Often people are shocked by the detail of the Negro League statistical compilations I do, especially when fielding statistics or relatively minor categories like sacrifice hits or hit batsmen are included. This comes mainly from two sources: 1) a basic unawareness (completely understandable) of the nature of baseball journalism, particularly the box score, in the earlier twentieth century; and 2) the legend of the Negro Leagues, which paints them as half-mythical enterprises that took place mostly in the realm of tall tales—an image I will have much more to say about in future posts.
I was first drawn into Negro League research when, on a whim, I looked up a few African-American newspapers, and found that there was vastly more material there than I had dreamed possible after reading Robert W. Peterson, John Holway, James Riley, and others, most of whom stressed the poverty and incompleteness of “objective” statistical data, or claimed that it was completely unknowable. When I followed up by looking into some mainstream daily papers, and realized that in the 1920s and before Negro League games were regularly reported in many of them, with box scores that generally surpassed the quality of the boxes in the black weeklies, it really cinched the deal. Realizing that you could count how many wild pitches Bullet Rogan was charged with, or how many times Jud Wilson was hit by a pitch, or that you could actually compile fielding statistics (thus range factors, fielding percentages, etc.) for the reputedly slick-gloved Dick Lundy, fired the imagination far more than any endlessly repeated anecdote or silly tall tale about how fast Cool Papa Bell was.
Anyway—I thought it would be interesting for those who’ve never actually seen Negro League box scores to occasionally post some scans here, maybe talking a little about the issues inherent in analyzing them. Since I just posted 1916 statistics, here’s a sample of box scores from that year.
First, from the St. Louis Globe-Democrat, a (white) daily paper, which, for every year I’ve researched in the 1910s and 1920s, featured generally excellent box scores of Giants and Stars games, here’s the box score for a July 27, 1916, game between the St. Louis Giants and (western) Cuban Stars:
Sometimes the boxes are a little approximate about innings pitched, which means you can’t in every instance follow what they say precisely. In the account of this game (which I couldn’t include in the image, due to limitations of my editing program), it is said that Pedroso was driven from the mound in the sixth. Together with the notation that Padrón (this would be Juan, the American lefty, not Luis, the Cuban) pitched 2 1/3 innings, this led me to credit Pedroso with 5 2/3 innings pitched, not six, as the box score gives.
Here’s a game account and box score from the Indianapolis Freeman (8-19-1916), for an August 13 game between the Lincoln Giants and (eastern) Cuban Stars (in separate images):
Note the difference between the line score, which gives the Lincolns 8 runs, and the table, which gives them 10 in the totals line—but the individual players’ runs scored only add up to 9! Typical box score fun and games. In this case the discrepancy originated with Cyclone Joe Williams’s walkoff three-run home run. As the Chicago Defender’s account of the game noted, “As only one run was needed to win the game the hit went for a single.” Plus the official score would have been only 8 to 7, and Williams and the other runner would not be credited with runs scored. This was the rule at the time in the major leagues, and many players “lost” home runs as a result, Babe Ruth among them (he really hit 715 home runs).
I have unapologetically played the revisionist in cases like this (this is actually the only one I’ve confronted so far, that I can remember), for these reasons:
1) There was obviously some disagreement about it at the time, as even then the rule was counter-intuitive: thus the confusion in the Freeman’s runs scored column.
2) There was no governing body to rule on such questions, since no Negro League actually existed. There was no “official” score, only the numbers and rulings the newspapers came up with; and they were simply (albeit unevenly) applying common practice.
3) Most importantly, the whole point of recovering Negro League statistics is to characterize as accurately as possible what happened; and Williams hit a three-run home run, not a single. Especially when you consider the smallness of the samples involved, taking away his home run could result in his achievements and abilities being badly misunderstood.
Lastly, I thought I’d post a sample of one of the play-by-play accounts published in both the Defender and Freeman, just to give you an idea of what they looked like. This is from an account of an American Giants/ABCs game in the September 2, 1916, Defender: