He's in the Mexican, Cuban, and American Halls of Fame. Will the Hall of Merit be next?

He's Willie Mays, Jackie Robinson, and Roger Clemens all in one. I hope he gets in.

Right now I have Waner #1 for 1950 but he is not an inner circle no-brainer type #1, so there is room for somebody to bump him. And it's not going to be Cronin or any other holdover. I am dying to find out just what Martin Dihigo is really all about. Kind of a Joe Rogan, only better? Or Cristobal Torriente and Bill Foster rolled up into one? Surely not Willie Mays and Jackie or Roger...that is a little hard to take.

So IOW he certainly could be #1, or #2. I will be shocked if he is not at least #3.

OTOH I'm not sure Cronin is #2 among MLers. A HoMer for sure but not clearly the #2 MLer available in 1950. Somewhere in the #2 to #5 pack.

"The greatest player I ever saw was a black man. He's in the Hall of Fame, but not a lot of people have heard of him. His name is Martin Dihigo. He's the only player I ever saw who could play all nine positions, could run, and was a switch hitter. I remember one year in winter ball, I was having a pretty good year myself down there, and they were walking him to get to me."

— HOF first baseman Johnny Mize
(perhaps slightly paraprased, as I don't have the quote right in front of me)

Sunny,

I know you are a compatriot of mine in the Best Freinds of Hughie Jennings club and I must say that Cronin was better the Hughie. The peaks are close with Hughie just ahead by Cronin destroys him in extended prime and career. So please to preach the gospel of Ee-Yah and tell me why you think that Hughie was better than Cronin!

My question will be Jennings or Dihigo for #3. Of course Dihigo could wind up higher than #3, maybe even #1.

Martin Dihigo Data

There's surely better data on Dihigo's Mexican and Cuban play out there, but here's the data from Holway and career data from MacMillan for us to get started with.

Born 1905

Teams

23-37 Cuban Stars (East), 28 Homestead Grays, 29-31 Hilldale Daisies, 32 ??, 33 Venezuela, 34 ??, 35-36 NY Cubans, 37 Santa Domingo, 38-44 Mexico, 45 NY Cubans

Batting and Position Data from Holway

NeL

1922 no ba data; 1b for Eastern Cuban Stars
1923 .242 for Cuban Stars (East); 1b
1924 .239 for Cuban Stars (East; ut & p
1925 .299 for Cuban Stars (East); 2b
9-25 in World Series vs. KC
1926 .327 for Cuban Stars (East); ba 5th, 12 hr (1st), 32 hr/550 (1st); 1b, all-star
1927 .312 for Cuban Stars (East); ss, all-star (at 2b?)
.223 for Homestead Grays; ut (late pickup)
6-16 vs. Major-League competition
1928 .167 for Homestead Grays; 3b & p
3-13 vs. major-league competition
1929 .259 for Hilldale; 18 hr (2nd), 37 hr/550 (2nd); ut & p, all-star (at ss)
1930 .404 for Cuban Stars (East); ba 5th, 7 hr (2nd), 39 hr/550 (4th), 3 3b (4th); 3b, all-star (at 2nd)
1931 .244 for Phi Stars; 5 3b (3rd); lf & p (0-1)
2-8 vs. major-league competition
1932-34 No Data
1935 .335 for NY Cubans; 9 hr (3rd), 30 hr/550 (3rd), 4 3b (4th), 6 sb (5th); ut & p
6-28 in playoff vs. Pittsburgh
1936 .331 for NY Cubans; 13 hr (4th), 60 hr/550 (2nd), 9 2b (1st); ut & p; all-star (as dh)
0-7 vs. major-league competition
1937 No Data
2-6 vs. major-league competition
1938-1944 No Data
1945 .100 for NY Cubans; of & p

Career, according to Holway
609-2034, .299
69 home runs, 18 hr/550 ab
12-49 vs. major-league competition
Mean avg. for 11 seasons with data, .287

Career, according to MacMillan 8th edition
415 g, 1435 ab, 453 hits, 53 2b, 18 3b, 64 hr, 32 sb, .316 ba, .511 sa

Cuba

1922 5-28
1923 no data
1924 15-50
1926 31-75
1927 54-130
1928 46-152
1929 51-180
1930–1934 No Data
1935 63-176 (batting title),
1936 74-229
1937 50-165
1938 25-99
1939 23-79
1940 20-110
1941 28-123
1942-44 No Data
1945 16-71

career, 501-1595, .314

Other Latin Play

1933 Venezuela (also possibly in 1932 and 1934)

1937 In Santa Domingo, 34-97
1938 In Mexico
1939 In Mexico
1940 In Mexico, 110-302, .364 (2nd), 9 hr (4th), 16 hr/550 (5th)
1941 In Mexico, 102-329, .319
1942 In Mexico, 89-279, .319
1943 In Mexico
1944 in Mexico

Pitching Data from Holway

NeL

1924 1-4
1925 no data
1926 2-2
1927 2-0
1928 1-1
1929 5-3
1935 7-3 (on pitching leaderboards)
1936 7-4 (on pitching leaderboards)
1-0 vs. major-league competition
1945 0-1

Career, 25-18
Holway gives his career record as 29-26
MacMillan 8th gives it as 27-21

Cuba

1924 2-3
1927 4-2
1928 2-1
1935 11-2 (led in wins)
1936 14-10
1937 11-5
1938 14-2 (led in wins)
1939 6-4
1940 8-3
1941 8-3
1942 4-8
1944 3-3
1945 5-4

Career, 78-50


Other Latin Pitching

1934 6-4 in Santa Domingo
1940 8-0 in Mexico
1941 9-10 in Mexico
1942 22-7 in Mexico, led league in ERA and Ks


Career, 45-21

Another where the rep may have outrun the reality. He will get a huge number of translated WS, because of having hit AND pitched, but may not have been that great at either. Fascinatedly awaiting Chris's conversions, which will be very difficult and controversial. Fortunately for us all, after this guy I think it gets a little easier. Even I agree on Josh Gibson!

1928 Martin Dihigo
Homestead Grays

Batting
G-19 (team 19)
AB-67
H-23
D-1
T-1
HR-4
R-15
W-8
HP-0
SF-1
SH-0
SB-0
AVE-.343 (NeL east .282)
OBA-.408 (.333)
SLG-.567 (.384)

Pitching
W-1
L-1
TRA-7.41
G-2
GS-2
CG-2
IP-17
H-27
HR-4
R-14
W-4
K-14
HB-0

Fielding-ss
G-9
DI-60
PO-16
A-14
E-1
DP-1
RF-4.50 (team 5.13; NeL east 5.07)
FPCT-.968 (team .938; NeL east .917)

Fielding-2b
G-5
DI-41
PO-7
A-13
E-1
DP-1
RF-4.39 (team 4.50; NeL east 5.39)
FPCT-.952 (team .952; NeL east .949)

Fielding-3b
G-4
DI-31
PO-1
A-6
E-2
DP-0
RF-2.03 (team 1.82; NeL east 2.73)
FPCT-.778 (team .842; NeL east .919)

Dihigo also played two games at pitcher (1 putout and 1 error in 17 innings) and one game at first base (12 putouts, 5 assists in 9 innings).

He will get a huge number of translated WS, because of having hit AND pitched, but may not have been that great at either.

That is a possibility, karlmagnus, though I have a feeling he will be at least Bullet Joe Rogan. Chris' numbers should point us in the right direction.

Another where the rep may have outrun the reality. He will get a huge number of translated WS, because of having hit AND pitched, but may not have been that great at either. Fascinatedly awaiting Chris's conversions, which will be very difficult and controversial

I have/am considering not attempting MLEs for Dihigo at all. The problems are many.

1) His career in the NeL is fragmentary, so defining a proper career shape and running proper regressions will be difficult.

2) Much of his play was in Cuba and Mexico, and I don' t have reliable conversion factors for those leagues.

3) His career was more profoundly shaped by the circumstances of Negro and Latin baseball than any other player we have considered, even Joe Rogan. There's no way he would have played nine positions in the majors, or gradually transitioned to full-time pitching around the age of 30. Clearly, all of this made sense for the leagues in which he was playing, but something else would have happened in the majors.

Will voters have to actually judge a Negro League player based on what he did for his teams rather than imagining him in the majors? Perish the thought!

I have/am considering not attempting MLEs for Dihigo at all.

I hope you make the attempt, Chris, nevertheless. It may not have the same weight as with your other MLE's, but it still may help us sort him out on our lists.

Will voters have to actually judge a Negro League player based on what he did for his teams rather than imagining him in the majors? Perish the thought!

We did it quite a few times with Frank Grant and others, David, so your sarcasm is not welcome.

I believe that we went on NeL play for Joe Rogan since it is highly unlikely he would have played both P and OF in the Majors. I dont' believe he was HOM material at either, but combined he was a first ballot guy (in a weak year if I recall).

Back to Dihigo, I believe he was better than Rogan and there is a very good chance he will beat out Jennings for my #3 spot.

David,

YOu must see the advantages of MLE's in ranking NeL players. If we are to pit them against MLB players of the same era it is nice to know how they would have performed in the same league. I also am not convinced that we have really distorted what any NeL player did in his league through MLE's that much. Maybe we gave Beckwith and Wilson a little less defensive value, but if we looked at the numbers stright up, I believe we would be overrating NeL players. No matter how godo we think the stars of the NeL were, I don't think any of us thin the NeL was as strong as either the NL or the AL.

At the very least (i.e. if you don't think Martin would have been a double duty player) could Dihigo have been a better hitting version of Wes Ferrell with more career and less peak?

First, as sarcasm goes, that was pretty mild.

Second, John is right, we have a heck of a lot more points of intersection of Dihigo with more or less known quantities than we did Grant or Bill Monroe or Rube Foster.

But third, yes, Dihigo is one of the tougher cases in a long while.

But back to David's point. I have said before that I think our job is to pick the 220 players who had the greatest value and impact *in reality,* not translated into some alternative universe or board game. So for me the questions re. Dihigo will be:

1. How much value did he have for his NeL and Cuban and Mexican League teams compared to the current alternatives--Mackey, Taylor, Monroe, Moore, Lundy, Beckwith, Suttles, Poles, Redding, Mendez, et al? If we can get the answer to that question right, we are more than halfway home.

2. But then, of course, we still do have to figure out how he (and they) compare with the best MLers. And if we feel like we can calibrate Suttles and Beckwith to the MLers, and Dihigo to Suttles and Beckwith, then we can place Dihigo among the MLers.

There will be more uncertainty in this case than in others, but I am not uncomfortable giving this the thought it deserves and then taking a stand.

This is not to say I have made up my mind, I have not. So, it might also be noted that he appears to be no more than the 10th to 15th best hitter available (not as good as Taylor who has never been on my ballot) and probably no better than 8th or 10th among the pitchers (certainly no Redding or Mendez, and perhaps not even John Donaldson). The trick, however, is to add them together. But even there, it is not clear he is as good as Joe Rogan in either area.

Think Bob Caruthers, I guess.

In summary: To me it is not entirely germane whether Dihigo would have made the MLs as a hitter or not, or as a pitcher or not. Clearly he had a lot of value for NeL teams, and that is what counts.

jschmeagol,

I think your postulate about dihigo being a ferrell with more career and less peak would be fraught with the same kinds of trouble that giving pitchers war credit might be fraught with. Namely that by not pitching as much early in his career, his arm might have avoided the kind of workload damage that did in Ferrell and so many others.

Of course, it's all speculation anyway, so I'll ask a similar question.

Were Dihigo in MLB, what position would he likely have come up at? I assume OF, but I don't know which OF spot, nor whether he might have come up as an infielder to maximize his value to his team.

Martin Dihigo was born in........
Ah, to hell with it.

I'm going to disagree with David (and I'm sure the majority of the electorate). Translations to me are all about how a player would have played in the majors. If that were not the case, then in cases like Dihigo, we would be disadvantaging white players, who certainly had no such opportunity to accrue double duty value.

Let's imagine two players, with the same number of PAs, with similar defensive values and similar career arcs:

- Player A (Caucasian) has a 115 OPS+
- Player B (African-American) has, as best we can tell, a translated 115 OPS+. He also pitched for about half his career at an estimated 100 ERA+ rate.

I am sympathetic to the argument that Player B had more value to his teams than Player A, but in the HOM case, I fail to see how it is fair to Player A to rank Player B ahead when it is plainly clear that Player A had *zero* opportunity to pitch.

