helped enormously by Chris' work, but if translations start telling me to elect the entire Kansas City Monarchs franchise, I ain't doing it.

There are plenty of VERY worthy MLers that we haven't yet elected; Beckley having more hits than Ruth is a statistically documented fact, and his increase to 3200 simply reflects adjusting to 154 game seasons as we have done for all 19C stars and for the NLers. He is a little better than Wilson, a lot better than Beckwith and HUGELY better than Bell. If you don't have him at least on your ballot, your analytical methods are out to lunch.

Chris Cobb-

Very interesting MLEs. As always, I will point out that, in my opinion, the MLEs understate the BA by 5 or 6 percent and the isolated slugging by about 10 percent or more. A more believable final career BA/SA to me would be about .315/.410 with my guess that the SA is still to low since you have decided to give the Negro Leaguers a 1930s low NL offense discount. But still, understated or not, 3710 hits?

That sure as hell doesn't look like Lloyd Waner or Willie McGee to me.

As for matching Ty Cobb's career stolen base mark, I figured it like this: Bell would have stolen 60 to 70 bases at his peak and was still stealing bases at 30 to 40 a year clip in the 1944 and 1945 NNL (19 stolen bases in 86 games).

[One interesting thing to note is that Bell, at the ages of 41 and 42 in 1944 and 1945 played EVERY NNL game for the Grays. Bell, who was known as a clean-living guy who really took care of himself, was awesomely durable.]

If you have a player who steals 30 bases a year at the beginning and the end of his career with a peak of 60 to 70 and plays for 22 years (1924 to 1945, Note to JSchmeagol, I meant years not age), you get about 800 to 900 stolen bases.

To steal over 892 bases in 20 years, a player would have to average 45 a year and, of course, 45 is the midpoint between 30 and 60. As for what was possible in the era of Ruth, I just looked at Max Carey who was averaging 45 to 50 stolen bases through his early 30s.

Of course, stolen bases are more of an 'elective' stat than any other (i.e. the player has to want to, and be allowed to, steal); but I think Bell would have been capable of that, especially considering that it would have probably been his meal ticket to a larger salary in the Majors (i.e. Bell, the stolen base king).

But, on the other hand, the times and style of play were against him as you point out. It would have certainly had a lot to do with what team and park Bell actually played on. But I can easily see Bell being the career SB record holder under the right circumstances.

Dr. Chaleeko-

Spot Poles was almost undoubtedly more valuable at his peak than Bell, although some of that is timeline related. I haven't read his thread; but Poles, in addition to being batting champion material with some good pop and speed probably almost equal to Bell, walked a ton.

He was a stronger faster better version of George Burns, the Giants lead-off man.

But his career was nowhere near as long as Bell's career. If you voting for peak, Poles is probably your man. On the other hand, if you are voting career, Bell buries Poles and Carey. Just consider that Carey had his last season as a quality Major League regular at age 35.

In my opinion, Bell would have had his last year as a quality Major League regular at 41 in 1944.

Karlmagnus,

Why are you establishing a quota for how many NLers you are willing to elect? Why not examine them on a case-by-case basis?

In the immediate post-war era, how many black players emerged who were at the HOM level? Does this suggest to you that in all the years prior to 1947, only 12 players played on the other side of the color line who were good enough to be in the HOM?

The fact that you were ignorant of Boojum Wilson's existence prior to a few weeks ago does not exactly inspire confidence that you are really giving these players their due consideration. The fact that, if Bell's WS come out to more than Beckwith, Suttles and Wilson, this will, rather than boost Bell's stock, devalue the other three's, is also not very encouraging. I mean, George Van Haltren has 344 WS, and last I checked he wasn't on your ballot. So since he has more Win Shares than Ernie Banks, shouldn't we expect you to disregard Banks when it's his turn?

It seems to me that you are applying a double standard. You cite Beckley's hits as a documented fact (leaving aside the wisdom of using career hits to determine value), when Chris posts MLEs for Bell, you praise him for it, declare that your comparison of Bell to Willie McGee is valid, and then ignore the fact that those same MLEs also suggest that Bell would have had many more hits than Beckley. It seems to me that you are using any excuse you can find to ignore Negro Leaguers. If you applied the same exacting standards to white major leaguers I suspect your ballot would look quite different.

I've explained my rationale ad nauseam; it's called demographics. You are misusing statistics to satisfy your guilty liberal conscience and then insinuating that those who don't agree with you are racist. Enough, already.

David:

karlmagnus is establishing a quota not for any bias against Negro Leaguers, but based on economic issues. He has also brought to our attention economic factors during the 1890's, so it's not a black-white thing. I converse with him online and can state that he is not bigoted.

Not that I agree with his economic theory in this instance, mind you, but he does write a column on economic matters and I would rather not compare my gray matter with his. :-)

John, thank you. I should not allow my irritation to get the better of me. Probably a sign that the gray matter is flaking at the edges:-))

Besides which, isn't it more than a little elitist to say that people shouldn't be participating here if they weren't already familiar with Boojum Wilson? It seems to me that David's ballot changed a bunch by his third one, right? Meaning he didn't know a whole lot about some candidates?

