Chadwick’s Choice: The Origin of the Batting Average
Did you know that slugging average is older than the batting average, and was tossed aside in favor of it? And if so, do you know why? I did not, until I came upon Henry Chadwick’s “The True Test of Batting,” in The Ball Players’ Chronicle of September 19, 1867. (I was rummaging through old newspapers, looking for something else in another early baseball weekly; more on that soon.) Chadwick’s article is a genuine crossroads in the history of baseball statistics.
Bases on balls were still uncommon events, having been introduced for the 1864 season, and no one thought of them as batters’ achievements, nor would they for decades to come. So the need for an on base average was not evident. Chadwick had already posited a primitive version of the slugging percentage, with total bases divided by number of games; change the denominator from games to at bats and you have today’s slugging percentage—which, incidentally, was not accepted by the National League as an official statistic until 1923 and the American until 1946. Chadwick’s “total bases average” represented the game’s first attempt at a weighted average—a huge conceptual leap forward from, first, counting, and next, averaging. The weighted average is in fact the cornerstone of today’s statistical innovations.
Chadwick’s bias against the long ball was in large measure responsible for the game that evolved. What he valued most in the early days was the low scoring game marked by brilliant fielding. In the early annual guides, he listed all the “notable” games between significant teams—i.e., those in which the winner scored fewer than ten runs!
What I did not recognize until now was that the triumph of the batting average was not merely a product of Chadwick’s preference for the scientific style over the brutish slugging. It was his recognition that most runs were scored through some combination of errors (muffs), which were easily counted, and misplays, which were not. In the 1860s a single might easily become a tainted extra-base hit, indistinguishable in the box score from a legitimate one. So in discarding the old practice of crediting batters with only Outs (hit into or run into) and Runs, Chadwick held—and correctly, given the state of the early game—that times reached base on undeniably safe hits was a superior measure to the bases gathered on that hit. Here is his reasoning, with spelling unaltered but paragraph breaks added to what appeared originally as a single block of text.
The True Test of Batting
Up to the present year, and, in fact, up to the inaugural number of The Chronicle, the only recorded test of the skill of a batsman in a match was the number of outs and runs in the score of the game. In The Chronicle, however, the plan of estimating the batting by the number of times a batsman made his base by clean hits was introduced, and now this plan, varied by taking in the number of bases secured on hits, has become general, all the daily press reporting the game having accepted it.
Our plan of adding to the score of outs and runs the number of times—not the number of bases—bases are made on clean hits will be found the only fair and correct test of batting; and the reason is, that there can be no mistake about the question of a batsman’s making his first base, that is, whether by effective batting, or by errors in the field, such as muffing a ball, dropping a fly ball, or throwing badly to the bases, whereas a man may reach his second or third base, or even get home, through errors of judgment in the out-field in throwing the ball to the wrong man, or in not properly estimating the height of the ball, &c—errors which do not come under the same category as those by which a batsman makes his first base.
For instance, the first striker goes to the bat, and, by a sharp ground hit between short stop and third base, out of reach of both those fielders, easily secures his first. The second striker hits a ball, which is easily fielded by the short stop, and were he to throw it to first, the second striker would easily be put out, but as the point is to send it to second, to cut off the player forced from the first, striker No. 2 gets his first, not from his good hit, but from the ball having to go to second first.
Striker 3 now comes to the bat, and sends a high ball to third base, and the ball is dropped, whereupon B, the second striker, makes his second, and C, the third striker, his first. D now takes the bat, and, hitting a high ball to centre field, which ball gives a chance for a catch, runs for his second, sending C and B before him; the ball being badly judged, and, when fielded, thrown in badly, D runs for his third, and, without stopping, he risks a home run, and gets his run from another high ball.
Now, how stands the record of this play as ordinarily scored ? Why simply as follows: The man who made the best hit of the four strikers is put out at second by the poor batting of his successor, while B and C, who made their bases by poor batting, arc credited with one base each, while D gets four bases through the lack of skill of the out-fielder in judging a high ball, the result of the play being a credit for seven bases on hits and three runs, when, by a just estimate, only one man made his base by a hit, and he was the only one put out.
Now, this is the average result of the batting score in a match game. But again, in estimating bases on hits, any scorer will find that it is quite a difficult task to sift the chaff from the wheat after the first base has been made; that is, he will find that the second and third bases are made more by lack of judgment in the outer fielding, and by errors of play which are not exactly “muffs,” viz., balls handled but not stopped or picked up neatly, overthrows or miscatches; while in the in-field these errors seldom occur, the ball, generally speaking, cither being palpably muffed, thrown wildly, or not held when touched on the fly. In the scores the number of bases made on hits should be, of course, estimated, but as a general thing, and especially in recording the figures by the side of the outs and runs, the only estimate should be that of the number of times in a game on which bases arc made on clean hits, and not the number of bases made.
Chadwick prevailed, and Hits Per Game became the criterion for the Clipper batting championship and remained so until 1876, when the problem with using games as the denominator in the ratio at last became clear. If a team played several weak rivals who committed many errors, the number of at bats for each individual in the superior team’s lineup would increase. The more at bats one is granted in a game, the more hits one is likely to have. The batting average used in the 1860s is the same as that used today except in its denominator, with at bats replacing games. The suggestion for that may be credited to Hervie Alden Dobson’s letter to the Clipper of March 11, 1871.
The batting average, of course, makes no distinction between the single, the double, the triple, and the home run, treating all as the same unit—a base hit—just as its prototype, Runs Per Game, treated the run as its unvarying, indivisible unit. This objection was met in the 1860s with Chadwick’s Total Bases Average (per game), but, as one reads above, was rejected. Looking at some other data, Chadwick’s choice now seems more reasonable, less idiosyncratic.
In 1871, the first year of professional play and a mere four seasons after Chadwick’s article, only 41 percent of runs scored were earned. The fielding percentage of the National Association clubs was .833. (In 2012 the MLB fielding percentage was .983.) The number of errors per game in 1871 was 7.6 and the runs scored per game was 10.47. And these figures were for the best clubs in the country; Chadwick made his choice of batting average as the “true test of batting” while considering hundreds if not thousands of clubs in 1867, professional and amateur.
The number of runs scored per game since the dawn of the 20th century has been remarkably consistent throughout baseball history. (The high figure for 1871 is attributable to the higher error rate.) Here’s a chart I developed a few years back.
1871: 7.61 per game
1911: 3.66 per game
1961: 1.82 per game
2005: 1.22 per game
1871: 20.94per game
1911: 9.03 per game
1961: 9.05 per game
2005: 9.18 per game
In effect, the relentless increase in home runs, doubles, strikeouts, and walks have balanced the decrease in errors and triples. Baseball is a game of delicate balance, and at the outset Father Chadwick was sensitive to its nature.