If we are not translating performance to the major leagues, then we are clearly giving an unfair bonus to those who played in inferior leagues, and thus had the opportunity to perform what in a better league would be rare or simply non-existant feats.

One final point of disagreement about David's statement:

Will voters have to actually judge a Negro League player based on what he did for his teams rather than imagining him in the majors? Perish the thought!

If we're not imagining players in the majors, then how the hell are we supposed to compare Negro League and Major League players at all? The only way this methodology would work is if we set aside a set number of NL players so they were only being compared to each other. Pass.

---------------------------------------

One final note: Despite my above statements, I did give Rogan credit for being a pitcher and hitter. I did this because I felt he was a sufficiently unqiue case that it could have happened in the majors.

He was above average in both hitting and pitching: Had he been average in either, I suspect the lesser would have been tossed aside. However, he was not a superstar in either. Had he been a superstar calibre pitcher or hitter, I suspect he would have been instructed to concentrate on that. As a 115 ERA+/OPS+ type guy, he was right in the middle where I can picture a team seeing his greatest value as sticking with both.

Whether Dihigo fits this description is something I'm excited to see. Mendez, as an example, is getting no credit from me for his hitting rebirth (except in so far as he gets credit for being a strong hitting pitcher).

Here's Dihigo's Mexican League batting:
Yr--G--AB---H--D-T-HR--R--BI-BB-SO-SH-HP-SB-AVE-OBA-SLG
37-07-026-010-01-2-01-07-12-03-00-00-00-00-357-448-643
38-42-142-055-08-2-06-37-27-26-04-01-00-09-387-482-599
39-51-187-063-11-3-05-32-31-27-13-00-03-05-337-429-508
40-78-302-110-17-6-09-60-73-17-25-03-02-09-364-402-550
41-92-329-102-25-4-12-74-59-57-53-02-00-07-310-412-520
42-85-279-089-12-4-08-56-51-74-21-00-03-09-319-466-477
43-75-238-066-14-3-07-35-50-51-24-01-01-05-277-407-450
44-60-189-047-10-2-04-34-29-33-17-00-03-05-249-369-386
46-66-177-056-09-2-03-27-32-32-21-00-02-08-316-429-441
47-20-046-009-03-1-00-04-06-07-06-00-00-00-196-302-304
(He also played one game in the field in 1950, without coming to bat.)

Career totals:
G-577
AB-1917
H-607
D-110
T-29
HR-55
R-366
RBI-370
BB-327
SO-185
SH-7
HP-14
SB-57
AVE-317
OBA-420
SLG-490

Unfortunately, the only fielding position the MeL encyclopedia gives is pitcher, though he was clearly playing many more games than he pitched.

Dihigo's Mexican League pitching:
Yr--W--L--ERA--G-CG---IP---H--BB--SO-HB-WP
37-04-00-0.93-05-04-038.7-022-03-008-03-00
38-18-02-0.92-22-16-167.0-104-32-184-04-03
39-15-08-2.90-23-20-202.0-169-42-202-09-01
40-08-06-3.54-17-08-109.3-106-48-065-00-00
41-09-10-4.01-23-09-157.0-177-43-093-00-00
42-22-07-2.53-35-26-245.3-244-77-211-00-02
43-16-08-3.10-26-18-194.3-181-64-134-02-05
44-12-10-3.14-31-18-212.3-207-88-090-01-02
46-11-04-2.83-20-11-140.0-134-49-063-02-09
47-04-02-4.37-10-03-055.7-077-13-016-00-01
50-00-00-0.00-01-00-002.0-001-01-000-00-00

Career totals:
W-119
L-57
PCT-.676
G-213
CG-133
SHO-19
IP-1523.7
H-1422
BB-465
SO-1109
HB-21
WP-23

Unfortunately, the MeL didn't count games started (until the last couple of seasons of his career), runs allowed (as opposed to earned runs), saves, or home runs allowed.

Sorry--his career MeL ERA was 2.84.

Dihigo seems to ahve had a decent bit of power in his earlier MeL years and Gary's stats show that he knew how to take a walk.

Dr.,

'The longer career but less peak' part of my above post was thrown in their at the last minute so that I didnt' get people saying that Dihigo wasn't the player that Ferrell was at his peak.

I guess my point was that IF he had been a pitcher in MLB would Ferrell have been a decent comp? Both were good pitchers and both could hit. If he is a decent comp for Ferrell then it looks like Dihigo would be an eventual HOMer in my book, though he may have to wait a while.

Right now it doesn't seem like Dihigo either pitched well enough to hit well enough to be an obivous HOMer. But he did both. 'Could he have done both in the Majors?' is a question that I think needs to be answered. If not I think we should give him the most credit possible (since he did do both) and I think that a decent starting pitcher (4.00-4.15 DERA) with a 110 OPS+ is more valuable than an OFer with a 115-120 OPS+.

The ERA+ and OPS+ estimates by Chris will be crucial. 100 ERA+/100 OPS+ doesn't do it; 110 ERA+/115 OPS+ most certainly does.

Martin Dihigo's Cuban League batting (the second 26/27 season was in a rival league):
Yr-----AB--H--D--T-HR--R-SB-AVE-SLG-POS
22/23-028-05-00-0--0-03-00-179-179-INF
23/24-002-00-00-0--0-00-00-000-000-INF
24/25-050-15-03-3--1-12-02-300-540-OF/P
25/26-032-11-03-2--1-11-00-344-656-INF/P
26/27-075-31-04-1--3-20-06-413-613-3B/P
26/27-020-09---------------450-----INF/P
27/28-130-54-12-3--2-32-05-415-600-OF/1B/P
28/29-152-46-14-4--2-29-04-303-487-OF/1B/P
29/30-180-51-06-6--0-23----283-383-OF/INF/P
30----054-18-02-1--0-08-04-333-407-OF/P
31/32-010-03-02-0--0-02-00-300-500-OF
35/36-176-63-08-8--0-42----358-494-OF/P
36/37-229-74-10-2--4-38-10-323-437-OF/P/INF
37/38-165-50-05-4--0-21-06-303-382-1B/P/OF
38/39-145-37-06-1--1-24-05-255-331-P/OF
39/40-079-23-02-1--1-10-03-291-405-P/INF
40/41-110-20-05-2--0-16-04-182-264-P/OF
41/42-123-28-06-0--1-19-01-228-301-OF/P
42/43-135-36-06-1--1-14-02-267-348-OF/P
43/44-087-22-01-2--0-11-00-253-310-P/INF
44/45-029-06-01-0--0-03-00-207-241-P/INF
46/47-011-01-00-0--0-01-00-091-091-P/INF

Career totals:
AB-2022
H-603
D-96
T-42
HR-17
R-347
SB-52
AVE-298
SLG-412

Martin Dihigo's Cuban League pitching:
Yr-----W-L, G-CG
24/25 2-3, 20-1
25/26 0-0, 1-0
26/27 2-0, 2-1
26/27 1-0, 2-0
27/28 4-2, 6-5
28/29 2-1, 5-2
29/30 1-2, 3-2
30 2-0, 4-2
31/32 DID NOT PITCH
35/36 11-2, 18-13
36/37 14-10, 30-22
37/38 11-5, 20-12
38/39 14-2, 21-14
39/40 6-4, 19-9
40/41 8-3, 13-10
41/42 8-3, 17-11
42/43 4-8, 14-7
43/44 8-1, 15-4
44/45 3-3, 13-2
46/47 1-3, 8-0

Career totals
W-102
L-52
PCT-.662
G-231
CG-117

Pitching stats are quite skeletal. Also, the career totals at the end of the Cuban Baseball book show a 107-56 record (he's the career leader in wins). Somewhere I missed 5 wins and 4 losses, but haven't been able to figure out where.

Dreaming about Mexico...

I was thinking about the conversion rates for Mexico in the shower this morning (too much info, I know), and it reminded me of a recent piece by Clay Davenport on the current Cuban leagues. CD said the Cuban leagues are roughly equivalent to A-ball leagues in the states.

Which got me thinking about Mexico back in the day. As a summer league, it attracted quite a bit of top-flight talent in the 1930s, including Dihigo, Serrell, CP Bell, Wells, and many others.

In fact, just skimming A-G plus selected well-known names in Riley right now looking for names and years I find this list of players who went to Mexico in the period 1939-1945.

1938: CP Bell, Bragana, Brewer, Paige
1939: CP Bell, Bragana, Brewer, B Brown, R Brown
1940: S Bankhead, Pepper Basset, CP Bell, Bragana, B Brown, W Brown, T Christopher, Dandrige, L Davenport, L Day, Dihigo, J Gaines, J Gibson, B Griffith, Matlock, W Wells, Strong
1941: S Bankhead, CP Bell,Bragana, B Brown, S Carter, B Clarkson, J Gibson, A Crespo, W Wells, Dandrige, Dihigo, S Garcia, Matlock
1942: Bragana, A Crespo, Dandrige, Dihigo, S Garcia, Irvin
1943: Bragana, Butts, Campanella, A Crespo, Dandrige, Dihigo, S Garcia, W Wells
1944: Bragana, Brewer, A Crespo, Dandrige, Dihigo, S Garcia, W Wells
1945: Bragana, B Brown, A Crespo, Dandrige, L Davenport, Dihigo, S Garcia, B Serrell

That's a pretty substantial cluster of top-flight talent. I don't know how many teams and players were in Mexico, but injecting 15 or more NgL all-stars or near all-stars into a league will do a lot for its quality. To give a sense of what that might be like today, imagine if the AA Texas League went independent, kept all of its players, trimmed rosters down to about 16-20 per team, and signed 15 to 30 MLB All-Stars. The league's roughly 10 teams in size, and its total number of players would now be 150-300 players. Its overall quality would rise substantially because about ten percent of its worst players would be replaced by the best ten percent of its players AND the new All-Stars would begin taking at bats away from the least effective starting players.

Again this is incomplete information, just eight letters of the alphabet a few other good players, so you can imagine that the talent moving south of the border for better pay and treatment was substantial.

If it would help, I could comb through Riley and report back on all the guys I can find who were down there in this period. Perhaps that would help begin the process of converting them? I figure if we can find stats for the seasons in quesiton, we could first figure out the strength of Mexico relative to the NgLs by comparing players who went between them, then convert players into the NgL environment, then finally convert them to MLB using our current conversion rates.

Let me know if that would be helpful and I'll put it into a spreadsheet for anyone who wants to undertake the comparison study. Might take me a couple days.

Chris, if you don't plan to do a full conversion for Dihigo (and I can understand why not) it would at least be helpful to get ballpark OPS+ and ERA+ equivalents for his NEL years. This would give us a good idea as to his prime, and whether he was truly elite status at his best as either a hitter or a pitcher.

I'll do some seasonal conversions for his hitting at least.

I'm not sure what I can do on his pitching -- it depends on how much team data is available. A decent assessment of pitching really depends on having a decent amount of team data -- a long enough season to make the team w-l record and the batting records meaningful. Dihigo did his main pitching in the NeL during the years for which data is terribly scanty.

My view of the matter is that Dihigo would never have been a pitcher in the majors, not because he lacked the talent to succeed (my rough estimate is that he would have been an average major-league pitcher) but because he was better as a hitter than he was as a pitcher and so would have been asked to specialize in that area.

In leagues where versatility was highly valuable (and was respected perhaps even beyond what it was worth) Dihigo's decision to pitch full-time and play the field the rest of the time made sense, and he must have been immensely valuable in that context: it was the right choice for the conditions in which he was playing.

But in the majors, it would have made sense for him to specialize. If he had specialized, it seems highly likely to me that he would have become a better hitter and fielder than he did in the NeL.

My concern with publishing MLEs at all is that, by putting stats generated in a context where the value of versality is high into a context in which the value of specialized, peak talent is high, I will be leading to a misassessment of Dihigo's merit.

I really don't know what the numbers will show, so maybe he'll look great regardless. But I guess I want to make sure that a thorough discussion of the context issue takes place prior to any assessment of MLE-type numbers.

Dr. Chaleeko,

A full listing of the NeL players who went to Mexico would be highly illuminating.

Even if it didn't get us to firm conversion factor for Mexico, it would give us a much clearer view of playing conditions both there and in the NeL.