Welcome to the club, David, we're all learning every day.

As for the demographic theory, however, karl probably knows I don't buy it. If I did, I'd be a lot more concerned about overrepresenting the Irish than the blacks.

HoMers are outliers. Demographics are largely irrelevant.

If I did, I'd be a lot more concerned about overrepresenting the Irish than the blacks.

I don't see a problem with a surplus of Irish lads, Marc. In fact, a surplus of Italian players will mollify my Italian ancestors, too. :-)

WARNING: LONGWINDED STATEMENT OF PURPOSE TO FOLLOW

I guess I'm a contrarian of a different sort from Karlmagnus. In spite of several messages in this thread (and in others as well) that downplay the importance of interpolating Negro League performance into the white leagues (and of comping), I do believe that it's important to understand how what we know about a Negro League player's game would have translated into the majors. In fact, I base my evaluation of Negro Leaguers on this very thing. There's a couple complicated reasons why.

1) With no one compendium of reliable Negro League stats at hand to base assumptions on, it's often difficult to have a firm sense of a player within the context of his own era. I don't think too many of us know all that much about the Negro Leagues in terms of pennant races or depth of talent or value distributions. More than we used to, yes, but probably not one-twentieth of what we know about the white leagues. With candidates coming along at an essentially random clip, and usually independent of one another, it's hard for a person in 2005 to understand how the stars of the league really fit into its larger contexts like team, league, community, etc....

2) Aside from statistics, contexts in the Negro Leagues appeared to shift so frequently that it's an easier thing to examine the type of player a guy was than to analyze much else about him. Consider these factors:
-the quickness with which leagues came and went
-the quickness with which franchises came and went
-the quickness with which franchises changed parks
-the (I believe) more extreme nature of NgL parks
-the helter-skelter scheduling
-the frequency with which some players changed teams and leagues, thus changing the league's competitive dynamics.
-the segmenting of careers due to play in Cuba, Mexico, and other locales.

I know you all know this stuff already, and I know Karl will say "that's why I think we're electing too many NgLs!" But what I'm trying to explain is that I disagree with both his conclusion about electing too many NgLs and with what seems to be a prevailing conclusion that thinking about NgLs in an integrated environment isn't an appropriate approach.

Instead, I see compelling reasons to identify each Negro League player with a type of major leaguer and to fit him within a continuum of similar players. In many ways, my voting system (such as it is) does the same thing for white players. I'd say, in fact, that we all do this for white players because that's what Bill James taught us to do with things like Sim Scores. It's simply an underlying assumption to our bag of tricks.

Whoever said it is right: that by relying on grouping a Negro Leaguer with similar white guys I'm not addressing how much value he had to his team. But the fact is that I feel utterly unqualified to do so. I feel much more qualified to figure out the type of player his stats and MLEs suggest he was and then to consider how that type of player fared in the majors as a group to create a proxy for his value and a picture of his game. This has the added benefit of avoiding basing decisions on either league contexts I don't know much of anything about or on assertions of value that are based on superlatives, anecdotal evidence, and outlying performance (like setting SB records or not taking up switch hitting...no offense Gadfly!).

As I've said before in another thread, I'm not telling anyone that Bell was Max Carey (or Suttles Willie Stargell or Wilson Rod Carew), but I am suggesting that he fits the same general profile as Carey as well as Poles, GVH, and others. And this profile creates a more comprehendable context of comparison from which to build and assess his case.

Sorry if that was long and dogmatic, I just felt like I ought to put my methods and feelings out there in response to concerns about comping.

Okay, so Primer just ate my last post so I will try and do it again.

Gad,

I find it very unlikely that Bell would have been a productive player at the age of 41 in 1944 in a Major League that wasn't war-depleted. This would bring down his hit and SB totals (not that such things really factor into my decision to vote for a guy).

Chris,

Do those MLE's take into consideration the fact that 1943-1945 the NL and AL were hardly Major League quality? And if not how much do you think we should then downgrade Bell's numbers?

Karl,

"If you don't have him [Beckley]at least on your ballot, your analytical methods are out to lunch. "

I think that if you vote for a borderline All-Star over a player who was the best player in the game for a few years (*cough* Hughie Jennings *cough*) then YOUR methods are out to lunch!

Sorry, just though that deserved a snarky comment ;-)

Cool Papa Bell relative stats are ready.

-First you have your year, team, PA.
-Second you have Chris's MLE's
-Third, in parentheses, you have pitchers-removed offense context. MLB for the 20s, then NL
-Fourth, you have AVG+/OBP+/SLG+
-Lastly, is the OPS+

All of the intermediate numbers are there. Let me know if something doesn't look right.

