By Travis Hoke
Webmaster: This article originally appeared in Esquire magazine’s October 1935 issue.
Every now and then, like most good Americans, I take a look at the baseball "averages." I suppose I haven't seen three games in five years, but I keep more or less in touch, and I glance with fair regularity at the tabulations of batting, pitching, and fielding percentages. I get something out of those tabulations—a laugh or a sigh, perhaps both—that I daresay no one else gets, unless it is Branch Rickey, the general manager of the St. Louis Cardinals.
The averages are supposed to measure a player's achievements, to gauge statistically his performances in batting, fielding, or pitching, and compare them with perfection–the “possible”—and with the performances of other players.
They do no such thing, of course—that, I suppose, is where the sigh comes in. They do not really indicate his ability at bat, or in the field, or on the mound. The players know it, unconsciously, at least. So do the makers of the statistics—there are three or four firms that compile them—for every few years they add such refinements as “Runs Batted In,” “Earned Runs,” etc., in the obvious effort to make the figure mean something. They have been conscious of the inadequacy of baseball statistics, they have been shooting all around the target and never hitting it, and what they strive for has been in plain sight all the time—that, probably is the cue for the laugh.
For there is a system of statistics that measures a player’s performance as fairly and accurately as figures can do it. The system has been in existence at least 30 years. It was offered to baseball and the public 25 years ago. It was used by a club in the American League for four or five years that I know of—secretly, however: it was worth money to that club to have information about players that the official averages did not reveal, and the very existence of the system was known to only three persons connected with baseball.
Well, I saw Branch Rickey the other day, and he gave the old hi-and-run sign so Now It Can Be Told. I sat with him and Frank Frisch, the Cardinals’ manager, and his Brain Trusters, Mike Gonzales and Clyde Wares, in a New York hotel room while they discussed the purchase of a pitcher from another club. They discussed his history , physique, temperament, habits, and averages, and about all they could dig up from the dope-books was that he had won so many games and lost so many more-none too fancy an average. Yet they all though him a good pitcher; they discounted the averages. Wares pointed out that his record was made with a losing club, Frisch thought he should win so many games more a season with a club like the Cardinals. Mike thought he would if he handled well. They had all the dope there was to be had on the player, yet they were forced to guess. Ricky took it all in, and then, in that intense way of his turned to me. “Judas Priest,” he exclaimed—the nearest to swearing that he allows himself—“if we had the old system working we’d at least know what this bird has done—we wouldn’t have to guess. Judas Priest, will they never give it to the fans?”
Let’s see what’s wrong with the official system, in case you haven’t discovered it for yourself. Let’s start with the batting averages. Basically, they are the record of how frequently a player succeeds in making a safe hit. If his average is .333, to make it simple, that means that on the whole he gets one hit in three times at bat. But what kind of hit—single, double, triple, home run? There is a difference in hits, a big difference, and the statisticians not too many years ago recognized the fact and added a column, “Total Bases,” to their batting tables.
That “Total Bases,” however, doesn’t have anything to do with the main average. A man might never hit anything longer than a single, and still have a batting average of .400 or anything up to 1.000.
And when does the player hit—with no one on base, or when it counts. A single with men on second and third is worth more to the team than a home run with the bases empty, but the averages do not make any distinction—each is a hit, and just that. The statisticians have felt the inadequacy here, too, and have tried to make up for it with a column headed “Runs Batted In,” but it means little because the best batter in the world cannot drive in runs if the batters ahead of him don’t get on base.
There are other things the matter with the batting averages, but that’s enough to give you the main idea of how little the official figures really tell about a player’s batting, in relation to perfection or the performance of other players.
It’s the same story with fielding averages. An error may permit a lone runner to get to first, or it may let in four runs—yet it’s just an error in the statistics.
But the pitching averages are the most hopeless of all. They represent, fundamentally, the number of games that a pitcher wins, out of the number he takes part in, and they are all cluttered up by rules as to who gets credit for a win and who gets charged with a loss. Several pitchers can work in one game for a club; only one gets credit or discredit for the final result. And what logic—to say that a pitcher wins or loses a game—when there are eight other men helping him win or lose it! The statisticians have felt that illogicality, and have tried to atone for it by keeping track of the “Earned Runs” scored against a pitcher, but to make that tabulation significant you still have to check it against the win and loss record of his club—and then guess.
So the official averages are no proper yardstick of a player’s worth, as I imagine you have often realized. There should be some true criterion, if there are to be any statistics.
It must have been thirty-five years ago that I began to worry about all this, and the blame rest with my nucle, William E. Hoke. W. E. H. is an inventor, and the son of an inventor, and he has invented everything from gauges to fables, but no gadget he ever thought up was so successful as the device by which he, a connoisseur of late sleeping, made me, a busting-around kid, keep quiet of mornings. His idea had the simplicity that inevitably attaches to greatness: he got me interested in professional baseball, and I kept myself silent and busy with the sport pages of the morning paper until he thought it expedient to arise. (To keep up my interest he had to take me to the game frequently, so perhaps I wasn’t so stupid, at that!)
Anyhow, I got to know enough about the game to realize that the averages didn’t do what they were supposed to do, and I told him so. Like any Hoke, my uncle was distinctly unimpressed with the opinions of any other Hoke, and besides, I was fresh, lippy kid—they used such horrendous words in those days, before the advent of experimental education and the tender treatment of juvenile bids for attention—but somehow the idea stuck with him and he mulled it over for a week.
He didn’t solve the problem for me: he made me tackle it for myself. If the official averages were wrong, he said, wherein lay the fault? Until I could put my finger on it I couldn’t fairly claim to know what I was talking about. And the deeper the error lay, the more surely was it the chief flaw. He led me along, step by step, making me think it out until I saw the answer.
The very name of the game itself provides the answer: base ball. A hit isn’t a hit—it is a one, two, three, four base hit. The unit of achievement in baseball isn’t a hit, it is a base. That is the simple, obvious fact, and the reason that the official statistics are no true criterion is simply and obviously that they do not take the base for their unit.
We worked out a system with the base as a unit. We gave the batter credit for the number of bases he achieved for his team, compared to the number of bases it had been possible for him to achieve. If he came to bat with no one on base his “possible” was four bases. If he singled, then, he got credit for 1 base—.250; if he hit a home run he earned 4—1.000. If there were men on base he got credit for the bases he advanced them—a homer with the bases full would give him 10 bases (count it for yourself) out of a possible ten. We were actually measuring a player’s performance in realistic terms, as accurately, we thought, as figures could do it.
Fielding averages we compiled with the same unit. A player was charged with the number of bases his error amounted to—it penalized him for misplays when they counted. And how beautifully it worked for pitching averages! No games won and lost, but bases gained against a pitcher—deducting those for which fielding errors were responsible, of course. It was a double-entry system, too, for the number of bases gained by a team naturally equalled those charged against the opposing pitcher plus bases made on errors.
It was the right idea, and a simple system, and we kept score with it at many a game. But there was nothing more we could do with it: you can’t patent an idea, and since no special form of score-card was necessary, there was nothing to copyright.
Some years later, when I was a cub reporter, I thought I saw the chance to bestow this dazzling conception on a stunned world, and with that end in view I became one-third of the staff of a weekly called Sport and Stage—the editor of which, Hal Lanigan, I had convinced of the soundness of the idea. He printed a full and detailed description of the system, and sent copies to every sports editor in the major league cities, and most of them reprinted it and praised it and said something ought to be done with it. After that single blast, there was an avalanche of silence.
So that was that until the winter of 1912-1913, when, at the home of a newspaper editor, I met Branch Rickey. He was a young lawyer, a graduate of Michigan, and had just come to St. Louis to be secretary of tha American League club, the Browns. He had had very little major league experience, but in my biased opinion he was then and still is the smartest man in baseball. He was perhaps too far ahead of the rest of baseball—that’s sometimes been his misfortune—but I could tell you of dozens of baseball tactics, now standard, that he originated in the days when there really was “inside” ball.
And how he leaped for the system! He saw at once that if there was any reason for baseball “dope” it must tell what it purports to tell: he saw that if he could get a guide to playing ability that other officials lacked he was that much ahead of them, and he engaged me to keep records the coming season, on all the players in the American League and on such minor leaguers as the Browns were watching. We took newspapers from every city in the league and compiled the averages from the play-by-play accounts of each game.
