archives

Kansas City Monarchs

This tag is associated with 2 posts

Evaluating Negro Leagues Pitcher, Part III: Rogan, Smith, and Williams

[All MLEs updated 7/4/18 to include (a) new 1938 and 1947 data (b) new baserunning-runs estimates(c) new, more objective playing-time estimates]

[Note: All MLEs here updated 1/1/18 due to changes in approach to pitcher batting.]

[Note: Updated 1/20/18 to include adjustment that accounts for general differences in pitcher-batting ability between MLB and Negro Leagues.]

We introduced you in our last two posts to eight of the eleven Negro Leagues honored by the Baseball Hall of Fame and the Hall of Merit. This time we’ll close the loop with Bullet Rogan, Hilton Smith, and Smokey Joe Williams.

[In case you want a reminder of the method we’ve outline, it’s here.]

Bullet Rogan

[Note: Updated 1/14/18. Presence of highly skewed league data caused z-score translations to be appear artificially low. Fixed.]

[Note: Updated 1/23/18 to fix a tiny calculation error that amounts to a change of -0.2 RAA/WAA/WAR.]

Wilbur “Bullet” Rogan is perhaps most famous for, like Martín Dihigo being a double-threat: A great pitcher and a great hitter. While Rogan lacked the versatility of Dihigo, he was probably a better pitcher. Rogan started his career in the US Army, not with a Negro Leagues team. In 1915, he was promoted to the 25th Infantry Wreckers so that he could join their ballclub, which featured future Negro Leagues stars Dobie Moore and Heavy Johnson among others. The Wreckers, stationed in the Pacific, took on all comers, and they beat a lot of PCL teams and pretty much everyone else.

From our perspective in the 21st Century, we might ask why a ballplayer wouldn’t hook on instead with a Negro Leagues team. Part of the answer is that there were no official Negro Leagues at that moment. Instead, blackball was a group of loosely confederated indy teams some of which might travel the country as a pair barnstorming their way to a paycheck or go it alone and take on the local yokels. This probably sounds to you like an unstable business model. Yup. With no central authority, there were no guarantees of payment, or at least prompt payment. That combined with playing multiple games a day in dusty towns you never heard of made Army baseball an attractive option. If you could hack basic training and could stand the hierarchy, you played ball; got paid in full, on time; got room and board; and led a predictable life. In the Wreckers’ case, a predictably sunshine filled life on an island.

As soon as the Negro National League formed in 1920, members of the Wreckers bought their way out of their service commitments and signed on with league teams. In fact, all three of Rogan, Moore, and Johnson were scooped up by J.L.Wilkinson and his Kansas City Monarchs. Actually, Rogan had played briefly with one of Wilkinson’s touring teams in 1917 but had returned to the army shortly after. Anyway, so at age 26, Rogan entered the Negro Leagues, and within two seasons, he was a star. He led the Monarch’s pitching staff as the team rumbled along to several pennants and Negro World Series appearances. In the mid-1930s, just as Satchel Paige joined the team and Hilton Smith emerged as a star, Rogan wound down his career, as did his long-time teammate Andy Cooper. He left behind a stellar 145 ERA+ (1303 innings) and a super 160 OPS+ (1721 PA).

Bullet Rogan
Negro Leagues Stats | Bio
Career: 1915–1937
Destination: NL 1915–1936
Missing data: 1915–1919, 1926–1927, 1929–1932
Honors: Baseball Hall of Fame, Hall of Merit
               PITCHING          |   BATTING   |  TOTAL
YEAR  AGE   IP  RAA   WAA   WAR  |   PA   WAR  |   WAR
=======================================================
1915   21  200   18   2.3   4.2  |   67   0.6  |   4.8
1916   22  210   20   2.6   4.6  |   70   0.7  |   5.2
1917   23  220   21   2.7   4.7  |   73   0.6  |   5.4
1918   24  220   21   2.7   4.7  |   73   0.6  |   5.3
1919   25  240   21   2.7   4.9  |   80   0.7  |   5.6
1920   26  270   22   2.6   5.2  |   90   0.8  |   6.0
1921   27  300   24   2.5   5.6  |  100   0.9  |   6.5
1922   28  280   46   4.7   7.6  |   93   0.8  |   8.4
1923   29  300   56   5.9   8.9  |  100   0.9  |   9.8
1924   30  270   41   4.5   7.2  |   90   0.8  |   8.0
1925   31  260   39   3.9   6.6  |   87   0.8  |   7.4
1926   32  260   34   3.6   6.2  |   87   0.8  |   7.0
1927   33  270   30   3.3   6.0  |   90   0.8  |   6.8
1928   34  210   17   1.8   3.9  |   70   0.7  |   4.6
1929   35  200   17   1.6   3.7  |   67   0.6  |   4.3
1930   36  200   17   1.5   3.6  |   67   0.6  |   4.2
1931   37  180    8   0.8   2.6  |   60   0.6  |   3.2
1932   38  150  - 6  -0.6   1.0  |   50   0.5  |   1.4
1933   39    1    0  -0.1   0.0  |    0   0.0  |   0.0
-------------------------------------------------------
TOTAL     4241  445  49.0  91.3  | 1414  12.6  | 103.9

Hypothetical MLB career rankings (1871–1960)
Innings pitched: 19th
Pitching Wins Above Average: 8th
Pitching Wins Above Replacement: 7th
Total Wins Above Replacement (pitchers only): 5th

Enough of Rogan’s record is missing from his Negro Leagues seasons, that we were a little concerned. For missing-data seasons, we use the pitcher’s career performance rate, but Rogan’s rate was very, very high and caused us to wonder if he was being inflated due to the lack of data. It was high enough that halving the career rate of run prevention still led to what you see above. As more data rolls in, we’ll update our MLE accordingly. In addition, Rogan’s 1922, 1925, and 1926 seasons required us to use the manual override to keep his performance in line with league norms.