1922  StL 296  0.308/0.348/0.422   (0.297/0.359/0.415)   104/ 97/102     99
1923  StL 336  0.234/0.283/0.330   (0.292/0.356/0.405)    80/ 79/ 81     61
1924  StL 580  0.269/0.329/0.343   (0.294/0.356/0.406)    91/ 93/ 84     77
1925  StL 645  0.293/0.350/0.423   (0.300/0.364/0.425)    98/ 96/100     96
1926  StL 684  0.295/0.351/0.434   (0.289/0.355/0.402)   102/ 99/108    107
1927  StL 693  0.269/0.329/0.334   (0.292/0.355/0.406)    92/ 93/ 82     75
1928  StL 693  0.269/0.329/0.357   (0.290/0.355/0.412)    93/ 93/ 87     79
1929  StL 693  0.260/0.328/0.343   (0.298/0.363/0.432)    87/ 90/ 79     70
1930  StL 666  0.308/0.366/0.430   (0.312/0.370/0.464)    99/ 99/ 93     92
1931  StL 626  0.287/0.367/0.365   (0.285/0.344/0.403)   101/107/ 91     97
1932  NeL 609  0.341/0.401/0.435   (0.284/0.337/0.412)   120/119/106    124
1933  Pgh 647  0.290/0.369/0.381   (0.274/0.327/0.376)   106/113/101    114
1934  Pgh 634  0.332/0.409/0.378   (0.287/0.342/0.408)   116/119/ 93    112
1935  Pgh 647  0.300/0.377/0.383   (0.286/0.341/0.407)   105/111/ 94    105
1936  Pgh 630  0.303/0.386/0.353   (0.286/0.345/0.400)   106/112/ 88    100
--
1942  Chi 588  0.339/0.413/0.396   (0.256/0.328/0.356)   132/126/111    137
1943 Home 480  0.313/0.388/0.434   (0.265/0.334/0.360)   118/116/121    137
1944 Home 472  0.308/0.377/0.378   (0.268/0.335/0.377)   115/113/100    113
1945 Home 323  0.257/0.325/0.319   (0.273/0.343/0.377)    94/ 95/ 85     79
1946 Home 283  0.347/0.399/0.368   (0.263/0.338/0.368)   132/118/100    118

I had to try to put together a list of candidates on the fly for my first ballot, as I had to travel that week. I admit I made mistakes and will do so again, but it was not because I was ignorant of the candidates. It was because I had not had the time to compress all the data necessary to make the most informed ballot possible.

I don't think that knowing of Boojum Wilson's existence prior to a few weeks ago is some sort of necessary qualification, but it makes me wonder how much karlmagnus is actually considering these candidates. It's easier to toss off a guy you've never heard of before or never really considered as a player than it is another guy whose stats you know well.

I never insinuated that karlmagnus was racist. That's absurd. I am insinuating that he is prejudiced against Negro Leaguers, not because they are black, but because they played in the Negro Leagues. Given all the talk about "demographics" and "economics" it seems to me that karlmagnus is indeed precisely saying that the Negro Leagues were very weak, and as a result of that he's only voting for the players who were obviously great, as if the quality of play in the league went right from the historically great to the mediocre with no stop in between. If I am wrong then prove it to me.

Demographics? Economics? Yeah, I'm aware. Great Migration, right? And lesser incentive to play baseball because you aren't as well compensated. Right. But this leaves a few things out. First, it fails to explain why knowledgeable baseball men like John McGraw thought the best black players were as good as the best whites. It does not explain why blacks did so well so quickly once they were brought to the major leagues, and it does not explain why blacks did so well against major league competition when they did get the opportunity to play them. It also doesn't explain why black stars did so well playing in other countries. There's a lot that it doesn't explain.

As for the migration North, the argument I suppose would be that since most blacks lived in the South prior to 1920, that therefore the quality of competition was not as good, as little scouting was done in the South and there were no top-notch black teams in the South. The problem with this analysis is that it fails to properly understand black culture at the time. During the first two decades of the twentieth century, for instance, the Chicago Defender was read throughout the South, carried there by blacks who worked on the railroads. Southern blacks were well aware of baseball up north, and without digging into all the numbers, it seems to me that more Southern black ballplayers found their way north than was the case in the major leagues. Just looking at some of the best NeLers of the 1900-1920 period, you have Rube Foster, from Texas, Bill Foster, from Texas, Spots Poles, from Virginia, Dizzy Dismukes, from Alabama, Dave Brown, from Texas, Charlie Grant, from Cincinnati, John Henry Lloyd, from Florida, Bruce Petway, from Tennessee, Louis Santop, from Texas, the list goes on. Indeed, one could argue, convincingly I think, that black clubs were better at recruiting Southerners during this period precisely because they knew that most blacks still lived in the South. They barnstormed through the South regularly and that's where they got many of their top players
.
Again, I repeat. I am not, in any way, suggesting that you are racist. That's a red herring that has nothing to do with the discussion at hand. Again, from what I've seen of your comments on the NeL threads as well as your ballots, it seems to me that you take the first opportunity to dismiss a Negro Leaguer's candidacy based on any chink in the armor; wheras major league players are given far more leeway. I don't think that's right.