It looks from the lists you've just generated, and from Dihigo's stats, that the quality of play jumped considerably between 1938 and 1940. Dihigo totally overmatched the league in 1938 (this is Jim McCormick in the Union Association type stuff), but his opponents are quite competitive in 1940-41. If your prelim data is representative, there were fewer NeL stars in 1942 and after, and Dihigo's mastery of the league rises again, at least for 1942 -- after that we'd expect age-related decline.

I really don't know what the numbers will show, so maybe he'll look great regardless. But I guess I want to make sure that a thorough discussion of the context issue takes place prior to any assessment of MLE-type numbers.

That's reasonable, Chris. It appears that your conversions will only take us so far with Dihigo, so less reliance will be given to them (not your fault, of course) than in the past with your other conversions.

In leagues where versatility was highly valuable (and was respected perhaps even beyond what it was worth) Dihigo's decision to pitch full-time and play the field the rest of the time made sense, and he must have been immensely valuable in that context: it was the right choice for the conditions in which he was playing.

Well stated Chris. I think is the major selling point for Dihigo. In his place and time he was immensely valuable. The leagues/teams he was with had things going against them. Instability meant that teams needed to be a draw - whether playing league games or barnstorming. You don't get some people to attend the games you go under. Dihigo helped in that he left you having to carry one less player - he could sub in and pitch or field. And you don't think there was a certain draw to a player who could throw a CG one day, hit a homer the next?

I'm a big fan of his so take my writings with a grain of salt. But he is hurt most by converting to a ML environment.

Here is some abridged info on Dihigo (the source is an expert):

He was born in 1906, not 1905 (his birth certificate is in his Cuban bio).

Up until 1930, He was on track to be perhaps greatest right-handed hitter of all time. But he evidently got bored with being so good and took up pitching and switch-hitting in 1931. In his defense, pitching, especially in the Latin American countries was a much higher paid position than hitter. Dihigo had a hell of an arm.

From 1932 to 1934, Dihigo was the star pitcher on a team sponsored by the dictator of Venezuela's son (One of his team mates was Tetelo Vargas, the best Dominican of that time). He wasn't the best hitter in the country, at first (Alejandro Oms and Vargas were); but he sure as hell was the best pitcher (number 2 was Cocaina Garcia, a forgotten but incredible, short and stocky, Cuban pitcher).

But, by 1934, he had evidently mastered switch-hitting and started to really hit again. But he never again showed the power that he had had previously. His hitting skills faded quickly after that as he concentrated on pitching and managing.

The Concordia Eagles, Dihigo's team, traveled to Cuba and played against Cuban All-Star teams. They imported Mize for one of the series. When Mize had to report to spring training, they replaced him with Josh Gibson.

In 1935 and 1936, he was a great pitcher and hitter for the New York Cubans. In 1937, Dihigo hit .351 in Trujillo's tournament (second best hitter to Gibson) and was the second best pitcher to Paige.

From 1938 to 1947, he played in Mexico.

***

To me Dihigo is somewhere between Carl Mays and Babe Ruth. I'll probably have him between 7 and 15 on my ballot.

Thanks, Daryn, and please thank the expert for sharing his expertise with us.

In evaluating Dihigo's batting stats from the 20s, the fact that he put them up before the age of 25 adds to their impressiveness.

But in the majors, it would have made sense for him to specialize. If he had specialized, it seems highly likely to me that he would have become a better hitter and fielder than he did in the NeL.

If further looks show your conclusion to be correct (that he would have specialized as a hitter), I completely agree with this statement. I should have added that to the above: If I'm taking a two-way NLer and converting him to a one-way majors, I need to boost his stats for that one way, if only a little, to account for the benefits of specialization.

If further looks show your conclusion to be correct (that he would have specialized as a hitter), I completely agree with this statement. I should have added that to the above: If I'm taking a two-way NLer and converting him to a one-way majors, I need to boost his stats for that one way, if only a little, to account for the benefits of specialization.

I agree, Michael.

Even I agree, provided it's only a little :-))

Chris,

You'll have the Mexico info soon, though the way things are going this week, it may take a few days.

You'll have the Mexico info soon, though the way things are going this week, it may take a few days.

Well, you're not getting paid until you deliver, mister! :-)

Quick link to an article about Negro Leaguers in Mexico that offers some nice background information on how and why the Mexican League was taking on so many players.

The article suggests that 30 players may have been down there in any given year, and that they dominated all the leaderboards and record books.

From the things that Gary A. has said about the Cuban Stars teams in the NeL, it seems that they were generally traveling clubs. Is their status in that regard known for the seasons in which Dihigo played for them? Gary, Gadfly?

If so, we can infer with some confidence a park factor of 100 for Dihigo in those seasons.

That would be nice.

I'm going to get to work on batting MLEs for Dihigo for his NeL seasons: they should be ready late Thursday or early Friday. I'll post playing time estimates for discussion asap.

I should note that the discussion has persuaded me that the MLEs will help us look at the shape of Dihigo's career as a hitter, and I feel like the issue of context has been sufficiently registered that the MLEs won't distort our view of him any more than they distort our view of Negro-Leaguers in general.

I'll report on whether I think meaningful pitching MLEs are possible as soon as I've reviewed the team data more carefully.

I know for sure that the Cuban Stars (East) were a travelling club in 1928. I believe there were some years in which they settled down in a home park, but I don't know which ones (Gadfly may know). By the 1940s the NY Cubans were using the Polo Grounds, but this was after Dihigo's time in North America.

Good news! The workflow waterfall I'm about to be crushed by ebbed for one evening, and I had a big chunk of unplanned time to devote to Mexico.

I just plowed through Riley noting each player he lists as having played in Mexico. I kept track through 1949 only, though many players went to Mexico in the early 1950s as the Negro Leagues were dissolving. In a couple cases, it's clear that the player was a Mexican man who only played a little in America's Negro Leagues.

Anyhow, in in the next post I'm just going to list them all by year in alphabetical order.

I would like to make one other quick note, and that is that in 1937 and only 1937, it appears that a Venezulan league played roughly the same role as the Mexican League did from 1938 through the mid 40s. I didn't take notes on which players were in that league, but I would estimate that 10-15 players spent the 1937 season there.

NEGROL LEAGUE PLAYERS
PLAYING IN MEXICO, 1938-1949
1938 (7)
Bell, Cool Papa
Bragana, Ramon
Brewer, Chet
Dials, Lou
Paige, Satchell
Salazar, Lazaro
Samplson, Ormond

1939 (14)
Bell, Cool Papa
Bragana, Ramon
Bremer, Eugene
Brewer, Chet
Brown, Barney
Brown, Ray
Dials, Lou
Porter, Andy
Salazar, Lazaro
Samplson, Ormond
Smith, Gene
Taylor, Johnny
Troupe, Quincy
Young, Tom

1940 (38)
Bankhead, Sam
Bassett, Pepper
Bell, Cool Papa
Bragana, Ramon
Brown, Barney
Brown, Willard
Christopher, Thadist
Dandridge, Ray
Davenport, Lloyd
Day, Leon
Dials, Lou
Dihigo, Martin
Direaux, Jimmy
Fernandez, Rudolfo
Gaines, Jonas
Gibson, Josh
Glover, Lefty
Griffith, Bob
Harvey, Bill
Hunter, Bert
Hyde, Bubba
Jefferson, Willie
Matlock, LeRoy
McDuffie, Terris
Pages, Pedro
Patterson, Pat
Porter, Andy
Radcliffe, Double Duty
Roberts, Specs
Ruffin, Leon
Strong, Ted
Taylor, Johnny
Troupe, Quincy
Wells, Willie
Welmaker, Roy
Williams, Chester
Wilson, Dan
Wright, Bill

1941 (37)
Bankhead, Sam
Barbee, Bud
Bell, Cool Papa
Bragana, Ramon
Brown, Barney
Carter, Spoon
Clarkson, Bus
Crespo, Alenjandro
Dandridge, Ray
Dials, Lou
Dihigo, Martin
Direaux, Jimmy
Fernandez, Rudolfo
Garcia, Cocaina
Garcia, Silvio
Gibson, Josh
Glover, Lefty
Harvey, Bill
Hughes, Sammy
Hunter, Bert
Jefferson, Willie
Matlock, LeRoy
McDuffie, Terris
Morney, LeRoy
Morris, Barney
Pages, Pedro
Porter, Andy
Ruffin, Leon
Salazar, Lazaro
Stone, Ed
Taylor, Johnny
Troupe, Quincy
Wells, Willie
Welmaker, Roy
Williams, Chester
Williams, Harry
Wright, Bill

1942 (15)
Bragana, Ramon
Crespo, Alenjandro
Dandridge, Ray
Dihigo, Martin
Direaux, Jimmy
Fernandez, Rudolfo
Garcia, Cocaina
Garcia, Silvio
Heredia, Ramon
Hunter, Bert
Irvin, Monte
Pages, Pedro
Porter, Andy
Salazar, Lazaro
Troupe, Quincy

1943 (20)
Blanco, Heberto
Bragana, Ramon
Butts, Pee Wee
Campanella, Roy
Crespo, Alenjandro
Dandridge, Ray
Dihigo, Martin
Direaux, Jimmy
Garcia, Cocaina
Garcia, Silvio
Green, Chin
Heredia, Ramon
Hunter, Bert
McDuffie, Terris
Pages, Pedro
Porter, Andy
Salazar, Lazaro
Troupe, Quincy
Wells, Willie
Wright, Bill

1944 (18)
Blanco, Heberto
Bragana, Ramon
Brewer, Chet
Canizares, Avelino
Crespo, Alenjandro
Dihigo, Martin
Formenthal, Pedro
Garcia, Cocaina
Garcia, Silvio
Green, Chin
Heredia, Ramon
Hunter, Bert
Pages, Pedro
Salazar, Lazaro
Smith, Theolic
Troupe, Quincy
Wells, Willie
Wright, Bill

1945 (22)
Blanco, Heberto
Bragana, Ramon
Brown, Barney
Crespo, Alenjandro
Dandridge, Ray
Duany, Claro
Formenthal, Pedro
Garcia, Cocaina
Garcia, Silvio
Green, Chin
Heredia, Ramon
Hooker, Lennie
Hutchison, Willie
McDuffie, Terris
Pages, Pedro
Rodriguez, Hector
Salazar, Lazaro
Serrell, Bonnie
Smith, Quincy
Smith, Theolic
Taylor, Johnny
Williams, Marvin

1946 (25)
Blanco, Heberto
Bragana, Ramon
Brown, Ray
Canizares, Avelino
Clarkson, Bus
Dandridge, Ray
Douglass, Jesse
Duany, Claro
Formenthal, Pedro
Garcia, Cocaina
Gulley, Nap
Kaiser, Cecil
McDuffie, Terris
Moreland, Nate
Pages, Pedro
Pennington, Art
Rodriguez, Hector
Salazar, Lazaro
Serrell, Bonnie
Smith, Theolic
Summers, Lonnie
Taylor, Johnny
Thompson, Hank
Williams, Jesse
Wright, Bill

1947 (18)
Bragana, Ramon
Brown, Ray
Canizares, Avelino
Carter, Spoon
Clarkson, Bus
Crespo, Alenjandro
Dandridge, Ray
Day, Leon
Douglass, Jesse
Garcia, Cocaina
Gulley, Nap
Pennington, Art
Ruffin, Leon
Salazar, Lazaro
Serrell, Bonnie
Smith, Theolic
Summers, Lonnie
Wright, Bill

1948 (19)
Blanco, Heberto
Bostock, Lyman Sr.
Bragana, Ramon
Brown, Ray
Canizares, Avelino
Crespo, Alenjandro
Dandridge, Ray
Day, Leon
Douglass, Jesse
Garcia, Cocaina
Garcia, Silvio
Gulley, Nap
Pennington, Art
Ruffin, Leon
Salazar, Lazaro
Serrell, Bonnie
Smith, Theolic
Williams, Marvin
Wright, Bill

1949 (10)
Bragana, Ramon
Brown, Ray
Crespo, Alenjandro
Garcia, Cocaina
Gulley, Nap
Pennington, Art
Salazar, Lazaro
Summers, Lonnie
Williams, Marvin
Wright, Bill

There's quite a lot of possible data points if someone is willing to do a study to figure out what the league quality of Mexico was at that time.