There’s not a ton to add to the story. Rogan was an excellent pitcher and great hitter for any batter, not just for a pitcher. You have to guess, though, that all of these great hitting pitchers in the Negro Leagues would probably have been lesser hitters than we translate, simply because they’d get fewer reps, fewer chances to play in the field between starts. When teams barnstormed and operated on shoestring budgets, they had to economize. One way to do so was to bring as few players as possible on the road. That meant rosters of 13 or so for traveling. Which, in turn, meant that pitchers had to be two-way players in order to spell on another and spell injured or tired position players.

We’ll see as we delve deeper into the candidate pool in posts down the line detailing other pitching candidates that these guys couldn’t all hit. But many could, and in this way, the Negro Leagues were 20 to 30 years behind the majors which saw a sharp reduction in pitcher batting as rosters expanded to cover longer schedules and specialization began to increase.

We’ll be providing an MLE down the pike for Rogan as if he had never pitched but only been a position player.

Fixing the issue with the skewed league data bumped Rogan up considerably because some seasons that showed up near average became above average.

Hilton Smith

[Note: Updated 1/14/18. Presence of highly skewed league data caused z-score translations to be appear artificially low. Fixed.]

[Updated 4/3/18 with additional 1937 data.]

For many years, Hilton Smith was most famous for following Satchel Paige to the mound. The great Paige, having been advertised near and far as pitching on a given day would go three innings, and Smith would finish things off. At least, when the Kansas City Monarchs traveled. In league games, Smith was more likely to start and finish his own games.

Other circumstances conspired to reduce Smith’s visibility. Rather than rise up through main Negro Leagues, he got his start in the Negro Southern League, which was major only in 1932 as a haven for teams bailing on the failing major leagues. He was 25 and stuck with his Monroe teammates for a couple more years then was recruited to play for the semipro Bismarck super team that beat back all comers in the mid-1930s. When he joined the Monarchs for 1936, they were mostly a barnstorming team with relatively few documented games against top rivals, and he became Satch’s shadow. Paige left the Monarchs for the majors in 1948. Smith stayed behind and then retired after the season, leaving behind appearances in six East-West All-Star Games in his wake.

Hilton Smith
Negro Leagues Stats | Bio
Career: 1932–1948
Destination: NL 1932–1948
Missing data: 1933–1936, 1948
Honors: Baseball Hall of Fame
               PITCHING          |   BATTING   |  TOTAL
YEAR  AGE   IP  RAA   WAA   WAR  |   PA   WAR  |   WAR
========================================================
1932   25  180   12   1.3   3.1  |   60   0.3  |   3.4
1933   26  210   13   1.5   3.6  |   70   0.4  |   4.0
1934   27  200   12   1.2   3.3  |   67   0.4  |   3.7
1935   28  220   13   1.3   3.6  |   73   0.4  |   4.0
1936   29  270   15   1.5   4.3  |   90   0.5  |   4.8
1937   30  260    6   0.7   3.3  |   87   0.5  |   3.8
1938   31  260   16   1.7   4.4  |   87   0.5  |   4.9
1939   32  260   24   2.6   5.2  |   87   0.5  |   5.7
1940   33  270   34   3.8   6.4  |   90   0.5  |   7.0
1941   34  210   24   2.8   4.8  |   70   0.4  |   5.2
1942   35  180    3   0.3   2.1  |   60   0.3  |   2.4
1943   36  160   19   2.2   3.7  |   53   0.2  |   4.0
1944   37   20    2   0.2   0.4  |    7   0.0  |   0.5
1945   38  180   13   1.4   3.3  |   60   0.3  |   3.5
1946   39  180    8   0.9   2.7  |   60   0.3  |   3.0
1947   40  180  -13  -1.3   0.5  |   60   0.3  |   0.9
1948   41   20    0   0.0   0.2  |    7   0.0  |   0.2
-------------------------------------------------------
TOTAL     3260  201  22.2  54.9  | 1088   6.1  |  61.0

Hypothetical MLB career rankings (1871–1960)
Innings pitched: 51st
Pitching Wins Above Average: t-56th
Pitching Wins Above Replacement: 40th
Total Wins Above Replacement (pitchers only): 30th

We made the manual adjustment for Smith’s 1943 season run-prevention rates to keep him in line with league norms.

A word about 1932–1936. The Lester/Clark HOF study included his stats for Monroe in the 1932 Negro Southern League in Shades of Glory. That Monroe team took on the Crawfords for a Negro Championship. They report nothing else until 1937. Riley and other bio sources indicate that Smith went to Bismarck with Satch and others in 1934/1935 somewhere, then from there became a Monarch in 1936. Given that in 1937 and 1939, Smith appears to be a finished product at age 30, it doesn’t seem unreasonable that we include 1932–1936 in our MLE. So we did at his known career average.