The comment about me being a liberal with a guilty conscience is completely absurd, as you would quickly recognize if you actually knew me. My main interest is the Deadball Era, not the Negro Leagues, which is a secondary area of interest for me. Negro Leaguers are on my ballot because they are the most deserving candidates. The best major league candidates get elected quickly. Hell, practically every player of note prior to 1890 has been shoved through the door by now.

Yikes... I didn't believe the Doc Cramer comparison at first, but there it is.

Those STL translations might be a little funny (though its probably tough to translate just a wild home park).

Could also be that he was just not the right guy for that park. Take your pick.

Chris:

Do you have any info other than the 1928 statistics posted by Gary A (# 6) on which to base your estimates of Bell's walks?

Anyone who knows:

Are there more NeL statistics available on runs scored? If so, I think it would be useful to post them for Bell.
Did Bell bat leadoff most or all of his career?

I ask these questions because for the 3 seasons in Bell's prime for which runs scored data are posted (NeL for 1928, Cuban L for 1928-29, 1929-30) he averages 1 R/G. Even for a high run scoring environment, that's fantastic. How many post-1900 leadoff hitters have ever averaged a run per game?

In his early Baseball Abstracts, Bill James made the point that although for most players runs scored are largely dependent on external factors, for leadoff hitters there are very predictable relationships between runs scored and the leadoff hitter's individual statistics. In Bell's case, we don't have a lot of info on important individual statistics (particularly walks and caught stealing). If we could get runs scored info, however, maybe we could try to back out some of the missing information.

Another comment. Although I think Chris is probably right when he says that Bell wouldn't have stolen 1000 bases in the majors, Bell clearly was a unique talent and unique players sometimes can change their circumstances. In 1957 I can't imagine anyone could have predicted that in only 5 years Cobb's single season stolen base record would be broken. Although the culture of 1930s baseball would have gone against such extensive base stealing, I can imagine that in the right league (the NL) and with the right manager (McKechnie?), Bell might have gotten the green light to steal as many bases as he could. And as Gadfly has argued, Bell was clearly capable of stealing 60+ bases per year.

The comment about me being a liberal with a guilty conscience is completely absurd

Karl,

Yeah, that one's misplaced,...because I'm the bleeding-heart liberal you're looking for. Did the College Republicans ever hate me! And don't get me started on redistributionism....

: )

But still, understated or not, 3710 hits?

That sure as hell doesn't look like Lloyd Waner or Willie McGee to me.

It's a question of career length. If Bell really could have sustained a career as a starter from 1924 to 1945 _and_ as a hitter of major-league average quality over that whole period, which was a period of rather high offense, then he would rack up a lot of hits. I have him down for 3230 games, as opposed to Lloyd Waner's 1992. If you project Little Poison into Bell's career length, he ends up with 3987 hits, but fewer walks.

Could Bell have been good for that many years? I don't know, but I'm doubtful. The NeL stats give every evidence of that degree of consistency, but _nobody_ actually played that many games in the majors before the 162-game schedule and the DH.

As for Jake Beckley, I'm not impressed by 19th century candidates who basically never led the league in anything (except triples once, in a year when offensive quality was way down because there were three major leagues.) Beckley had a nice long career. I'd say he was slightly better than Mark Grace, all in all.

It's a question of career length. If Bell really could have sustained a career as a starter from 1924 to 1945 _and_ as a hitter of major-league average quality over that whole period, which was a period of rather high offense, then he would rack up a lot of hits.

It seems to me your numbers indicate that Bell played quite well when he was older. We know he won the Triple Crown in the Mexican League in 1940. According to your numbers he's a better than .300 hitter every year in the '40s except for '45.

Maybe he was on PEDs?

:)

That his best OPS+'s come during the war yeasr makes me ask again, "were the MLE's adjusted for the war years?"

David,

But those years that Bell was hitting .300 were war yearsa nd should be discounted as such. And the Mexican league? I can't imagine the Mexican league as better than the Cuban league, let alone the NL, AL, NNL, etc.

Right now my tentative conclusion is that Bell's ML career would have lasted until the late 30's, maybe 1940, with maybe a comeback during the war years (43-45, 42 is kind of a transition year). However, that comeback would be of dubious value because of the competition.

That his best OPS+'s come during the war yeasr makes me ask again, "were the MLE's adjusted for the war years?"

Two points:

1) Bell's best OPS's come during the war years, but we don't have OPS+ scores yet.

2) The MLEs are calculated so as to project the player's performance into the major-league environment of that season. From 1930, that is a specifically National League environment. So all of my MLEs are for 154-game seasons and for whatever the offense and competition level of the relevant major league was.

And the Mexican league? I can't imagine the Mexican league as better than the Cuban league, let alone the NL, AL, NNL, etc.