There's just a ton of talent moving over the border during this period. I guess the question it leaves me with is not only how strong did this make Mexico, but how weak did it leave the NgLs? What does it mean to be x number of games over replacement when there's an All-Star team's worth of guys playing in Central America?

And given the huge number of players south of the border in 1940 and 1941, it seems like we should be very very leery of big seasons put up stateside in those years by NgL players.

e.g by Monte Irvin!

The large (though not huge) number of American black and Cuban players in Mexico from 1944-1949, is also good evidence that the quality of play in the Negro Leagues was probably higher prior to 1937 (in addition to Venezuela, Santa Domingo drew a number of players out of the league that year) than at any time after that.

The conversions factors I have calculated are based on 1944-49, so it is probable that they underrate to some extent pre-1937 seasons. Gadfly has argued for this point for some time. While I have agreed with that point in principle, I was doubtful that the talent drain was large enough to matter after WWII. Dr. Chaleeko's lists show that it continued to be quite significant. Conversely, the conversion factors probably overrates the 1940-41 seasons, when the talent pool of the NeL was most heavily tapped by the Mexican League.

I don't know if there's any way to get a precise estimate of how much difference the presence or absence of, say, 5 stars and 5 regulars makes terms of league quality?

jimd, you've looked into the effects of adding and subtracting star players to help explain WARP's league strength assessments -- is there a theoretical model we could look at for this?

One thing about Dihigo: his career is actually more extensively documented than any other NeL career to this point. Counting up everything from the NeL and the Cuban, Mexican, and Dominican leagues (as well as the 49 at bats Holway gives him against major leaguers), we have 6,119 at bats (1,865 hits for a .305 average, and 142 home runs) as well as a 261-143 record as a pitcher (.646). And we don't even have any information from Venezuela, where he played when he was 26-28 years old.

I found the career statistics for Bill Wright and Bus Clarkson in Minor League Baseball Stars Vol III. I don't know if these additional data points will help with figuring out the appropriate levels of adjustment between leagues.

Bill Wright, born 1914, 6'5" 230, switch hit, OF
year gms abs rn ht 2b 3b hr rbi sb avg slg.
1933 021 078 xx 19 02 2 00 xx 00 .244 .321
1934 012 050 xx 06 00 1 03 xx 02 .120 .340
1935 021 082 xx 20 01 0 02 xx 04 .244 .329
1936 021 074 xx 25 02 5 01 xx 02 .338 .541
1937 031 100 xx 41 04 2 07 xx 00 .410 .700
1938 022 081 xx 19 02 0 00 xx 03 .235 .259
1939 027 099 xx 40 01 3 03 xx 02 .404 .566
1940 087 350 94 126 30 10 8 67 29 .360 .571
1941 100 387 98 151 25 9 17 85 26 .390 .633
1942 044 163 xx 53 08 2 01 xx 02 .325 .417
1943 088 352 65 129 25 5 13 70 21 .366 .577
1944 087 334 59 112 24 7 10 60 14 .335 .539
1945 044 165 xx 62 12 5 03 xx 05 .376 .564
1946 085 316 47 95 11 8 05 52 17 .301 .434
1947 079 249 36 76 10 4 03 38 13 .305 .414
1948 066 258 47 86 16 3 00 32 12 .333 .419
1949 074 293 46 81 14 3 07 43 03 .276 .416
1950 063 248 75 73 10 1 02 31 05 .300 .367
1951 030 104 23 38 10 2 02 25 04 .365 .558
1952-54 no listing
1955 068 250 49 75 12 3 03 40 06 .300 .408
1956 079 297 48 102 8 3 08 45 07 .343 .471

Listed as leading league in triples in 1936 and 1939, batting average in 1937 and 1941.

Teams and league:
1933 - 34 Nashville, Negro National
1935 Columbus, Negro National
1936 - 37 Washington, Negro Nat'l
1938 - 39 Baltimore, Negro Nat'l
1940 Santa Rosa, Mex
1941 Mexico City, Mex
1942 Baltimore, Negro Nat'l
1943 - 44 Mexico City, Mex
1945 Baltimore, Negro Nat'l
1946 - 47 Mexico City, Mex
1948 Monterrey, Mex
1949 Torreon and Mexico City, Mex
1950 Mexico City, Mex
1951 North Laredo and Mexico City, Mex
1955 - 56 Aguascaliente, Central Mexico League (minors?)

Bus Clarkson, 5'11" 200, SS-3B
year gms abs rn hts 2b 3b hr rbi sb avg slg.
1937 no record listed
1938 no record listed
1939 005 019 xx 008 xx x xx xx xx .421 xxx
1940 011 045 xx 018 01 0 05 xx xx .400 .756
1940 019 080 12 027 04 3 01 13 01 .338 .500
1941 082 326 67 109 23 3 19 83 07 .334 .598
1942 002 007 xx 002 02 0 00 xx xx .286 .573
1943 - 45 World War II (I assume most of 1942 also)
1946 037 131 29 039 08 1 09 32 07 .298 .580
1946 038 146 26 045 07 2 02 34 xx .308 .425
1947 112 390 75 118 19 7 17 68 20 .303 .518
1948 080 276 93 110 12 0 28 68 xx .399 .746
1949 056 192 32 060 xx x xx xx xx .313
1950 033 108 21 032 10 1 04 18 xx .296 .519
1950 059 205 34 062 11 1 07 33 00 .302 .468
1951 097 283 52 097 12 4 05 49 06 .343 .466
1952 014 025 03 005 00 0 00 01 00 .200 .320
1952 074 242 49 077 14 2 12 68 10 .318 .541
1953 137 445 91 147 32 1 18 87 11 .330 .528
1954 157 543 109 176 21 2 42 135 07 .324 .602
1955 100 316 40 093 08 0 13 46 01 .294 .443
1956 116 353 71 095 16 0 18 69 04 .269 .467
Underlined are Negro League years
Italicized are Mexican League
Bold are National League
Plain are minors

Led his league in homers in 1948 and 1954.

Team and League by year:
1937 - 38 Pitt, Neg NL
1939 Toledo, Neg NL
1940 Indianapolis and Newark, Neg Am
1940 Nuevo Laredo, Mexico
1941 Tampico, Mexico
1942 Newark, Neg Am
1946 Vera Cruz, Mex
1946 Philadelphia, Neg Am
1947 Vera Cruz, Mex
1948 St. Jean, Provicial (?)
1949 Philadelphia, Neg Am
1950 Baltimore, Neg Am
1950 Milwaukee, American Assoc.
1951 Milwaukee, American Assoc.
1952 Boston, NL
1952 Milwaukee, American Assoc.
1953 Dallas, Texas
1954 Beaumont and Dallas, Texas
1955 Los Angeles, PCL
1956 Los Angeles, PCL / Tulsa, TX / Des Moines, Western

Here's a quick take on Martin Dihigo's pitching in the Negro Leagues. The record won't support precise conclusions, but I think it's enough for us to form an impression of Dihigo's general quality as a pitcher.

In the early part of his career, 1924-31, his wins above team are as follows:

Year Record Team WP WAT % of team decisions*
1924 1-4 .308 -0.5 10%
1925 4-7 .423 -0.7 21%
1926 2-2 .564 -0.3 10%
1927 2-0 .500 +1.0 4%
1928 1-1 .615# -0.2 10%
1929 5-3 .538 +0.7 10%
1930 no data
1931 0-1 .733 -0.7 2%

*Based on listed decisions for pitchers, not on team record in the standings
#winning percentage based on Holway's 13 recorded games. Team winning percentage in Gary A.'s 19 recorded games may be different

From this data, it is clear that Dihigo was generally being used as an occasional starter, not a front-rotation pitcher, with 1925 being the one exception. His decision percentage stays pretty constantly at 10%: the greater number of decisions in 1929 is a product of a larger number of recorded team games, not a heavier usage pattern. It is also clear that Dihigo was unexceptional as a pitcher. His winning percentage follows his teams' closely, and he makes no appearances on any league leaderboards during these years.

This does not look like a record for which Dihigo should receive MLE credit.

In 1935 and 1936, however, Dihigo's record is considerably different:

Year Record Team WP WAT % of team decisions*
1935 7-3 .538 +1.1 19%
1936 7-4 .489 +1.6 23%

Here Dihigo is clearly a front-line starter, with a substantial number of wins above his teams. He was 4th in the league in winning percentage, fifth in TRA in 1935, 5th in strikeouts in 1936.

As we're still dealing with a small number of games, I don't think an MLE ERA+ can be projected with precision, especially in the absence of run support data, which might have been significantly affected by Dihigo's own hitting. However, I would suggest that Dihigo's MLE ERA+ would most likely have been in the 115 to 125 range, with 110 and 130 being the absolute high and low possibilities. He'd have been a good to very good pitcher in the majors at this point in his career. He was definitely a very good pitcher in the Negro Leagues, somewhere around 5 to 15 overall, I'd guess.

Even in '35-36 though he's nothing like a full time pitcher. With only about 50 decisions for the team as a whole, 10 is a VERY low season's load -- we're not talking 162 game seasons and 5 man rotations here. Doesn't this mean that he was never considered anything like the ace of his NEL team staff? If that's the case, I would have thought an MLE ERA+ of 110 was too high. Do you have info on who else was pitching for those teams? Surely a pitcher with an MLE ERA+ of 120, the midpoint of your estimate, would have been the ace of the staff, unless it inluded Paige, and used accordingly.

Karl,

Three responses.

1) For a major-league usage pattern, 20% of decisions is high in the 1930s. I know you continue to doubt that the non-league games NeL teams played actually figure into pitcher workloads, but they do, and teams were using 4-5 pitcher rotations just like they were in the majors. 20% of team decisions is clearly a full-time starter.

2) The data for the Cubans shows that they were certainly using a 4-5 pitcher rotation during these seasons.

1935 decision totals (Dihigo bold): 12, 10, 10, 9, 7, 2, 1, 1
1936 decision totals: 12, 11, 9, 7, 5, 1, 1, 1

Dihigo is clearly in a rotation, one of his team's regular starters.

The Cuban teams tended, for what reason I don't know, to use a more regular rotation that some of the other teams, but even so no pitchers of teams with a ful NeL schedule were recording a much greater percentage of team decisions.

Here's the data for 1935

Team--Top Pitcher--% of decisions--team games
Pgh Crawfords--Leroy Matlock--27% (64 team games)
NY Cubans--Luis Tiant, Sr.--23% (52 team games)
Nas Elites--Bob Griffith--21% (32 team games)
Chi Am Giants--Ted Trent--29% (42 team games)
Home Grays--Ray Brown--27% (59 team games)
Phi Stars--Webster McDonald--24% (72 team games)
Bkn Eagles--Leon Day--48% (27 team games)
Nwk Dodgers--Bob Evans--30% (50 team games)

Team--distribution of decisions
Pgh--17, 16, 13, 8, 4, 3, 2, 1
NY--12, 10, 10, 9, 7, 2, 1, 1
Nas--7, 6, 5, 5, 5, 3, 1, 1
Chi--12, 8, 7, 6, 4, 2, 1, 1, 1
Home--16, 15, 11, 10, 6, 1
Phi--17, 15, 12, 11, 10, 3, 2, 1, 1
Bkn--13, 4, 3, 2, 1, 1, 1, 1, 1
Nwk--15, 8, 8, 7, 4, 4, 2, 1,1

Overall, Mackey is 13th among pitchers in an 8-team league in percentage of team decisions. That looks like a full-time starter to me.

He was 5th this year in TRA, and 4th in winning percentage. I said he was among the top 5-15 pitchers in the league. I think that is a reasonable estimate.

There were at least 20 pitchers in the majors in 1935 with an ERA+ of better than 120.

I don't see that Dihigo's usage pattern advances any evidence that would make my estimate of his ERA+ unlikely.

OK, I stand educated. It's a very odd economic decision, to have 4-5 frontline pitchers for only 50 league games, but non-league games obviously played a much more important role, like in the early 1870s. I still think 105-120 is more like it (he wasn't the ace, and there weren't 30 ML pitchers with an ERA+ of 130) but I "see where you're coming from" as they say. If his hitting for NEL teams was in the 115-120 OPS+ range then I agree, he's probably a HOMer. Odd that there aren't more Caruthers-type careers in the majors; don't really see why it's impossible, especially with the DH.

Odd that there aren't more Caruthers-type careers in the majors; don't really see why it's impossible, especially with the DH.