Smith is another good-hitting Negro Leagues pitcher who adds a lot of value that way. I’m not sure what’s up with 1944. There’s some mention of an injury in 1943 in some sources, though I see no evidence of it in the stats. But in 1944, he does appear to have been unable to start as often as usual, so maybe he hurt himself in winter ball, affecting his summer performance. It seems to have affected his durability more than his effectiveness.

It is also possible that Smith spent part of 1945 in the war, though we can’t find corroboration. He is alleged to have tipped off J.L. Wilkinson, leading to Jackie Robinson’s signing with the Monarchs immediately after his discharge.

Smokey Joe Williams

[Note: Updated 1/1/18 to fix transcription errors in the mid-teens.]

[Note: Updated 1/6/18 to correct calculations for Joe’s first two and last years.]

[Note: Updated 1/14/18. Presence of highly skewed league data caused z-score translations to be appear artificially low. Fixed.]

In The New Bill James Historical Baseball Abstract, Bill ranks Joe Williams 52nd among the greatest players in baseball history. I can buy that, or even a better ranking. Williams was an early link in a long historical chain of Texas fireballers that includes “The Ryan Express” and the “The Rocket.” He must have been extremely intimidating back in the 1900s and 1910s. He stood 6’3” tall, with a chiseled face that bore the high cheek bones, angular nose, and strong chin gifted him from the Comanche heritage of one of his parents. That great fastball, likened by a promoter to a pebble blown by a storm, must have erupted from his hand a lot closer to home plate than most hitters were used to during a time when the average American male was an inch and a half shorter back then than today. Cyclone Joe, as he was called earlier in his career, gained fame for his prodigious strikeout totals. His career variously included no-hitters and a 27-strikeout performance (at night, in 12 innings) among other gems. He beat PCL teams by the bushel in a California swing, went 20-7 against white major league teams (8-3 documented), and had a 140 ERA+ in Cuba. His ERA+ of 149 trails only Dave Brown (150 ERA+) among hurlers with 1000 innings in the Negro Leagues Database, and only Brown and Satch (193 ERA+) among players anywhere near 1000 innings. His 1240 strikeouts rank first in the Database. His 1862 innings are third in the Database, his 132 victories are third, his 196 complete games are third, and his 20 shutouts rank fifth. He could bring it.

Smokey Joe Williams
Negro Leagues Stats | Bio
Career: 1907–1932
Destination: NL 1907-1930
Missing Data: 1907–1908, 1925–1927, 1929
Honors: Baseball Hall of Fame, Hall of Merit
               PITCHING          |   BATTING   |  TOTAL
YEAR  AGE   IP   RAA  WAA   WAR  |   PA   WAR  |   WAR
=======================================================
1907   21  100   12   1.6   2.5  |   33   0.1  |   2.7
1908   22  200   22   3.1   4.9  |   67   0.3  |   5.2
1909   23  250    8   1.0   3.4  |   83   0.4  |   3.8
1910   24  300   17   2.0   4.9  |  100   0.4  |   5.4
1911   25  310   32   3.6   6.7  |  103   0.5  |   7.2
1912   26  300   52   5.7   8.7  |  100   0.4  |   9.1
1913   27  300   25   2.9   5.9  |  100   0.5  |   6.4
1914   28  270   34   4.2   6.7  |   90   0.5  |   7.2
1915   29  250   18   2.3   4.7  |   83   0.4  |   5.1
1916   30  250   31   4.2   6.5  |   83   0.4  |   6.9
1917   31  300   39   5.2   7.9  |  100   0.4  |   8.3
1918   32  270   28   3.7   6.2  |   90   0.3  |   6.5
1919   33  280   20   2.5   5.1  |   93   0.4  |   5.5
1920   34  300   30   3.6   6.5  |  100   0.4  |   6.9
1921   35  300   35   3.8   6.9  |  100   0.5  |   7.3
1922   36  250   41   4.2   6.7  |   83   0.5  |   7.2
1923   37  240   10   1.0   3.5  |   80   0.4  |   3.9
1924   38  200   29   3.1   5.1  |   67   0.3  |   5.4
1925   39  180   19   1.8   3.7  |   60   0.3  |   4.0
1926   40  150   18   1.9   3.4  |   50   0.3  |   3.7
1927   41  130   16   1.7   3.0  |   25   0.2  |   3.2
1928   42   50    4   0.4   0.9  |   18   0.1  |   1.0
1929   43   20    3   0.3   0.5  |    7   0.0  |   0.5
1930   44   10    2   0.2   0.3  |   17   0.1  |   0.4
-------------------------------------------------------
TOTAL     5210  545  63.7 114.6  | 1732   8.2  | 122.8

Hypothetical MLB career rankings (1871–1960)
Innings pitched: 4th
Pitching Wins Above Average: 6th
Pitching Wins Above Replacement: 3rd
Total Wins Above Replacement (pitchers only): 3rd

We employed the manual override on 1912, 1914, 1921, and 1922 to keep Williams’ MLEs in line with league norms.

In terms of their relative standings, he is precisely to Paige as Pete Alexander is to Walter Johnson. The thing about Johnson and Paige is that everyone in the greatest-ever conversation gets compared to them. Roger Clemens? How does he stack up to the Big Train? Joe Williams? How does he compare with Satchel? That’s not a dis on anyone, either. It’s simply an acknowledgment of how great those guys were. Similarly, whoever is juxtaposed to them in any serious discussion of GOATedness (that is, Greatest of All Time) must be an awfully good pitcher to even merit the comparison. Williams was a really great pitcher. Yet, he needs a strong bat to get by Paige. In terms of measuring his pitching performance, Williams finishes behind Paige, but he needs 400 more innings to do it.

That said, if you were GM for a big league club, and someone said they could clone Smokey Joe Williams and have him ready to pitch for you starting next year, you’d do it quicker than you can say medical ethics. He was the towering figure among moundsmen in the early Negro Leagues era, and anyone would take the kind of peak we are estimating in his MLEs.

Note: When we adjusted our league stats to remove pitchers with very low innings and very high RA9, several of Williams’ seasons drastically improved, and his MLE value along with them. He now looks to be right on par with Satch.

And now, we’ve reached the end of our walk to the mound to meet with our Negro League Hall of Fame/Merit hurlers. We hope you’ve enjoyed getting to know them a little and learning about some of baseball’s best players who are virtually unknown historical figures. Often we talk about underrated players in MLB. Tony Gwynn and Tim Raines were the subject of such talk while I was growing up, and Miller and I write about underrated players like Dwight Evans or Bobby Grich all the time. But those guys are better known by many orders of magnitude compared to anyone in the Negro Leagues, with the possible exception of Satchel Paige. It’s more likely that the average fan on the street knows who Sean Casey is than Buck Leonard. Or Josh Collmenter than Josh Gibson (and certainly Kirk Gibson over Josh Gibson). Every Negro Leagues player is underrated, so we hope we’re able to give them a little spotlight time. If you’re interested to learn more, we recommend not only the amazing Negro Leagues Database and the equally amazing SABR Bioproject, but also books such as Shades of Glory and The Biographical Encyclopedia of the Negro Baseball Leagues.

Next time out, we’re going to document for you our MLE method for position players, then we’ll dive into the Hall of Fame/Merit players at each position en route to a first sweep through the Negro Leagues.

Advertisements

Evaluating Negro Leagues Pitchers, Part I: Brown, Cooper, Day, Dihigo

[All MLEs updated 7/4/18 to include (a) new 1938 and 1947 data (b) new baserunning-runs estimates(c) new, more objective playing-time estimates]

[Note: All MLEs here updated 1/1/18 due to changes in approach to pitcher batting.]

[Note: Updated 1/20/18 to include adjustment that accounts for general differences in pitcher-batting ability between MLB and Negro Leagues.]

We recently described to you our method for creating Major League Equivalencies (MLEs) for Negro Leagues pitchers. Today we start unveiling the results. This is the first of three posts in which we’ll share MLEs for Hall of Fame and Hall of Merit Negro Leagues pitchers in alphabetical order.

Today we’re going to cover Ray Brown, Andy Cooper, Leon Day, and Martín Dihigo.

Each MLE should not be too literally interpreted. There’s a lot of moving parts, and, yes, human error creeps in sometimes. We suggest looking first at the player’s career numbers than at the seasonals, the latter of which are more volatile and more likely to look funny to you. We’ll include information about some of the thornier issues in each man’s record along the way as well as provide some sense of what these players would like compared to MLB players. The reason for the latter is not to demean their actual performance in the leagues they played in, but rather to give you a sense of what their performance reminds us of and as a sanity check to be sure that we’re doing our job correctly. We invite you to tell us what you think in the comments so we can refine these estimates.

Ray Brown

[Note: Updated 1/14/18. Presence of highly skewed league data caused z-score translations to be appear artificially low. Fixed.]

Ray Brown was a durable righty in the 1930s and 1940s whose record on the Negro Leagues Database currently stands at 111-37 (.750) and a 145 ERA+ that sits eleventh among pitchers with at least 200 innings in the database. His 35.2 WAR rank second in among hurlers and his 1310 innings place ninth. Brown threw the kitchen sink at hitters and was most well known for his curveball. The mainstay of the dynastic Homestead Grays, he was known as a tough competitor but pleasant and quiet. He was also a favorite in Cuba and Puerto Rico where he racked up impressive records and was known as Jabao (freckled one). Not much is known about him personally because he died in 1965, just before intensive research got underway to interview Negro Leagues players before they passed on.

Ray Brown
Negro Leagues Stats | Minor Leagues Stats | Bio
Career: 1932–1951
Destination: NL 1932–1951
Missing data: 1950–1953, any Cuban or Puerto Rican winter seasons
Honors: Baseball Hall of Fame, Hall of Merit
               PITCHING          |  BATTING    |  TOTAL
YEAR  AGE   IP  RAA   WAA   WAR  |  PA    WAR  |   WAR
========================================================
1932   24  210  - 7  -0.7   1.5  |   70   0.4  |   1.9
1933   25  240  - 2  -0.3   2.1  |   80   0.5  |   2.6
1934   26  230  -14  -1.4   1.0  |   77   0.5  |   1.5
1935   27  220   25   2.6   4.8  |   73   0.4  |   5.2
1936   28  270   33   3.5   6.2  |   90   0.6  |   6.8
1937   29  260    7   0.8   3.4  |   87   0.5  |   4.0
1938   30  260    9   1.0   3.6  |   87   0.6  |   4.2
1939   31  260   21   2.3   4.9  |   87   0.5  |   5.4
1940   32  270   27   2.9   5.7  |   90   0.6  |   6.2
1941   33  250   29   3.3   5.7  |   83   0.5  |   6.3
1942   34  250   22   2.7   5.1  |   83   0.5  |   5.6
1943   35  160   22   2.6   4.1  |   53   0.3  |   4.4
1944   36  260    3   0.3   3.0  |   87   0.4  |   3.4
1945   37  210  - 3  -0.4   1.8  |   70   0.3  |   2.1
1946   38  200  -15  -1.7   0.3  |   67   0.4  |   0.7
1947   39  160  - 2  -0.2   1.4  |   53   0.3  |   1.7
1948   40  110    3   0.3   1.4  |   37   0.2  |   1.6
1949   41   40    2   0.2   0.9  |   23   0.1  |   1.1
1950   42   40    1   0.2   0.6  |   13   0.1  |   0.6
1951   43   10    1   0.1   0.2  |    3   0.0  |   0.2
-------------------------------------------------------
TOTAL     3940  160  18.0  58.0  | 1314   7.5  |  65.4

Hypothetical MLB career rankings (1871–1960)
Innings pitched: 26th
Pitching Wins Above Average: 78th
Pitching Wins Above Replacement: 32nd
Total Wins Above Replacement (pitchers only): t-20th

At this time, Brown’s best MLB comps are Red Ruffing and Burleigh Grimes. They also had the same strengths and weaknesses: Long careers, good-to-great great bats for a moundsman, and pretty good on the bump, itself, but not exactly Lefty Grove either. Ruffing is the upside, Grimes is the downside. Though he looks like more Ruffing than Grimes, lacking Old Stubblebeard’s tendency to ying and yang between good and poor seasons, and being at least twice as good a hitter. More like Ruffing in that sense as well.

I expected Ray Brown to come out looking like a higher-tier pitcher than he has. But there’s a couple things of interest here that appear to militate against that:

  1. Brown pitched in front of several teams with HUGE DRA totals, which he is debited for (though we cap at +/- 0.50 runs per nine innings).
  2. Because we use for our innings estimates the workload of a typical MLB pitcher, Brown may show fewer MLE innings than he might have racked up in MLB. He doesn’t appear to have much of an injury history and may well have been able to shoulder an ace’s load.
  3. He’s getting hurt by his 1934 season, which was short 22.33 IP and bad (60 ERA+). We haven’t manually adjusted for that here, but even just making him a league average pitcher for that year would add 4 WAR to his career MLE total. It’s possible that step is in order to bring him closer to the historical consensus.
  4. Brown was excellent in winter league play but because that data (especially the league-wide pitching data) is not available yet on the Negro Leagues Database, we haven’t included it. Brown seems unlikely to be elected in our first round of HoME Negro Leagues elections, but he will surely make it into the subsequent rounds. It may be that taking an independent dive into his Cuban seasons could help, though Puerto Rican information remains sketchy at this time.

 Andy Cooper

[Note: Updated 1/1/18 to fix a transcription error in 1925 (formerly 8.0 WAR, not corrected to 5.4 WAR).]

[Note: Updated 1/14/18. Presence of highly skewed league data caused z-score translations to be appear artificially low. Fixed.]

One of a pair of brothers to pitch in the Negro Leagues, Cooper had the profile of a big-league lefty: a wide repertoire of pitches all of which had wrinkle to them, thrown with pinpoint control, at a variety of speeds. And, of course, a good move to first. In other words, a classic finesse pitcher. Stylistically, players like Jimmy Key and Mark Buehrle spring instantly to mind. He made his bones with the Detroit Stars in the 1920s Negro National League but eventually signed on with the famous Kansas City Monarchs, forming an impressive trio with Hilton Smith and Satchel Paige at one juncture. He also managed the team.

Andy Cooper
Negro Leagues Stats | Bio
Career: 1920–1939
Destination: NL 1920–1936
Missing data: 1926–1927, 1929–1932
Honors: Baseball Hall of Fame
                PITCHING         |   BATTING   |  TOTAL
YEAR  AGE   IP  RAA   WAA   WAR  |   PA   WAR  |   WAR
=======================================================
1920   24  140    5   0.6   1.9  |   47  -0.4  |   1.5
1921   25  220    6   0.6   2.9  |   73  -0.4  |   2.4
1922   26  250   46   4.7   7.2  |   83  -0.3  |   6.9
1923   27  210   14   1.4   3.5  |   70  -0.4  |   3.1
1924   28  200   22   2.4   4.4  |   67  -0.5  |   3.9
1925   29  210   32   3.2   5.3  |   70  -0.3  |   5.1
1926   30  200   30   3.2   5.2  |   67  -0.4  |   4.8
1927   31  210   23   2.4   4.5  |   70  -0.4  |   4.2
1928   32  210   12   1.2   3.4  |   70  -0.3  |   3.1
1929   33  200   18   1.7   3.8  |   67  -0.2  |   3.6
1930   34  200   20   1.8   4.0  |   67  -0.2  |   3.7
1931   35  220   24   2.6   4.8  |   73  -0.3  |   4.5
1932   36  210   22   2.4   4.5  |   70  -0.3  |   4.2
1933   37  210   21   2.5   4.5  |   70  -0.4  |   4.1
1934   38  170   21   2.2   4.0  |   57  -0.2  |   3.8
1935   39   30    4   0.7   0.7  |   10   0.0  |   0.6
1936   40   10    2   0.3   0.3  |    3   0.0  |   0.3
-------------------------------------------------------
TOTAL     3100  320  33.5  65.0  | 1034  -4.9  |  60.0