The Cuban League was clearly the best Latin American league from about 1900-30, but after 1930 political instability and economic troubles led to a big drop-off. I don't know exactly how it compares with the other leagues, but after 1930 I wouldn't presume that it was necessarily the best.

1) Bell's best OPS's come during the war years, but we don't have OPS+ scores yet.

I posted them above #64 (with short comments at #66). Traffic has been heavy here that they might have been missed.

The OPS+'s are a bit odd. Let me know if something isn't right.

Thank you for the clarification Chris! Right now it looks like Bell's offensive peak was in the mid 120's.

Are there more NeL statistics available on runs scored? If so, I think it would be useful to post them for Bell.
Did Bell bat leadoff most or all of his career?

I know for sure that Bell batted leadoff as late as 1934. I'm pretty certain he did for years after that, and his Mexican League record seems to indicate that he was still leading off then.

As for scoring one run per game: for what it's worth, here are the runs scored for the top five men in the Stars 1928 lineup:
1. Bell 84 runs in 83 games
2. B. Russell 72 in 82
3. Wells 85 in 82 games
4. Suttles 71 in 80
5. Redus 71 in 82

Also, an incomplete accounting of Bell's 1934 record with the Pittsburgh Crawfords:
G-40 (team 40)
AB-163
D-5
T-0
HR-1
R-35
W-14
SB-10
AVE-.350
SLG-.399

I wasn't sure if anybody else has the Mexican League encyclopedia. Here are Bell's stats there, which also give us some W/K data:

YR--G---AB--H---D--T-HR--R--RBI-W--K-HP-SH-SB-AVE-SLG
38-040-160-051-07-06-06-051-17-20-12--1--1-09-356-588
39-058-226-080-16-02-05-055-24-26-19--1--6-12-354-509
40-089-382-167-29-15-12-119-79-45-24--0--3-28-437-686
41-100-421-132-21-15-04-085-48-65-26--1--4-14-314-463

Thanks, Gary.

The walk rates look similar to what Chris was using.

I'm trying to make sense of statistics like those for 1938 - 51 runs scored in 72 times on base. Even with him reaching scoring position on his own power 28 times, that is still a pretty amazing rate of runs scored. You just don't see records like that in the majors. Do you happen to know who would have been batting behind him?

karlmagnus wrote:
I've explained my rationale ad nauseam; it's called demographics.

I haven't seen a rationale; I've only seen a dogged insistence on a population-based quota (10 percent! 10 percent!), interspersed with ideology-driven insults (guilty liberals! affirmative action! you're calling me a racist! you're cuckoo!). If you actually had a demographic argument (as jimd does) and could present it without cherry-picking facts to fit your preconceptions, maybe it would be a contribution to the discussion.

Bringing the discussion back to Cool Papa Bell:

DavidFoss wrote:

Those STL translations might be a little funny (though its probably tough to translate just a wild home park).

Could also be that he was just not the right guy for that park. Take your pick.

I'm inclined to think that it is the latter, produced in part and exacerbated by the change to switch-hitting. Consider: you are a right-handed hitter playing in a great park for right-handed hitters, especially in terms of power. Bell didn't have much natural power, but he placed among the league leaders in home runs in 1926, so clearly the park was working for him: a 107 OPS+ at 23 years old is quite promising.

Then he's converted to switch-hitting. Whoops! Now he's hitting left-handed most of the time in a park that doesn't favor left-handers, while he's learning how to hit left-handed. It's a team-inflicted case of sinusitis, from which it took Bell three years to (mostly) recover, and (I think) it's not until he gets out of a park to which his new hitting style is intensely unsuited that his talents as a hitter really show.

That's the story I'd tell now, anyway.

When I do conversions for Willie Wells, we'll have another opportunity to see if the StL park-adjustments look excessive. I don't think they will.

Brent wrote:

Do you have any info other than the 1928 statistics posted by Gary A (# 6) on which to base your estimates of Bell's walks?

I didn't, but now I do :-), thanks to Gary A.'s Mexican League data. I got lucky in guessing Max Carey-style plate discipline, I guess.

jschmeagol wrote:

That his best OPS+'s come during the war yeasr makes me ask again, "were the MLE's adjusted for the war years?"

To return again to Bell's 1940s MLEs: The war effect would certainly be at work in 1943-45, but in 1942 not many players had gone into the service yet.

Three other factors could be contributing to this effect:

1) Bell was playing for the first time in parks that were highly favorable to his set of skills: I think this is likely in 1942. I don't know about the parks the Grays called home in 1943-45, but it's possible that they favored Bell above other players.

2) Park-adjustments might off. I park-adjusted Bell in 1942 for most of the Schorling park figure: I don't know if that's still valid in the early 1940s. I don't know anything about the Gray's parks, so I made no park adjustment there. If they were playing in hitters' parks, Bell's numbers might be inflated for that reason.

Does anyone know where the Grays were playing, and the reputation of the parks?