But the system had been changed. I had realized by then that it was not accurate to credit a hitter with one unit for each base, because all bases are not equally important. It is more than twice as valuable to the team, for instance, to hit a double than a single, because a man on second is in position to score on a following single. It is more valuable to hit safely with a man on second that with one on first, and there should be recognition of the difference, in the figures. So I had revised the system to fit. If a man singled with the bases empty he was credited with one—from home to first counted one base—but from first to second counted two bases, so that if he doubled he received credit for three, not two bases. A triple got him six bases, and so on. A home run with the bases empty meant ten out of a possible ten bases.
If there was a man on first and the batter singled, sending the runner to third, the batter got one base for his hit and five for the bases he advanced the runner—six in all, out of a possible 19. If the batter scored a runner from third with a single, he got five out a possible 15 , and so on. The converse, of course, was the case for pitching averages. A pitcher who yielded a hit with men in position to score automatically had more bases charged against him than if the hit came with the bases empty.
That, briefly, was the final, the perfected, system, and it was not as complicated as it may sound—in fact, it was surprisingly easy to operate. It showed everything about the performance of a player that could be shown by figures, and justly and accurately. It couldn’t of course, make a distinction between one run and another, between the winning run and a superfluous run, any more than it could distinguish between a game that would win a pennant and any other game, but up to that point, it worked.
I used it with the Browns for two years. Those were the days when Miller Huggins was with the other St. Louis team, the Cardinals, and when we had on the Browns Bobby Wallace, just about the finest fielding shortstop in the game—and Jimmy Austin and Dell Pratt and Sam Agnew and Tillie Walker and Clarence Mitchell and Burt Shotton and Bert Hamilton and that gang; the Federal League days when we never knew what moment we wouldn’t have a team, the days when Ty Cobb was tops and Babe Ruth was just coming up. I used the system, and it showed surprising things—rather, it proved some things that wise baseball men believed in, despite the official figures, and it disproved other beliefs. Some people in the American League thought, for example, that Duffy Lewis of the Red Sox was underrated as a hitter, that he was about the best man in the League in a pinch. My figures proved it: In one year Lewis, by my system, batted a little better than Cobb, as I remember it, and the other year he was just behind Ty. In the official averages no such thing occurred, of course. And certain other players, stars to the public and to many baseball men by reason of the official averages, became just ordinary players in my book.
As for pitching ability, well, as Branch said the other day, we didn’t have to guess. Whether a pitcher was on a winning or a losing club, the bases scored against him, not games won and lost, showed what he had accomplished.
Of course, that wasn’t enough for Branch Rickey. He realized that a lot of guess work and hearsay becomes, in time, accepted as fact. Let a batter bite on a low one outside two or three times in a row, and soon it is accepted all over the circuit that he bites on low ones outside. A mind like Branch Rickey’s didn’t want to rely on such data. So I sat in the grandstand behind home plate for at least one series each season with every other club in the League and recorded what batters actually struck at and what they accomplished—by using a system of dots and dashes I could tell Branch what had been pitched to any batter at any time, and whether he had hit, and if so, to where—and it was surprising to see how many myths those records disproved.
Both that and the average system were secret—only Rickey and Colonel Hedges, the owner, and I knew about them, though I suspect that Charlie Barret, the scout, had a pretty good idea all the time; if he didn’t it was about the only thing in baseball he didn’t know. The players never knew what I was up to: when I saw Clyde Wares the other day he learned for the first time what I was doing with all those papers in the old office in Sportsman’s Park.
The baseball writers, naturally, were curious, and one of them tried to read my score-book through field glasses from the press box above me, and failing, wrote a rather acidulous bit about “Fort Hoke, the impregnable, etc.”
All the fun, but it was doing what amounted to bookkeeping, and like any youth of my disposition. I couldn’t take it. I fell behind with the figures, and when, for the fourth or fifth time I couldn’t honestly give Branch the up-to-date dope on some player or other, he gently let me resign at the end of my second season. I presented him with the system, and he used it thereafter, I don’t know for how long.
But with the chain-store scheme he developed for the Cardinals, it would take a big staff of accountants to keep track of all Cardinal prospects and the rest of the National League. He doesn’t use it any more.
So baseball and the public can have it. If averages are worth compiling and publishing they should be kept with the base as a unit. But they won’t be, I fear. There will be the same old Runs Driven In and Games Won and Lost and Earned Runs, and they will continue to mean what they have always meant, which is nothing, net. As to my base-unit system, the history of 25 years ago will probably repeat itself, and the silence will be deafening.