It's a paradox, karlmagnus. While a DH/pitcher is definitely much more possible than a position player/pitcher today, the DH has affected the average batting skill of pitchers, IMO, since a hurler doesn't get a chance to hit in the minors.

Are there any pitchers that hit enough today that a team would want as a DH?

Now, a a true hitter who becomes a pitcher in the majors might be feasible. Having him as a DH and on the mound sounds like a win-win proposition to me.

Plus the DH rule isn't really conducive to batting your pitcher anyway. If you start your DH/P and need to pull him for relief later on, you have to bat your scrubby hitting pitchers(or use a ph for them) for the rest of the game. The pitcher would have to hit alot better than the alternative DH to make that worthwhile.

The pitcher would have to hit alot better than the alternative DH to make that worthwhile.

Agreed.

How many guys have there been in history who could have put up both a 90+ ERA+ and a 90+ OPS+? I have no idea. The answer might be 20 or it might be 800. I suspect the number is higher than we would guess, however.

I suspect the potential for 90/90 is pretty common, but 110/110 much less so. Ferrell is after all unique in the 20C (correct me if I'm wrong) in being a top flight or even above average pitcher with an OPS+ of over 100. Which is why, if Dihigo is really 120/110 or 120/105, he's a HOMer (I'm not sure about 115/105, though.)

WFerrell 117/100
CMays 119/82
DNewcombe 114/85

George Uhle 105/86
Brandon Backe 88/128 :-) (last year he was 100/128)

Well, I'm glad I'm back because now I get to talk about Martin Dihigo, one of my favorite players. Daryn posted some info from me above (#33), but here it is complete:

Dihigo may be the weirdest baseball player I've ever studied.

He was born in 1906, not 1905 (his birth certificate is in his Cuban bio).

Up until 1930, He was on track to be perhaps greatest right-handed hitter of all time. I think he would have matured into a 40 to 50 HR, .350 and up hitter in the Majors if he could have actually played.

But he evidently got bored with being so good and took up pitching and switch-hitting in 1931. In his defense, pitching, especially in the Latin American countries was a much higher paid position than hitter. Dihigo had a hell of an arm.

From 1932 to 1934, Dihigo was the star pitcher on a team sponsored by the dictator of Venezuela's son, the Concordia Eagles (One of his team mates was Tetelo Vargas, the best Dominican of that time).

He wasn't the best hitter in the country, at first (Alejandro Oms and Vargas were); but he sure as hell was the best pitcher (number 2 was Cocaina Garcia, a forgotten but incredible, short and stocky, Cuban pitcher).

But, by 1934, he had evidently mastered switch-hitting and started to really hit again. But he never again showed the power that he had had
previously. I think he could have hit .330 to .350 in the Majors with 30 Home Run power from 1934 to 1938 or so, but his hitting skills faded
quickly after that as he concentrated on pitching and managing.

The Mize quote in Dihigo's thread comes from this time. The Concordia Eagles, Dihigo's team, traveled to Cuba and played against Cuban All-
Star teams. They imported Mize for one of the series. When Mize had to report to spring training, they replaced him with Josh Gibson. I always liked that fact.

In 1935 and 1936, he was a great pitcher and hitter for the New York Cubans. In 1937, Dihigo hit .351 in Trujillo's tournament (second best
hitter to Gibson) and was the second best pitcher to Paige.

From 1938 to 1947, he played in Mexico, and those stats should be available.

One thing to remember about Dihigo is that, like a lot of teenage superstars (Kaline and Yount come to mind), he was better in his 20s than in his 30s. I think this is because it comes so easy to them that they don't work to maintain their talents in their 30s or something.

I think Dihigo would have been a 20 game winner in the Majors from 1931 to 1942 or so (obviously not every year, but in his good ones).

I don't know if this will help you, but I always think of Dihigo as Babe Ruth in Reverse. He would have been number 1 on my ballot, no question.

One final weird note: I read a story about some Americans that traveled to Cuba a couple of years ago. They went to see where Dihigo was buried
and found his remains (bones) in a box in a mausoleum. They actually ended playing around a little bit with his bones.

Probably the most disturbing Negro League tidbit I've ever read.

Some other notes:

41: Chris Cobb-

The Cuban Stars were pretty much traveling teams, playing whereever, though the Eastern Cuban Stars' team (run by Alejandro Pompez) always had some odd park where they would play some games as the home team.

However, the 1935 and 1936 New York Cubans had Dyckman Oval in New York as their primary home field. Dyckman was a bandbox.

But, as always, they also had some other odd parks that they called home. There used to be an interesting article on the net about how the club used Belmar, New Jersey, as one of their home fields in 1935.

47. Chris Cobb-

There is one other thing to remember about the quality of play in the Negro Leagues. The Negro Leagues of the 1920s generally tried to produce 8 team Leagues (though the eastern usually had six with the Homestead Grays and several other good teams abstaining from joining).

On the other hand, the Negro Leagues of the 1940s settled down to two 6 team leagues, a large difference.

One of the interesting aspects of the Negro Leagues is the Latin influence. Originally, in the 1920s, it is almost 100% Cuban (with Oscar Levy, the Panamanian who may have actually been Jamaican, the great exception). At the end of the 1920s, the Dominicans arrive.

Throughout the 1920s, there were two full Cuban squads in the Negro Leagues.

But, after 1930 and the Depression, much of this Latin talent goes back to Mexico, Venezuela, and other Latin American countries.

The 1940 Negro Leagues have one Latin team, the New York Cubans, which (despite the name) is loaded with Puerto Ricans and Dominicans, not just Cubans.

From 1931 or so on, the Latin baseball demographic, which was increasing at a great rate (a trend still happening now), pretty much stayed in Latin America until after integration.

In my opinion, the Mexican League which siphoned even more Latin talent than African-American talent from the Negro Leagues was of about Triple-A quality from about 1940 to 1948, with peaks of quality in 1940-41 and 1946-47.

Very obviously, the Mexican League (which started in 1937 with something like 10 teams) was of lesser quality from 1937 to 1939. But, in 1940 the League stablized at 6 teams. In 1940, the Mexican League teams offered top salaries of 400 to 500 dollars a month.

Top salaries in the Negro Leagues at this time were 200 to 250 dollars. They couldn't compete until the World War Two econonmic boom hit. So, in 1940 and 1941, the Mexican League was able to draw a lot of talent out of the Negro Leagues.

From 1943 to 1945, this is not as true, though many (perhaps most) Latins stayed in Mexico.

Of course, from 1946 to 1948, Jorge Pasqual had his little war with the Majors and once again the Negro Leagues could not compete economically. However, there is a large mitigating factor in 1946. The Mexican League expanded from 6 to 8 teams in 1946 which kept the quality from rising even more.

The reason for the expansion was that Jorge Pasqual did not non-white players to lose their jobs because of his importation of white players from American organized ball.

For a very good example of all this, all you need do is study the career of Raymond Dandridge, who went from Negro Leagues to Mexican Leagues to Triple-A in a crazy-quilt pattern throughout the 1940s. The Major League Stars book referenced by Kelly above has his stats.

Another way of looking at this is that, in the 1920s, there were more or less 16 Negro League teams operating that I believe were, all combined, of Triple-A quality. In the 1940s, there were 3 non-white Leagues operating of Triple-A quality, consisting of 18 teams.

I also believe that there was more, not less, non-white talent in the 1940s, especially due to the opening up of Venezuela, the Dominican, and especially Puerto Rico; but that is just an opinion.

I have a study on Mexican League quality here somewhere and I'll try to post it this weekend.

53. Karl Magnus-

One of the odd things that you find when you research baseball is that pratically all professional teams have always played 150 or more games per season.

Major League teams from the 1870s and 1880s played 60 or 70 or 84 or 112 or 140 League games, but they also played enough non-league games that they would still play 150 or more games per season, no matter how many league games they played.

The Cuban Giants, in fact, were established to take advantage of this fact. One of the reasons for the demise of the Cuban Giants is that teams began playing many more League games and less exhibitions as the years went by.

The Cuban Giants, in the 1880s, sometimes published their full season records. The team was playng 150 games per year.

This is also true of Negro League teams, in some ways to an even greater extent because of economics. The Homestead Grays and Hilldale, for instance, often published their full records in the 1920s.

These teams were playing well over 150 games per season, sometimes approaching 200.

And their roster size was limited.

If you were a pitcher in a Negro League rotation, you were getting the crap pitched out of you.

The Homestead Grays in the 1920s published the starting records of their four main pitchers. Each man had won 30 games and started over 40 games. This was a usual thing.

Plus the DH rule isn't really conducive to batting your pitcher anyway. If you start your DH/P and need to pull him for relief later on, you have to bat your scrubby hitting pitchers(or use a ph for them) for the rest of the game. The pitcher would have to hit alot better than the alternative DH to make that worthwhile.

This is an interesting point of difference between the rules of Organized Baseball and the rules of the NCAA. Under NCAA rules, a starting pitcher who is in the batting order can be removed from the game as a pitcher but remain in the game as the DH. This makes pitcher/DH a viable NCAA position, and if you've got access to NCAA statistics, you can probably pick out a couple.

One of the odd things that you find when you research baseball is that pratically all professional teams have always played 150 or more games per season.

In the U.S., except for Florida and the West Coast, the season is limited by weather to 7-8 months, so that implies 5-6 games per week. (Watched by what paying audience, during the week with no lights? Oh, well.) Since I mentioned the NCAA above, I should comment that they're different than the pros: they only play 3-4 games per week, and stop playing before summer.

Teams in Mexico and the Caribbean did not have that weather-imposed limitation on season length. Was the main response seems to consider winter and summer to be different territories populated by different teams and leagues?

Gadfly,

If you were going to guess at a park factor for Dyckman Oval, what would it be?

In the absence of data and the reputation as a strong hitter's park and with the added uncertainty of there being possibly a second home ballpark in this case, I'd be inclined to go with something like 103 or 104.

Very helpful information about the changing sizes of the leagues and the movements of the talent pool. The talent situation was a bit more stable than I had thought from Dr. Chaleeko's lists.

63. OCF-

You are forgetting one thing, double-headers. The Negro League teams almost invariably played double-headers on both Saturday and Sunday, not to mention every Holiday. They also, especially when times were tough, sometimes played triple-headers.

In one of the Reach Guides, there is an actual account of every game Hilldale played. The team was playing every day. Surely weekday games, which were usually played late in the day and called twilight games, did not draw much; but the players were on salary and the owner (Ed Bolden) wouldn't (and couldn't) let any chance to make a quarter pass by.

64. Chris Cobb-

I've never seen measurements for Dyckman Oval, but it was small and would have helped a power hitter, i.e. Dihigo, most. But, of course, the New York Cubans did play all over. I guess I would use 105, but it might be as high as 110.

One more thing-

This is just my opinion; but I think, if Dihigo had played in the Majors, they would have stuck him in right field, told him to cut out that switch-hitting crap, give up his pitching aspirations, and concentrate on his stroke.

Just like they did to Ruth.

How on earth you would measure that, I have no idea. Probably the only way to measure Dihigo would be through Win Shares, while realizing that, if he had just concentrated on one talent, he would have been even greater.

Another complication with Dihigo's career is managing. Basically, from 1933 on until 1955, Dihigo was a manager. At that time, there was more money in managing than there was in playing, and many players took on the managing job for this reason.

This decision almost always impacted their actual playing career (see Chance, Bresnahan, Cobb, etc) in a negative way. If Dihigo doesn't become a manager, the second half of his career (his 30s) is probably much better.

I like to make allowances for this, but that is just a personal preference also.

Dihigo looks sort of like the anti-Caruthers. IOW the main argument for his induction will be his hitting, and the pitching presents a significant bonus. At least that seems to be the pattern in the NeL.

In Cuba and Mexico, OTOH, he pitched a lot more relatively speaking and could even be said to have been a better pitcher than hitter.

So the key questions seem to be:

1. How does he compare as a hitter with Suttles or Beckwith or even Taylor or Poles?

2. And how good (or bad) were the Cuban and Mexican Leagues. Why was he more effective (or at least used more) as a pitcher in Cuba and Mexico? If simply because the Cuban and Mexican Leagues were not as good, then why does he not appear to have been as good a hitter in Cuba and Mexico?