3) A league-offense adjustment might be needed. I've used median starter batting averages to compare NeL offense levels to the majors for 1930-39, and I have league-average data from gadfly from 1944 on, but 1940-43 remains a gap in the record. I haven't filled it in because I've had other things to do and it didn't matter to the cases of the players I was studying. Clearly, it does matter in the case of Bell, since we have to make a determination of how long he could have lasted in the majors.

Gadfly, do you have NNL league data for the early 1940s? If not, I will do a median-average study, which is less exact but, I think, better than nothing.

Overall, because of the absence of league-offense data, there's a somewhat higher degree of uncertainty about these numbers, but it should be possible to lessen that uncertainty.


These issues make Bell a hard player to evaluate: I'm not at all sure what I'm going to do with him. We have to consider the odd career shape produced by the switch-hitting project; we have to consider the long stretch in Mexico; we have to consider the effects of his base-running; we have to consider the nature and value of his career length; his defensive value (always hard to estimate precisely) will be a big factor.

I'll do win shares through 1936 tonight, hoping for more info on the level of play in Mexico, on ballparks, and on 1940s offense levels. If none of the latter is available, I'll do a median-average study for those seasons.

When I do conversions for Willie Wells, we'll have another opportunity to see if the StL park-adjustments look excessive. I don't think they will.

Thanks for the sanity check, Chris.

Also, would it be possible to get updated yearly MLE's for Suttles? He also played for STL for a while. Beckwith would be another good choice because he's near the top of the backlog. Of course, I realize you don't have infinite free time. :-)

For now, Bell is off my ballot. Those OPS+ nubers are just too hard to make up. Tons of hits with a low OPS+ because of a long career. Is Cool Papa the black Sam Rice?

OCF #20, seconded by Dr Chaleeko #49:
I'm also wondering if he's got any of the Harry Stovey effect working for him: speed showing up as power, and scoring more runs than the raw offensive numbers justify.

Unmeasured speed "works against" this type, too. Teammates advance fewer bases than usual on his hits and probably on his outs.

As things stand I will have Bell somewhere between 23-30 on my ballot. This puts him below Carey (were he eligible), GVH (#15), Averill (17), and Roush (21) However, he is above guys like Fielder Jones, Roy Thomas, and George Burns. It is possibe that he will move up, but I agree with andrew, that he would have had to be a combination of Henderson and Andruw Jones with the rest of his game to sufficiently overcome his offense.

One problem with extrapolating 70-80 game seasons to 154 games is that it doesn't take account of wear and tear over a career (qv barry Bonds) nor the probability of severe injury (which must surely double if the number of games doubles.) Early on I had Joe Start projected at 4,784 hits, being 2,910 on his ML stats adjusted to 154 games plus an appropriate allowance for 1861-70. In reality, while Start was a HOMer, he presumably wasn't a greater hitter than Ty Cobb or even probably than Rose. Bell's 3700, and other very high MLE counting stats based on 22-25 year careers are an error of a similar type, I think.

Karlmagnus,

I agree in principle with your theory about the small sample of games, though I have one question about it. If Negro League teams played as many barnstorming and exhibition games as it seems, this mitigates the injury and wear and tear fear a little bit.

I won't claim that it mitigates your conern entirely, because it's possible that in barnstorming the NgL teams could have coasted in games against less difficult opponents or that they may have rotated more players through the lineup in those games to rest star players. I would defer to our experts on that one.

Anyway, just testing your theory a little bit.

Does anyone know where the Grays were playing, and the reputation of the parks?

Griffith Stadium (reference).

Negro Leaguers played, if anything, more demanding and physical grueling seasons than major leaguers did. The travel accomodations were far worse, and the barnstorming games played in a year often exceeded the 154 major league schedule, plus many of them spent their winters playing competitive ball in Cuba or other Latin American countries. I don't think you can fault Bell for not playing enough baseball.

I did this for Suttles, so I'll do it for C.P. as long as I have the script ready.

Chris's totals don't include 1922/1923 (in parentheses in his MLE's). I'll do that here, too.

Totals (plus/minus 1 from Chris due to rounding)
10593 PA
9570 AB
2846 H
3658 TB

Averages
CPBell -- 0.297/0.365/0.382
Context -- (0.285/0.347/0.403)
Plusses -- 104/105/ 95

OPS+ = 100

Totals (plus/minus 1 from Chris due to rounding)

10593 PA
9570 AB
2846 H
3658 TB

Averages
CPBell -- 0.297/0.365/0.382
Context -- (0.285/0.347/0.403)
Plusses -- 104/105/ 95

OPS+ = 100

Digging around for someone with similar numbers doesn't get me any centerfielders. Here's someone in the neighborhood

Lave Cross

Totals
9710 PA
9072 AB
2645 H
3467 TB

Averages
LCross -- 0.292/0.328/0.382
Context -- (0.277/0.341/0.370)
Plusses -- 105/ 96/103

OPS+ = 100

Give Lave some bumps for schedules slightly under 154 games you get close. The differences of a good fielding 3B versus excellent CF, Cool Papa gets a plus on baserunning. Still, to look at the support for Cross shows that CPB shouldn't be getting his acceptance speech ready.