Hypothetical MLB career rankings (1871–1960)

Innings pitched: 65th
Pitching Wins Above Average: 26th
Pitching Wins Above Replacement: 22nd
Total Wins Above Replacement (pitchers only): 32nd

Cooper’s 1925 translated RA9 is considerably lower than the qualified league leader in the NL. So we adjust him to 10% above the leader so we aren’t overshooting. His last two seasons also translated better than the leader, and even though we are nipping the innings down from an initial projection, we are still capping his RA9.

Cooper might look better in this MLE than he does on the Negro Leagues database. Part of the reason could be that our protocol substitutes the pitcher’s MLE-career average performance in instances where we have no data available to us. Much of the heart of Cooper’s career is missing, so he either currently looks worse on the Negro Leagues Database than he ought to, or he looks better in our MLE. We won’t know until we know. So the combination of lopping off a couple of lesser end-of-career seasons, giving him MLE-career average performance in missing seasons, and giving him more innings in his salad days makes him come forward a bit.

On the other hand, I translated him in 2005 at the Hall of Merit for 500 more innings than this MLE and a .560 win% (223-176). We didn’t have team and league stats at that time, nor did we have WAR. But if .560 represents his true performance level, then that’s 22 WAA in shorthand (10 runs per win, 22 wins or so above .500). So I’m coming in higher now than previously, but by enough to be covered by things like park factors and team defense. Another poster on BTF chipped in that they figured 223 Win Shares, if you remember them from early 2000s. Divide Win Shares by 3 for wins, and that’s 74, and maybe nip off some wins for sub-replacement performance since Win Shares doesn’t really deal with replacement much, if at all, and you’re in the same neighborhood we’ve arrived at here.

Overall, Andy Cooper has long particularly reminded me of Andy Pettitte: Long-career lefties who didn’t have great fastballs, who played on the dominant team of their times (the Monarchs being the Yanks of the Negro Leagues), who weren’t notably durable in-season but who generally stayed in the rotation, who had a big year or three, who were generally not All-Star type pitchers but nonetheless were above average for a long time in run prevention, and who were not typically the ace of their staff. Mark Buehrle might also fit this description. We’ve projected Cooper as a #3 starter for most of his career based on the number of starts he had on his real teams, but on virtually any other team he would have been a solid number two man. In that way, we could be coming in low on him bulk-wise.

Two final notes. First, some of the biographical data indicates that Andy Cooper might have hurt his arm in late 1930, though that’s contradicted by his participation and performance in the California Winter League that winter. Still, I hedged on 1931 and took his innings down a bit. Second, Cooper had an excellent record during a brief run as the Monarch’s manager in the late 1930s, winning three pennants.

Leon Day

[Note: Day’s 1952 MLE updated on 1/1/18 to correct a transcription error. It previously read 2.2 Total WAR and is now, correctly, listed at 1.5.]

[Note: Updated 1/6/18 to correct STDEV information for seasons in late 1930s to include all pitchers, not just ERA qualifiers; to use the manual adjustment on his 1943 season.]

[Note: All MLEs here updated 1/13/18 to include newly released data for 1946, which add about two WAA/WAR to his pitching totals; also updated batting for consistency of approach.]

[Note: Updated 1/14/18. Presence of highly skewed league data caused z-score translations to be appear artificially low. Fixed.]

[Note: Updated 1/23/18 to correct a tiny calculation error that resulted in +0.3 WAA/WAR and +3 RAA.]

Day struck out lots of hitters and was selected for seven East-West All-Star Games as the main man in the Newark Eagles’ rotation. In 586 innings, he struck out 402 hitters in the Negro Leagues, a rate of 166 per year, good for 9th among documented Negro Leagues seasons. He served two years during World War II, spent a couple seasons in Mexico, and in his final years toiled in the minor leagues.