More questions than answers at the moment, not that that is a surprise. That just puts him into a lump with Bell and Mackey, though for now I guess I will have Dihigo somewhere between 3 and mid-ballot rather than somewhere between 10 and 20 with Bell and Mackey.

But given these two I think pertinent questions, I can't quite get Dihigo into an "elect me" slot.

Dihigo looks sort of like the anti-Caruthers. IOW the main argument for his induction will be his hitting, and the pitching presents a significant bonus.

Like George Van Haltren? Different timing, of course. Van Haltren pitched from the short-distance box at the very beginning of his career, and Van Haltren's pitching wasn't all that valuable.

These questions focus our attention very effectively:

1. How does he compare as a hitter with Suttles or Beckwith or even Taylor or Poles?

I hope to have some data on this tonight or tomorrow morning. I'd say he was definitely better than Taylor or Poles in that he was a complete hitter: decent average, very good power, and plate discipline. Neither Taylor nor Poles were true power hitters, though both may have hit for higher average than Dihigo.

2. And how good (or bad) were the Cuban and Mexican Leagues. Why was he more effective (or at least used more) as a pitcher in Cuba and Mexico? If simply because the Cuban and Mexican Leagues were not as good, then why does he not appear to have been as good a hitter in Cuba and Mexico?

It seems to me that his usage patterns as a pitcher were a result of his inclinations as much as his effectiveness. In his first stint in the NeL, he was a part-time pitcher, so he was ok, but nothing special in effectiveness. He had 3 years to learn to pitch in Venezuela and Cuba, because he wanted to, before he returned to the NeL in 1935. Then he was a very good pitcher _and_ he was the manager, so he decided what he would do. It was his decision to concentrate more on pitching. The extent to which that was strategic and the extent to which that was a personal preference is hard to know, but it seems likely that both strategy and preference were factors.

His decline as a hitter is partly explainable by his concentration on pitching, partly explainable by aging. Dihigo was 34 in 1940, so we would expect to see signs of decline in most hitters by this age, even if they weren't devoting their attention to pitching and to managing in addition to hitting and playing the field.

His lower hitting stats in Cuba are partly attributable to its being in general a pitchers' league.

Gadfly says Dihigo took up switch-hitting in 1931 (one can see a big drop in his NeL stats), but I wonder if he was trying it out in Cuba earlier than that -- the big drop in his batting and slugging seems to come between 1928 and 1930 (though the small sample sizes make it dangerous to place too much weigh on trends here).

In both the Cuban Leagues and the NeL, he was a great hitter in the late 1920s, then he tailed off sharply before going to Venezuela; he was a good hitter when he returned to these leagues in 1935.

Martin Dihigo MLEs

Well, nothing is easy with Mr. Dihigo. There are three serious issues with his MLEs for his Negro-League seasons, in addition to their being split up and giving us only half of his career, and the whole pitching thing.

1) The data provided by MacMillan 10th edition and by Holway are greatly at variance for Dihigo’s play in the 1920s, such that Dihigo’s value appears radically different, depending on which set of statistics you accept. I have not seen this degree of variance before. Two seasons in particular are at issue, 1926 and 1929.

In 1926, the difference may simply be explained by Holway having more data. Here are the two lines.

Mac 10th – 76 ab, 32 hits, 5 2b, 0 3b, 8 hr, .421 ba, .803 sa
Holway – 205 ab, 67 hits, ? 2b, ? 3b, 12 hr, .327 ba, .576 sa

The latter line, which has nearly 3x the at bats, makes for a smoother pattern of development in Dihigo’s career (sa 1925-28 .403, .576, .610, .567) so I’m inclined to accept Holway over Macmillan’s .803 sa season here.

In 1929 however, the data are simply and radically incommensurate.

Mac 10th – 207 ab, 77 hits, 8 2b, 1 3b, 13 hr, .372 ba, .609 sa
Holway – 266 ab, 69 hits, ? 2b, ? 3b, 18 hr, .259 ba, .515 sa

Holway has more at bats and fewer hits. The home run totals vary about as one would expect given the different at bat totals, but there’s clearly an error in one data set or the other with respect to the hits. I’m not sure whom to trust, though in this case I am inclined to favor the Macmillan data.

Given the short length of Dihigo’s career, the difference in these seasons has a huge impact on his career totals and on his peak. I’ve run the totals using only Mac 10th data and the totals using Holway’s data whenever possible (my usual practice), and here are the two career lines:

Mac 10th MLEs .312 ba, .387 obp, .491 slg
Holway MLEs .287 ba, .367 obp, .464 slg

I have also done a set of MLEs using Holway for 1926 and Macmillan for 1929 (Because of regressions, changes in one season affect surrounding seasons, so one can’t simply pick and choose from the regressed data lines, though one can from the unregressed data.)

2) Regression is also a problem for Dihigo, especially in the Macmillan version. In Macmillan, Dihigo’s early career is saddle-shaped low-high-medium-high-low. This distribution causes the regression program to totally erase his peaks, creating one long, even plateau of very good. Compare the straight translations from MacMillan to the regressed translations below:

year ba/sa – reg ba/sa (age)
1925 .280/.344 -- .283/.365 (19)
1926 .391/.693 -- .332/.532 (20)
1927 .340/.516 -- .332/.500 (21)
1928 .312/.479 -- .333/.521 (22)
1929 .325/.489 -- .330/.502 (23)
1930 .367/.615 -- .329/.509 (24)
1931 .248/.289 -- .271/.351 (25) – Switch-hitting experiment begins

The regression is much more than usually harsh on Dihigo’s peaks.

I’ve provided regressed and unregressed MLEs for Macmillan, Holway, and Macmillan/Holway data sets.

So, there are an unusual number of issues with this data, so you get to look at it in six different versions!

3) Park Effects.

The data for 1935 and 1936 have fewer problems than the 1920s. Dihigo’s production was, relative to the 1920s, consistent in these seasons, so the regression works fairly well, and Holway and Macmillan’s data match reasonably well. I’ve used Holway’s data because it includes more at bats, but the difference between the two is not large.

Gadfly tells us, however, that the team’s home park was a bandbox. Given all the uncertainty about park factors, I’ve gone with a park factor of 104 for these two seasons, but it’s possible that the factor was as high as 110. A park factor of that size would lower Dihigo’s totals by 5%. Those two seasons would still be very good even with the reduction, but they would stand out less than they do.

In fact, I think it will turn out that, when offensive context is fully adjusted for, Dihigo was a significantly better hitter in 1935 and 1936 than in the late 20s, which is what we’d generally expect from a player in his age 29 and 30 seasons when they are compared to his play ages 19-25. Some of Dihigo’s gaudy early numbers are attributable, I think, to the period being one of exceptionally high offense in the NeL.

I’ll be very interested to see the OPS+ numbers for Dihigo, and I hope David Foss will be willing to run several sets.


Two final notes prior to posting the MLEs:

1) On playing time estimates. Since Dihigo’s career totals aren’t really at issue here, I’ve been fairly casual about my playing time projections. I’ll post his MLEs from 1923 on, but for career totals I count 1925 as his first MLE season. I’ve modeled his playing time on Mel Ott, the other teen hitting prodigy of the mid-1920s. For his 1935-36 seasons, I’ve assumed that he would start about 32 games, and that he would have a game off after each start, so he plays about 120 games a season.

2) On walks. Since the Mexican League data includes walks, my estimates of Dihigo’s walks are much more firmly grounded in data than for most NeL players. So I think the walk estimates are quite reliable . . .

Martin Dihigo MLEs

Version 1 – Unregressed Macmillan 10th 

Year Team EqG  PA   BB  Hits  TB  BA  OBP  SA
(1923 Cub  50  200  11   47   51 .248 .288 .268)
(1924      50  200  13   43   59 .227 .276 .315)
1925       82  318  21   83  102 .280 .328 .344
1926      124  521  35  190  336 .391 .432 .693
1927      150  630  56  195  296 .340 .398 .516
1928 Home 148  622  69  173  265 .312 .388 .479
1929 Hill 150  630  68  183  275 .325 .398 .489
1930 Cub  154  647  71  212  354 .367 .437 .615
1931 Hill 135  567  71  123  143 .248 .342 .289
1932-34 In Venezuela – No Data   
1935 Cub  122  488  71  111  203 .267 .374 .486
1936      120  480  69  123  256 .300 .401 .622
Total    1185 4902 532 1392 2230 .319 .392 .510
N.B.  Career totals do not include 1923-24

Version 2 – Regressed Macmillan 10th

Year Team EqG  PA   BB  Hits  TB  BA   OBP  SA
(1923 Cub  50  200  11   47   55 .249 .289 .290)
(1924      50  200  13   43   59 .228 .277 .317)
1925       82  318  21   84  108 .283 .330 .365
1926      124  521  39  160  257 .332 .381 .532
1927      150  630  56  191  287 .332 .392 .500
1928 Home 148  622  67  184  289 .333 .404 .521
1929 Hill 150  630  68  185  282 .330 .402 .502
1930 Cub  154  647  75  188  291 .329 .406 .509
1931 Hill 135  567  69  135  175 .271 .360 .351
1932-34 Venezuela        
1935 Cub  122  488  71  114  215 .274 .379 .516
1936      120  480  70  119  238 .290 .394 .579
Total    1185 4902 535 1361 2142 .312 .387 .491
N.B.  Career totals do not include 1923-24

Version 3 – Unregressed Holway (Mac 10th in seasons for which Holway’s data lack at bats)

Year Team EqG  PA   BB  Hits  TB  BA   OBP  SA
(1923 Cub  50  200  11   47   51 .248 .288 .268)
(1924       50  200  13   43   59 .227 .276 .315)
1925       82  318  21   83  102 .280 .328 .344
1926      124  521  40  146  239 .303 .357 .497
1927      150  630  56  195  296 .340 .398 .516
1928 Home 148  622  69  173  265 .312 .388 .479
1929 Hill 150  630  77  125  229 .227 .321 .414
1930 Cub  154  647  71  209  350 .364 .434 .609
1931 Hill 135  567  71  123  143 .248 .342 .289
1932-34 Venezuela        
1935 Cub  122  488  69  122  203 .290 .391 .486
1936      120  480  69  122  260 .298 .399 .633
Total    1185 4902 544 1298 2088 .298 .376 .479
N.B.  Career totals do not include 1923-24

Version 4 – Regressed Holway (Mac 10th in seasons for which Holway’s data lack at bats)

Year Team EqG  PA   BB  Hits  TB  BA   OBP  SA
(1923 Cub  50 200   11   47   55 .249 .289 .290)
(1924      50 200   13   43   59 .228 .277 .317)
1925       82 318   21   83  108 .281 .329 .364
1926      124 521   40  143  226 .298 .352 .470
1927      150 630   58  178  275 .311 .374 .480
1928 Home 148 622   70  165  267 .299 .378 .483
1929 Hill 150 630   75  138  243 .249 .338 .438
1930 Cub  154 647   77  173  279 .303 .386 .490
1931 Hill 135 567   71  125  168 .252 .345 .338
1932-34 Venezuela        
1935 Cub  122 488   69  122  214 .292 .392 .511
1936      120 480   70  121  242 .296 .398 .590
Total    1185 4902 551 1248 2021 .287 .367 .464
N.B.  Career totals do not include 1923-24

Version 5 – Unregressed Holway with Mac 10th for 1929

Year Team EqG  PA   BB  Hits  TB  BA   OBP  SA
(1923 Cub  50 200   11   47   51 .248 .288 .268)
(1924      50 200   13   43   59 .227 .276 .315)
1925       82 318   21   83  102 .280 .328 .344
1926      124 521   40  146  239 .303 .357 .497
1927      150 630   56  195  296 .340 .398 .516
1928 Home 148 622   69  173  265 .312 .388 .479
1929 Hill 150 630   68  183  275 .325 .398 .489
1930 Cub  154 647   71  209  350 .364 .434 .609
1931 Hill 135 567   71  123  143 .248 .342 .289
1932-34 Venezuela        
1935 Cub 122 488    69  122  203 .290 .391 .486
1936     120 480    69  122  260 .298 .399 .633
Total   1185 4902 535 1355 2134  .310 .386 .489
N.B.  Career totals do not include 1923-24

Version 6 – Regressed Holway with Mac 10th for 1929

Year Team EqG  PA   BB  Hits  TB  BA   OBP  SA
(1923 Cub  50 200   11   47   55 .249 .289 .290)
(1924      50 200   13   43   59 .228 .277 .317)
1925       82 318   21   83  108 .281 .329 .364
1926      124 521   40  143  226 .298 .352 .470
1927      150 630   57  187  282 .326 .387 .492
1928 Home 148 622   68  178  278 .321 .396 .502
1929 Hill 150 630   68  185  282 .329 .402 .502
1930 Cub  154 647   75  187  290 .327 .405 .506
1931 Hill 135 567   69  135  174 .271 .360 .350
1932-34 Venezuela        
1935 Cub  122 488   69  122  214 .292 .392 .511
1936      120 480   70  121  242 .296 .398 .590
Total    1185 4902 537 1342 2095 .307 .383 .480
N.B.  Career totals do not include 1923-24

Thanks again Chris! From my own feeble attempts to do MLEs for minor leaguers, I know a lot of time and effort goes into these.

I don't know how good the Cuban League was during parts of Dihigo's career (it apparently fell apart for a couple of years during the early 1930s due to political upheaval, and I'm not sure of the quality after it first reconstituted). But Dihigo does show up on a lot of the Cuban career leader lists in Figueredo. Ranks 4th in years played (22), 8th in runs scored (356), 9th in doubles (100), and tied for 4th in triples (44). As a pitcher he ranks 2nd (behind Luque) in years as a pitcher (19), 4th in games (248), 1st in complete games (121), 1st in wins (107), 10th in losses (56), and 4th in winning percentage (.656).

If you think it would be useful in evaluating him as a pitcher, I could take a look at his wins above team in Cuba. (It may be hard to know what to do with that information, though, if we don't know the quality of the league.) It seems like, in principle, we ought to be able to use Luque's major league and Cuban career records to infer the quality of the Cuban League.

I'm in the midst of trying to estimate some league totals for the Cuban League in the 1920s, using Figueredo's book. It'll be a little tricky, since in some years a lot of players are omitted, and he pretty much doesn't include pitchers unless they played a significant amount at another position. But I think it's better information than Holway gives for NeL seasons.

Eventually I want to collect Cuban League box scores to try to reconstruct these seasons from the ground up. Luckily I have access to a run of the Cuban newspaper La Lucha, which carried great box scores in the 1900s (including things like batters' BBs and Ks, sac hits, hit by pitch, etc.). Haven't checked on the 1920s, though.

Also--I happen to have access to yet another count of Dihigo's 1929 season (not compiled by me):

AB-194
H-62
HR-11
AVE-.320
SLG-.552 (but I don't have totals for doubles and triples)

This looks to be less complete than either the Mac or Holway totals, and isn't really consistent with either; it does, however, tend to lend credibility to the Mac's higher averages (and maybe also its lower hr figure). To get to Holway's totals from these, Dihigo would have to have hit 7 for 72 in the missing games, with all 7 hits being home runs.

Chris, can us simple folk have OPS+ estimates from these? I have to say Macmillan looks more likely than Holway -- career BA significantly under .300 without Suttles' high HR totals, seems a little unlikely for someone with the repuation of an ace hitter. (Gadfly faints as I make a point supportive of a marginal NEL candidacy!)

Brent:

I think comparison of Luque's Cuban pitching record to Dihigo's would be very instructive. All the data needed should be on the site: it's just a matter of digging up Luque's.

Gary A.:

Thanks for the 1929 data! It tips the scale pretty firmly against Holway's numbers for that season. Having slept on the numbers, I'd say that I myself am going to use Version 5/6 for my own rankings, keeping in mind that Dihigo's experiment with switch-hitting in 1931 skews his development curve (and his abbreviated career totals) in ways that would never had happened in the majors.

I am persuaded, though, that the regression itself is not as violent to Dihigo's peak as I was first thinking. His _very_ high totals all occur in seasons with small data sets. Regression reminds us that those totals are _likely_ to return towards the mean when spread to a full season, just as they seem to in the move from Macmillan's small data set for 1926 to Holway's larger one.

Karlmagnus:

DavidFoss is the man with the data to produce OPS+ figures. I'm sure he'll produce them as soon as he can. When I post MLEs in the middle of the night, I don't make it easy for him to get OPS+ figures up right away.

I’ll be very interested to see the OPS+ numbers for Dihigo, and I hope David Foss will be willing to run several sets.
...
DavidFoss is the man with the data to produce OPS+ figures. I'm sure he'll produce them as soon as he can. When I post MLEs in the middle of the night, I don't make it easy for him to get OPS+ figures up right away.

Yeah... The multiple sets is no problem because I have scripts which semi-automate the process, but I didn't catch this last night and I'm already at work and away from my HOM stuff. I have to run somewhere right after work this evening, too. Probably won't get to this until Saturday morning, sorry. A bit frustrating because it should only take me a few minutes -- oh well.

The impatient can take the context information from my previous reports as there are no new years up there.

Martin Dihigo Version 6 OPS+

Count me among the impatient, I guess.

These numbers may vary slightly from David’s official ones, and the layout isn't as complete or elegant, but it gives us an idea.

First three columns are Version 6 MLEs for ba, obp, slg
Second three columns in parentheses are league averages for same.
Paired columns are obp+/sa+
Final column is OPS+

1923 .249 .289 .290 (0.292/0.356/0.405)   81/88     69
1924 .228 .277 .317 (0.294/0.356/0.406)   78/78     56
------------------------------------------------------
1925 .281 .329 .364 (0.300/0.364/0.425)   90/86     76
1926 .298 .352 .470 (0.289/0.355/0.402)   99/117   116
1927 .326 .387 .492 (0.292/0.355/0.406)   109/121  130
1928 .321 .396 .502 (0.290/0.355/0.412)   112/122  134
1929 .329 .402 .502 (0.298/0.363/0.432)   111/116  127
1930 .327 .405 .506 (0.312/0.370/0.464)   109/109  118
1931 .271 .360 .350 (0.285/0.344/0.403)   105/87    92
1932-34 NO DATA
1935 .292 .392 .511 (0.286/0.341/0.407)   115/126  141
936  .296 .398 .590 (0.286/0.345/0.400)   115/148  163
tot. .307 .383 .480

Count me among the impatient, I guess.

Thanks! :-)

No worries, of course. I'll run my scripts as soon as I can in the next day or two.

Knocking out 1923-25, when he was very young and wouldn't have been an MLer (probably) that gives 8 seasons with an average extended-peak (presumably) OPS+ of 127. Given the pitching as well, that scrapes over my HOM bar, but it doesn't vault it. Also 1931 is anomalously low and 1936 anomalously high -- further data on either of those two seasons might change my view.

Some comments on 1931 and 1936

1931 is low because Dihigo took up switch-hitting that season: the problem is not with the data but with the man's career choices . . .

1936 is high either because a) in isolation a reliable mean for regression can't be calculated or b) Dihigo, at 30, had reached his prime as a power hitter and had a career year or c) both.

It is unlikely that there are data errors for 1936, as Holway and Macmillan 10th agree pretty closely, but since we have only 118 at bats on record that year the sample size problem is rather small. However, Dihigo's data for 1935 and 1936 combined (which covers 280 at bats) look to be better than his play in the 1920s, so I think a larger base for the regression would be highly unlikely to lower his OPS+ for 1936 below 150.

But karl, it really was an 11 year extended peak because he played in Venezuela from 1932-1934. While he did pitch a lot those three years, according to Gadfly he was still a pretty good hitter.

I guess, on the other hand, if he was learning switch hitting those year it is hard to tell where between 92 OPS+ and 141 OPS+ he would have performed.

Yes, I'm not implying that he had only an 8 year career. Length of prime would depend heavily on any data we may be able to convert from 1937-40, obviously, too.

I should reiterate also for 1935-36 that it is possible that the park factor is too low, although a substantial hitting park factor (104) was used in the conversion for those seasons.

One reason that ages 29-31 isn't a player's peak (as previously thought) is because a player's defensive abilities are beginning to erode and may take away from their offense. It is entirely possible that 1936 was the best offensive season for Dihigo but that he wasn't the outfielder that he was a few years back.

Dihigo's reputation suggests a question and the data above suggests an answer. The question is, was Dihigo so good that he had not one but TWO HoM careers.

The only player I'm aware of now who had two HoM careers is Cy Young, one in the '90s and another in the '00s. Caruthers certainly didn't have two, neither of his parts (pitching or hitting) was a HoM career by itself. Ditto Babe Ruth, who was not a HoM pitcher (see Smokey Joe Wood). George Van Haltren was mentioned as a distant comp of Dihigo but he certainly was not a HoM pitcher, nor George Sisler, and both of their better careers (as position players) have thus far not been adjudged to be HoM careers anyway.

Back to Dihigo. First, was he a HoM hitter? Well, if Suttles is not a HoM hitter then Dihigo is not a HoM hitter, but the likelihood is that Suttles will be a HoM hitter sometime and it is not clear that Dihigo is not a HoM hitter.

OTOH he is not a HoM NeL hitter (only 11 years) nor is he a HoM Mexican and Cuban League hitter--the latter if only because we don't know quite what to make of it.

Then, is he a HoM pitcher? Probably not, again because we really don't know what to make of his pitching mostly in the Mexican and Cuban Leagues.

So taking the different parts, there is not an obvious HoM career in any of the individual pieces or parts. So his case depends on putting the pieces together.

Still looking at the pieces suggests a HoMer to me. IOW his NeL hitting looks half as good as Suttles'. His Cuban and Mexican hitting, assuming we know from his NeL record just how good he was, adds almost another decade to the mix at (presumably) the same level. I mean, it is the same person.

His Cuban and Mexican pitching looks as good as, say Mendez,' who gets votes on that single little piece of achievement alone. Add his NeL pitching, which is not quite as good as Mendez,' but that adds something.

So I would say that Dihigo is 80 percent of Suttles plus 90 percent of Mendez. We don't have to elect either Suttles (though we will) or Mendez in order to elect 80 percent of the one plus 90 percent of the other.

I guess the real comp here for the odd make-up, though nothing for style, is John Ward, who was 50 percent of a HoM pitcher and 50 percent of a HoM IF and who, adding the pieces together, just barely squeezed over the bar. Dihigo's case is more scattered and more complex, but the parts are both more numerous and also bigger on average than Ward's.

As far as Dihigo's 1929 stats go, I would just use the published ones (which are pretty reliable):

Batting-
AB-R-H-2B-3B-HR-BA-SA-SB-SH
251-72-97-15-4-18-.386-.693-21-6
Finished 2nd in HR
Finished 8th in BA
Finished 6th in SB

Fielding-
PO-A-E
287-153-29

Looking back at my post, I want to make clear that, when I say pretty reliable, I am only referring to the 1929 stats that were compiled by the American League, not all published Negro League stats.

As for evaluating Dihigo's career, I still think a Win Shares approach is the way to go. Figure out win shares for his hitting and his pitching and then combine the two into one long, and very odd, career path.

Gadfly, I've heard about those published 1929 stats; where were they published? In Reach or Spalding guides, or in newspapers?

Gary A-

Pittsburgh Courier, September 28, 1929

When I said American League in Post 87, I meant Negro American League (NAL).

One of the newspaper writer's great complaints about the 1923-1928 Eastern Colored League was that there never enough statistical info. The NAL, lead by Cum Posey, tried to remedy this and actually kept pretty good stats.

I just wish they had published games played.

Martin Dihigo

-First you have Year, Team(s), 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+

Version 1 – Unregressed Macmillan 10th

1923  Cub 200  0.248/0.290/0.268   (0.292/0.356/0.405)    85/ 81/ 66     48
1924  Cub 200  0.227/0.280/0.315   (0.294/0.356/0.406)    77/ 79/ 78     56
--
1925  Cub 318  0.280/0.327/0.344   (0.300/0.364/0.425)    93/ 90/ 81     71
1926  Cub 521  0.391/0.432/0.693   (0.289/0.355/0.402)   135/122/172    194
1927  Cub 630  0.340/0.398/0.516   (0.292/0.355/0.406)   116/112/127    139
1928 Home 622  0.312/0.389/0.479   (0.290/0.355/0.412)   108/110/116    126
1929 Hill 630  0.325/0.398/0.489   (0.298/0.363/0.432)   109/110/113    123
1930  Cub 647  0.367/0.437/0.615   (0.312/0.370/0.464)   118/118/133    151
1931 Hill 567  0.248/0.342/0.289   (0.285/0.344/0.403)    87/ 99/ 72     71
--
1935  Cub 488  0.267/0.373/0.486   (0.286/0.341/0.407)    93/109/119    129
1936  Cub 480  0.300/0.400/0.622   (0.286/0.345/0.400)   105/116/156    171

Version 2 – Regressed Macmillan 10th

1923  Cub 200  0.249/0.290/0.290   (0.292/0.356/0.405)    85/ 81/ 72     53
1924  Cub 200  0.228/0.280/0.317   (0.294/0.356/0.406)    78/ 79/ 78     57
--
1925  Cub 318  0.283/0.330/0.365   (0.300/0.364/0.425)    94/ 91/ 86     77
1926  Cub 521  0.332/0.382/0.532   (0.289/0.355/0.402)   115/108/132    140
1927  Cub 630  0.332/0.392/0.500   (0.292/0.355/0.406)   114/110/123    134
1928 Home 622  0.333/0.404/0.521   (0.290/0.355/0.412)   115/114/126    140
1929 Hill 630  0.330/0.402/0.502   (0.298/0.363/0.432)   111/111/116    127
1930  Cub 647  0.329/0.406/0.509   (0.312/0.370/0.464)   105/110/110    120
1931 Hill 567  0.271/0.360/0.351   (0.285/0.344/0.403)    95/105/ 87     92
--
1935  Cub 488  0.274/0.379/0.516   (0.286/0.341/0.407)    96/111/127    138
1936  Cub 480  0.290/0.394/0.579   (0.286/0.345/0.400)   101/114/145    159

Version 3 – Unregressed Holway (Mac 10th in seasons for which Holway’s data lack at bats)

1923  Cub 200  0.248/0.290/0.268   (0.292/0.356/0.405)    85/ 81/ 66     48
1924  Cub 200  0.227/0.280/0.315   (0.294/0.356/0.406)    77/ 79/ 78     56
--
1925  Cub 318  0.280/0.327/0.344   (0.300/0.364/0.425)    93/ 90/ 81     71
1926  Cub 521  0.303/0.357/0.497   (0.289/0.355/0.402)   105/101/124    124
1927  Cub 630  0.340/0.398/0.516   (0.292/0.355/0.406)   116/112/127    139
1928 Home 622  0.312/0.389/0.479   (0.290/0.355/0.412)   108/110/116    126
1929 Hill 630  0.227/0.321/0.414   (0.298/0.363/0.432)    76/ 88/ 96     84
1930  Cub 647  0.364/0.433/0.609   (0.312/0.370/0.464)   117/117/131    148
1931 Hill 567  0.248/0.342/0.289   (0.285/0.344/0.403)    87/ 99/ 72     71
--
1935  Cub 488  0.290/0.391/0.486   (0.286/0.341/0.407)   101/115/119    134
1936  Cub 480  0.298/0.398/0.633   (0.286/0.345/0.400)   104/115/158    174

Version 4 – Regressed Holway (Mac 10th in seasons for which Holway’s data lack at bats)

1923  Cub 200  0.249/0.290/0.290   (0.292/0.356/0.405)    85/ 81/ 72     53
1924  Cub 200  0.228/0.280/0.317   (0.294/0.356/0.406)    78/ 79/ 78     57
--
1925  Cub 318  0.281/0.327/0.364   (0.300/0.364/0.425)    94/ 90/ 86     75
1926  Cub 521  0.298/0.351/0.470   (0.289/0.355/0.402)   103/ 99/117    116
1927  Cub 630  0.311/0.375/0.480   (0.292/0.355/0.406)   107/106/118    124
1928 Home 622  0.299/0.378/0.483   (0.290/0.355/0.412)   103/106/117    124
1929 Hill 630  0.249/0.338/0.438   (0.298/0.363/0.432)    84/ 93/101     95
1930  Cub 647  0.303/0.386/0.490   (0.312/0.370/0.464)    97/104/106    110
1931 Hill 567  0.252/0.346/0.338   (0.285/0.344/0.403)    88/100/ 84     84
--
1935  Cub 488  0.292/0.391/0.511   (0.286/0.341/0.407)   102/115/126    140
1936  Cub 480  0.296/0.398/0.590   (0.286/0.345/0.400)   103/115/148    163

Version 5 – Unregressed Holway with Mac 10th for 1929

1923  Cub 200  0.248/0.290/0.268   (0.292/0.356/0.405)    85/ 81/ 66     48
1924  Cub 200  0.227/0.280/0.315   (0.294/0.356/0.406)    77/ 79/ 78     56
--
1925  Cub 318  0.280/0.327/0.344   (0.300/0.364/0.425)    93/ 90/ 81     71
1926  Cub 521  0.303/0.357/0.497   (0.289/0.355/0.402)   105/101/124    124
1927  Cub 630  0.340/0.398/0.516   (0.292/0.355/0.406)   116/112/127    139
1928 Home 622  0.312/0.389/0.479   (0.290/0.355/0.412)   108/110/116    126
1929 Hill 630  0.325/0.398/0.489   (0.298/0.363/0.432)   109/110/113    123
1930  Cub 647  0.364/0.433/0.609   (0.312/0.370/0.464)   117/117/131    148
1931 Hill 567  0.248/0.342/0.289   (0.285/0.344/0.403)    87/ 99/ 72     71
--
1935  Cub 488  0.290/0.391/0.486   (0.286/0.341/0.407)   101/115/119    134
1936  Cub 480  0.298/0.398/0.633   (0.286/0.345/0.400)   104/115/158    174

Version 6 – Regressed Holway with Mac 10th for 1929

1923  Cub 200  0.249/0.290/0.290   (0.292/0.356/0.405)    85/ 81/ 72     53
1924  Cub 200  0.228/0.280/0.317   (0.294/0.356/0.406)    78/ 79/ 78     57
--
1925  Cub 318  0.281/0.327/0.364   (0.300/0.364/0.425)    94/ 90/ 86     75
1926  Cub 521  0.298/0.351/0.470   (0.289/0.355/0.402)   103/ 99/117    116
1927  Cub 630  0.326/0.387/0.492   (0.292/0.355/0.406)   112/109/121    130
1928 Home 622  0.321/0.395/0.502   (0.290/0.355/0.412)   111/111/122    133
1929 Hill 630  0.329/0.402/0.502   (0.298/0.363/0.432)   110/111/116    127
1930  Cub 647  0.327/0.405/0.506   (0.312/0.370/0.464)   105/109/109    118
1931 Hill 567  0.271/0.360/0.350   (0.285/0.344/0.403)    95/105/ 87     91
--
1935  Cub 488  0.292/0.391/0.511   (0.286/0.341/0.407)   102/115/126    140
1936  Cub 480  0.296/0.398/0.590   (0.286/0.345/0.400)   103/115/148    163

Re: #91:

OK... sorry for the delay guys. Huh, it looks so much prettier on the preview screen... when you click submit it goes to that squish font where the 6's look like 5's.

There is a lot of missing data so I won't post totals for Dihigo.

WFerrell 117/100
CMays 119/82
DNewcombe 114/85
George Uhle 105/86
Brandon Backe 88/128 :-) (last year he was 100/128)

Guys, I got a whole page on this pitcher OPS+ stuff.

Submitted for your approval:

George Mullin: 101/100
Al Orth: 100/92
Jack Stivetts: 120/105
Jim Whitney: 112/105
Bob Lemon: 119/82


Jack Stivetts for HoM?

Dihigo prime OPS+ averages 127, ERA+ I prefer to take 110 from Chris' suggested range. He's Stivetts with a longer career, and batting better than pitching (Caruthers was 135/123, an order of magnitude better, and his ERA+ was dragged down by one lousy year at the end.) In but only just, in my view. He also doesn't seem to have combined excellence in both areas in the same season as Caruthers did.

Sorry, withdraw last comment; he pitched well in 1935-36.

One thing to think about is that three years of his prime, 1932-34 (ages 26-28) are left out of all these figures.

Hmmmm. Thinking out loud about Martin D.

The Many Faces of Martin Dihigo:

1a) As an 18 and 19 year old, he's over his head.

1b) Coming of age quickly, he struggles as a rookie, becomes a solid regular at 21, then becomes a very good but not all-time-great hitter from age 21 through 25. Somewhere around age 25, he takes up switch hitting and his numbers plummet. All along, he pitches, but mostly seems average.

2) He goes to VZ from ages 27-29, where his play is undocumented but where he apparently hones his pitching skills to become a star pitcher and hones his switch hitting to the point where...

3) He returns to the states for ages 30 and 31 to post two of his best offensive seasons and pitches well too.

4) At age 32, he goes to Santo Domingo along with Paige and Plant, er Paige and Gibson, has a great year pitching and hittng (per Riley).

5) Plays in Mexico from age 33-42(or longer?). He mostly pitches, being a slightly above-average pitcher when accounting for the quality of Mexico versus the NgLs (as evidenced in post 21 and discussion of Mexico that follows), and ultimately declines and retires around 1947 or so.

THOUGHTS ABOUT HIS HITTING
In Phase 1 he finds success early and becomes a very good player prior to his peak. His peak goes mostly undocumented, then when he returns at age 30, he has two excellent seasons that suggest a typical growth pattern.

In fact, if you plotted his OPS+ this way it looks somewhat typical

75 116 130 133 127 118 91 XXX XXX XXX 140 163

Since we know that the 91 (and part of the 118?) may represent his learning curve for switch hitting, the line that goes through his age-23 127 OPS+ figure and extends through his age-31 163 OPS+ seems very normal based on what we know about career arcs. I know speculatio is dangerous, but it's easy to imagine the above line looking something like this if
a) he'd not tried switch hitting
b) we had data for his peak

75 116 130 133 127 [135 140 160 150 145] 140 163
[The brackets are my speculatory data points.]

Of course I'm using smooth data points in the interest of simplicity, but this kind of progress would be a reasonable fit for the aging model we understand today.

Additionally, his hitting skills remained pretty sharp. With the big influx of NgL players in the early and mid 40s, his hitting rates go from 350/450/550 to 310/420/475, from MVP to mere All-Star, then he finally declines into retirement.

I tend to get caught up in speculation, and so I think we might need some Mexican league translations to better understand his career. Nonetheless, there's enough here to suggest that he was a really good hitter with a broad base of skills.

THOUGHTS ABOUT HIS PITCHING
So from post 50, we know he's not a great pitcher before leaving for VZ. But in VZ where, presumably, the league quality isn't quite as high, he looks studly. Yet when he returns in 1935, he pitches well and becomes an above-average hurler in the souped-up Mexican League of the early-to-mid 40s.

Frankly, the pitching doesn't impress me all that much. He's not Wes Ferrell, closer to Stivetts, I guess, but I'm left thinking a 110 MLE ERA+ might be generous.

IN SUM
OK, so I'm starting to get a clearer picture here.

If I take the value-to-team approach, he's obviously well qualified in the Rogan sense, but I don't know that he's as good of a combined candidate as Rogan because I think Rogan was a better pitcher. If Rogan was 115 OPS+/110 ERA+, Dihigo feels like a 125/103 kind of guy.

If I take the what-would-he-be-like-in-MLB approach, then I don't think he pitches, but I do think he probably turns out looking like an Al Kaline type of player with a good glove, good power, decent average, good OBP. However, I can't justifiably rank him as high as I'd rank Kaline because I'm speculating and I don't have quite enough data to support that assertion.

If I had to vote right now, he's in the top 8 on my ballot, though I'm not certain exactly where yet. Probably nearer Cravath and GVH than Cronin and Waner, but there's time and room for him to move up.

One problem that I see with Dr. Chaleeko's speculative OPS+'s is that he may have still been 'adjusting' to switch hitting. It is possible that he would have worked up to his 1935 and 1936 numbers.

So instead of:
135 140 160 150 145

We may get:
115 130 155 150 145

or something like that again my numbers are just wild ass guesses, but I think he may have still had an adjustment period before getting back to his actual numbers. I just don't see him going from a 91 OPS+ in back to his career arc.

jschmeagol,

I'm sorry, I should have qualified that by saying that in MLB I don't think he would have suddenly adopted switch hitting like he did in the NgL, so he wouldn't have had that adjustment period, thus no 91 OPS+. sorry i wasn't more clear about that.

oh, my bad, I thought you were just saying what we can expect from Venezuela. Part of the misunderstandingis my fault as well.