In looking at Bell's career OPS+ and thinking about his electability, it's important to keep in mind two things:

First, his 5 years out of the NeL from 1937 to 1941 aren't included in that measure. From everything we can tell from his MeL records and his NeL records on either side, Bell was _likely_ hitting in the 105-120 OPS+ range during these seasons. This would a) raise his career OPS+ somewhat (to, say 102-104) and would add 2500-3000 pa to his career.

Second, his career rate is drawn down noticeably by the years in which he was learning to switch-hit, so he will do better on a cumulative metric like win shares than he will on a career-average metric. Given the OPS+ profiles of Lloyd Waner and Doc Cramer, it seems likely that Bell would have kept his starting CF job even with these OPS+ numbers, so I wouldn't remove or deemphasize those seasons in his career, but I notice that, if these seasons were removed and his Mexican seasons added in, he would have a career OPS+ of 107-110 in 11000-11500 PA, which is a bit more than Max Carey's entire career, at (or a bit above) Carey's career OPS+.

All that's just to say that Bell's career is unusual, and there are ways in which a career OPS+ number might be a bit misleading about his value as a player.

Third, I've been looking at major-league outfielders playing past the age of 40, and I think that it is quite possible that Bell could have played until age 43. There are several examples of players with a Bell-type skill set lasting that long, and playing an effective corner outfield (Bell was in left field [a large left field] 1943-46). I think it is not possible that he would have been playing every day, however, as _none_ of the major-league players I looked at (key examples Willie Mays, Ty Cobb, Ted Williams, Sam Rice, Rickey Henderson), were starting more than 120 games in the outfield in any one season after age 40.

All that's to say that unless we have cause to re-evaluate Bell's offensive performance in the mid-40s (and it should receive a war-time discount), I see no reason not to give him credit for those seasons. with playing time appropriately scaled down to reflect the extra rest needed at 40+. (I tried to do that for the MLEs, but it might be that 20 games a season should be shaved off, 1942-46).

I'll be interested to see what his win shares look like, but after thinking through all these matters, I am back to thinking of him as Max Carey-like, with his longer career balancing out his period of Doc-Cramer-like offense early in his career.

Now I just have to figure out where I'd have Max Carey if he were still eligible and run the win shares to see if it bears out my hypothesis . . .

Thanks Chris.

The OPS+ reports are a simple and fun way of taking your great MLE work and presenting it in a manner than mimics the bb-ref read-outs of AL/NL players. But, as you remind us, OPS+ is only the beginning of the story -- defense, base-running & durability are not included.

OPS+ was certainly not why we inducted Max Carey. I wasn't a big Max Carey fan, myself, but he did have strong enough support here to get inducted.

OPS+ is only the beginning of the story.

And it makes for a really great beginning -- I look at the raw MLE stats for the NeL players and have little idea how to evaluate them without a clearly defined context. Running win shares provides the context, but they are very slow to produce. You get us to OPS+ _much_ faster, and they get the story going!

Chris Cobb-

I have no published League-wide data for the Negro Leagues after 1929 and before 1944. Basically from 1929 to 1944 is a statistical wasteland, although Cum Posey usually tried to collect some stuff and get it out there. But usually it was simply of individual players and of highly dubious quality.

By the way, I could not have stated the park effect on Bell's OPS any better. In the 1940s, Bell played for teams with home fields that were very well suited to his talents: Chicago 1942 (Comiskey Park) and Homestead 1943-1946 (Forbes Field in Pittsburgh and Griffith Stadium in Washington).

As for Bell's aging pattern, it has been proven that players with speed age far better than slower players. Bell had extreme, off-the-charts, speed and may have had a unique aging pattern.

it has been proven that players with speed age far better than slower players.

Otis Nixon was never even a particularly good major league ballplayer. His best single-season OPS+ was 94, and his playing time came and went according to the needs of the teams he played for - he seldom showed up to spring training with a guaranteed starting job. But he had a 17-year career, and over his last three years (age 38-39-40 for four different teams) he went 112-26 as a base stealer.

On the subject of base stealing; If Otis Nixon could have 620 lifetime SB, including over 250 after the 1990's HR explosion began, then Bell could have piled up a bundle himself.

Cool Papa Bell Projected Win Shares

I hope these will help to move the study of Bell forward. Working on these win share estimates raised as many questions for me as they answered, so I hope you’ll take a look at the notes following, which explain the estimates and address the questions they raise.

Year Team BWS FWS Total
1924 Stl  7.2 5.5 12.7
1925     13.5 7.8 21.3
1926     16.4 8.2 24.6
1927      8.9 6.5 15.4
1928      9.2 7.0 16.2
1929      6.7 7.0 13.7
1930     12.4 4.5 16.9
1931     15.4 5.7 21.1
1932  *  21.2 5.3 26.5
1933 Pgh 17.2 6.7 23.9
1934     16.5 6.0 22.5
1935     17.1 4.3 21.4
1936     12.8 4.8 17.6

1942 Chi 21.5 3.7 25.2
1943 Home16.7 2.3 19.0
1944     13.0 2.0 15.0
1945      3.5 1.3  4.8
1946      7.8 1.1  8.9
        237  89.7 326.7

My Estimates for 1937-1941
1937 SD  12.5 4.2 19.7
1938 Mex 17.0 4.2 21.2
1939     13.0 4.2 17.2
1940     20.5 4.2 23.6
1941      8.0 4.2 12.2
         71  21   92

* Multiple teams

Notes

1) On the value of stolen bases: there are few seasons during Cool Papa’s career in which a player stole 60-70 bases. It is therefore hard to estimate the contribution of sb to his win shares, if we wish to estimate that he would have been the top base-stealer of his era by a large margin. I have raised his seasonal win shares from 1924-36 by an average of one win share per year to account for his stolen-base advantage over his otherwise most similar batters. This estimate was arrived at by comparing the win shares of top base-stealers 1924-27 to their most comparable players in terms of OBP/SLG and seeing how much they tended to pull ahead. Those who think this estimate is too high or too low should adjust the win share estimates for those seasons accordingly.

2) On total win shares: having looked more closely at Bell’s comparable players on offense throughout his career, I find it unlikely that Bell would have played as many games in the majors as I have projected him for, even if he played as many seasons, which I still consider possible. For my own placement of him, I’ll be including his Mexico years but reducing his total games by 10%. I thought I should show win shares that correspond to the playing time listed in the posted MLEs, however, so I haven’t included those reductions in the posted data.

3) On fielding win shares: I estimated Cool Papa as an A+ outfielder for his career, with a career average of 4.03 ws/1000 innings. This estimate was arrived at by bringing together Lloyd Waner and Doc Cramer as models. Waner covers the 21-34 years, Cramer the 35-43 seasons. I modeled his peak on Little Poison and his decline on Cramer, and this is the result. This puts him towards the middle of career center-fielders with more than 20000 outfield defensive innings. That group:

Ty Cobb, 25834 innings, 3.20 ws/1000
Willie Mays, 25205 innings, 4.11 ws/1000
Tris Speaker, 23880 innings, 4.93 ws/1000
Max Carey, 21683 innings, 4.37 ws/1000
Willie Davis, 202020 innings, 3.88 ws/1000

My original MLEs would give Bell 27492 outfield innings, which is one reason that I consider my playing time estimate overall to be a bit too high. Without lost time to military service, Willie Mays could have reached that total, so it’s not clearly impossible for a major-leaguer to reach it, but I am doubtful.

It’s clear that Bell _should_ be in the group with the longest-career outfielders because he’s exactly the type of player to fit into it. The top 10 outfielders in innings are Ty Cobb, Willie Mays, Hank Aaron, Tris Speaker, Rickey Henderson, Lou Brock, Max Carey, Dave Winfield, Vada Pinson, and Al Kaline. All the players who are the close matches for Bell’s skill-set and demonstrated longevity – Henderson, Brock, Carey – are here. It’s Bell’s place within this group, not that he should be in it in the first place, that needs to be settled.

For the record, Lloyd Waner averaged 4.33 ws/1000 in 15849 defensive innings; Doc Cramer averaged 3.85 ws/1000 in 18806 innings.


4) On Mexican win shares: these are rough estimates, arrived at by looking at his win shares in surrounding seasons and by guessing at the competition levels in Mexico, based on the reductions that would be necessary to bring Bell’s performance there into line with his career totals. I did not attempt a full study, with regression and everything. If my conversion factors for the NeL are correct, I think the conversion factors for Mexico, based on Bell’s evidence, would be between.85/.72 and .87/.75.

I would find more discussion of the Mexican League helpful, and I think, actually, that it’s crucial to Bell’s case. In looking at his career, he looks a lot like Max Carey. One of the features of Carey’s career is that he peaked late, at ages 30-35. A speedster who holds his speed advantage a long time, and gradually develops his “old player” skills of plate discipline and hitting for power would probably peak later. If Bell has a late peak (and the data we have don’t contradict that idea – it looks like 1940, Bell’s age 37 season, was probably his best year as a hitter), some or all of it would fall in his Mexican League years. To place him fairly, an accurate estimate of the height of his peak is crucial, and I don’t feel that I have that at present.

Thanks for the Easter present, Chris!

One question, though: do you have Bell's non-10% reduced projected games for each season that he played in the Mexican League handy? Thanks!

John,

My very rough projections for Bell for 1937-41 are based on 142 games played per season.

My very rough projections for Bell for 1937-41 are based on 142 games played per season.

That will work for me, Chris. BTW, I agree with your reasoning concerning the 10% reduction for those games.

Now to see if he's on my ballot or not.