Leon Day
Negro Leagues Stats | Minor Leagues Stats | Bio
Career: 1934–1953
Destination: NL 1935–1953
Missing Data: 1949, 1950
Honors: Baseball Hall of Fame, Hall of Merit
               PITCHING          |  BATTING    |  TOTAL
YEAR  AGE  IP   RAA   WAA   WAR  |   PA   WAR  |   WAR
=======================================================
1935   18   50    5   0.5   1.0  |   17   0.1  |   1.1
1936   19  100    9   0.9   1.9  |   33   0.2  |   2.1
1937   20  170  - 3  -0.3   1.4  |   57   0.3  |   1.7
1938   21  140  - 1  -0.1   1.3  |   47   0.3  |   1.6
1939   22  260   20   2.2   4.8  |   87   0.5  |   5.3
1940   23  230   20   2.2   4.5  |   77   0.4  |   4.9
1941   24  210   12   1.3   3.4  |   70   0.4  |   3.8
1942   25  180   27   3.3   5.0  |   60   0.3  |   5.4
1943   26  220   26   3.1   5.2  |   73   0.3  |   5.5
1944   27                        |             |
1945   28                        |             |
1946   29  260    4   0.4   3.0  |   87   0.5  |   3.5
1947   30  180  - 8  -0.8   1.1  |   60   0.3  |   1.4
1948   31  210  - 9  -1.1   1.2  |   70   0.4  |   1.6
1949   32  260    0   0.0   2.7  |   87   0.5  |   3.2
1950   33  160    2   0.2   1.8  |   53   0.3  |   2.1
1951   34   40    1   0.1   0.5  |   13   0.1  |   0.5
1952   35  160  - 6  -0.7   0.9  |   53   0.3  |   1.2
1953   36   30  - 5  -0.5  -0.2  |   10   0.1  |  -0.1
-------------------------------------------------------
TOTAL     2860   93  10.9  39.8  |  954   5.3  |  45.1

Hypothetical MLB career rankings (1871–1960)

Innings pitched: 90th
Pitching Wins Above Average: t-151th
Pitching Wins Above Replacement: 92nd 
Total Wins Above Replacement (pitchers only): 74th

Day hurt his arm in the offseason of 1937 and missed some time in 1938, which we’ve accounted for. He also missed 2 years and one September to the war. The tough spot in this MLE is 1937. In his SABR bio, Day and the author claim it as an outstanding year, something like 13-0. But on the Negro Leagues Database, we only have a few starts from him. Those starts are good not great, and when translated are roughly average. It is entirely possible that the Database simply doesn’t yet have his complete numbers for 1937. Day’s bio also says that he had a dead arm in 1946 after a two-year army layoff. Despite that, he pitched the whole year, and his stats that year from the Negro Leagues Database suggest he was in great form. His performance in Mexico in 1947 and 1948 certainly suggests he had lost something. The apparent rally in 1949 and 1950 is probably only because our protocol uses his career average due to no data for those seasons. He pitched reasonably well in the minors at ages 34 and 35, but at 36 was very clearly done.

As a hitter, he was excellent. Not as good as Ray Brown, but really good. In fact, in 1943, he spent a little time with the Philadelphia Stars in addition to the Eagles, but he didn’t pitch once for them. He was too busy in their lineup at second base.

These MLEs suggest that value-wise, he comps to a pitcher such as Howard Ehmke, only with a good bat. Or perhaps Red Lucas who had a great bat. Style-wise, not so much. Generally, that feels too flat for me, but until we see the rest of the data on him, we can’t say with certainty. But now that we have the 1946 data, we are missing only two summer seasons of information and whatever in-season data is not yet available. We are closing in on the point where we have a high degree of confidence that Day was not a Hall level pitcher.

Martín Dihigo

[Note: Updated 1/14/18. Presence of highly skewed league data caused z-score translations to be appear artificially low. Fixed.]

Last licks today for Martín Dihigo (pronounced Mar-TEEN DEE-go), a most interesting case. He’s famous for playing all the hell over the diamond, but he also had a substantial pitching career, and a good one. His evolution to starting pitcher, however, was slow. From 1923 (age 18) through 1931 (age 26), his appearances at pitcher amounted to 17% of his total games played, and he appears to have been used as a #4/5 or swing starter since he was incredibly useful elsewhere on the diamond. Then there’s a three-year blank spot where he toured with a team out of the Dominican, with no stats available. When he returned to the Negro Leagues in 1935–1936, he pitched in almost 30% of his games played, still holding down a regular job as a position player too. Beginning in 1937 he went down Mexico way, and in La Liga, he pitched in 40+% of his games, and, yes, continued to play the field as a regular until the very end. Which means that there’s two ways we can approach Dihigo’s career with MLEs. The first is that he was a pitcher all the way and would have followed a typical pitcher’s path in terms of workload. The other is just to run the numbers as is.

We’re presenting both versions so you can get an idea of the spread of potential we’re talking about here.

Martín Dihigo
Negro Leagues Stats | Bio
Typical pitching arc
Career: 1924–1947
Destination: NL 1924–1947
Missing data: 1929, 1932–1934
Honors: Baseball Hall of Fame, Hall of Merit, Cuban Baseball Hall of Fame, Mexican Baseball Hall of Fame
               PITCHING          |   BATTING   |  TOTAL
YEAR  AGE   IP   RAA   WAA   WAR |   PA   WAR  |   WAR
========================================================
1924   19   50    6   0.6   1.1  |   17   0.1  |   1.2
1925   20   80    5   0.4   1.3  |   27   0.2  |   1.5
1926   21  100  - 7  -0.7   0.4  |   33   0.3  |   0.6
1927   22  200  - 4  -0.4   1.7  |   67   0.6  |   2.3
1928   23   50  -14  -1.3  -0.8  |   17   0.2  |  -0.6
1929   24  180   12   1.1   3.0  |   60   0.5  |   3.5
1930   25  200   44   4.0   6.1  |   67   0.5  |   6.6
1931   26  220   15   1.6   3.9  |   73   0.6  |   4.5
1932   27  210   16   1.7   3.8  |   70   0.6  |   4.5
1933   28  210   12   1.4   3.5  |   70   0.6  |   4.1
1934   29  200   15   1.5   3.6  |   67   0.6  |   4.2
1935   30  190   24   2.5   4.4  |   63   0.5  |   4.9
1936   31  200   21   2.2   4.2  |   67   0.6  |   4.8
1937   32  260   23   2.5   5.1  |   87   0.8  |   5.9
1938   33  260   39   4.4   6.9  |   87   0.8  |   7.8
1939   34  260   19   2.0   4.7  |   87   0.8  |   5.5
1940   35  190    8   0.9   2.8  |   63   0.6  |   3.4
1941   36  190    8   0.9   2.8  |   63   0.6  |   3.4
1942   37  250    9   1.0   3.5  |   83   0.8  |   4.3
1943   38  220   33   4.0   6.1  |   73   0.4  |   6.5
1944   39  230   28   3.2   5.5  |   77   0.3  |   5.8
1945   40  210    3   0.3   2.4  |   70   0.6  |   3.0
1946   41  130  -25  -2.5  -1.2  |   43   0.2  |  -0.9
1947   42   50    7   0.7   1.2  |   17   0.4  |   1.6
-------------------------------------------------------
TOTAL     4340  297  32.2  75.9  | 1448  12.4  |  88.3

Hypothetical MLB career rankings (1871–1960)
Innings pitched: 17th
Pitching Wins Above Average: 28th
Pitching Wins Above Replacement: 13th
Total Wins Above Replacement (pitchers only): 8th

In both the versions we’re presenting, Dihigo’s 1930 translated RA9 is considerably lower than the qualified league leader in the NL. So we adjust him to 10% above the leader for reasonableness.

Worth noting that in this and the following MLE, we have given Dihigo 50 innings in 1928 because, well, he stank it up. Seems unlikely that he would have been given a long leash while allowing that many runs.

Here we have a player with a good if not phenomenal peak plus lots of bulk, and, of course, an excellent bat. This version of Dihigo looks like a little peakier version of Ted Lyons but with three times the bat. Heckuva package.

Martín Dihigo
As-is estimate
               PITCHING          |  BATTING    |  TOTAL
YEAR  AGE  IP   RAA   WAA   WAR  |  PA    WAR  |   WAR
=======================================================
1924   19   50    6   0.6   1.1  |   17   0.1  |   1.2
1925   20   80    5   0.4   1.3  |   27   0.2  |   1.5
1926   21  100  - 7  -0.7   0.4  |   33   0.3  |   0.6
1927   22  150  - 3  -0.3   1.3  |   50   0.4  |   1.7
1928   23   50  -14  -1.3  -0.8  |   17   0.1  |  -0.6
1929   24  150    9   0.9   2.4  |   50   0.4  |   2.9
1930   25  150   33   3.0   4.5  |   50   0.4  |   4.9
1931   26  140    9   1.0   2.4  |   47   0.4  |   2.8
1932   27  150   11   1.1   2.7  |   50   0.4  |   3.1
1933   28  130    8   0.9   2.2  |   43   0.3  |   2.5
1934   29  140   11   1.1   2.5  |   47   0.4  |   2.9
1935   30  150   18   1.9   3.4  |   50   0.4  |   3.8
1936   31  170   18   1.8   3.5  |   57   0.5  |   4.0
1937   32  260   23   2.5   5.1  |   87   1.7  |   5.9
1938   33  260   39   4.4   6.9  |   87   1.7  |   7.7
1939   34  260   19   2.0   4.7  |   87   1.7  |   5.4
1940   35  190    8   0.9   2.8  |   63   0.6  |   3.4
1941   36  190    8   0.9   2.8  |   63   0.6  |   3.4
1942   37  250    9   1.0   3.5  |   83   1.7  |   4.2
1943   38  220   33   4.0   6.1  |   73   0.4  |   6.5
1944   39  230   28   3.2   5.5  |   77   0.3  |   5.8
1945   40  210   12   1.3   3.5  |   70   0.6  |   4.1
1946   41  130  -25  -2.5  -1.2  |   43   0.2  |  -0.9
1947   42   60    8   0.9   1.5  |   20   0.4  |   1.9
-------------------------------------------------------
TOTAL     3870  267  29.1  67.9  | 1291  10.4  |  78.4

Hypothetical MLB career rankings (1871–1960)
Innings pitched: 28th
Pitching Wins Above Average: t-34th
Pitching Wins Above Replacement: 18th
Total Wins Above Replacement (pitchers only): 11th

Still kind of like Ted Lyons, only a little flatter peak this time. Same great bat as before, just in slightly less playing time.

Again, the difference here is all in the innings pitched. In the first version, we see a rough estimate of what he might have done with a more typical MLB career arc as a pitcher. The second version is much less interpretative and shows us how his pitching career was bifurcated between his American (through 1936) and Mexican (after 1936) experiences. Either way, he appears very strong, and it’ll be very interesting to see how his MLEs as a position player come out by comparison.

OK, that’s it for part one. Next time out, we’ll have a go at messers Foster, Foster, Mendéz, and Paige. And if you’re worried that you don’t see enough peak value in these performers, wait til part two. You won’t be disappointed.

Institutional History

Advertisements
%d bloggers like this: