Suddenly last summer Senior Editor Larry Keith decided it was time to create some new statistics for relief pitching. "There's a plethora of batting data," Keith said to himself. "Teams now keep stats on things like how often a guy hits behind the runner or scores a man from third with fewer than two outs. But relief pitchers are judged solely by saves, wins and ERAs; their importance has far surpassed the indices for grading them. Everybody talks about this problem, but nobody does anything about it."
This is an article from the April 12, 1982 issue
Keith assigned Staff Writer Jim Kaplan to do something: namely, work out a better system for evaluating the men in the bullpen and write an article about it. The result is The New Way to Spell Relief, beginning on page 78.
"Baseball is as fascinating as chess," says Kaplan, "and creating a new formula for gauging relief pitching is like developing a new variation or defense in chess." He found similar enthusiasm among the players, coaches and managers he sought out during the last two months of the 1981 season. "Everybody had a system," he says, "and it wasn't usually a selfish one created for contract negotiations. There was general agreement that players who pitch well without getting a win or save should be awarded some sort of statistical credit, and that relievers who let a previous pitcher's runners score should be held accountable. Our formula reflects both views."
Kaplan devised his system with assistance from America's foremost sports statistics service, the New York-based Elias Sports Bureau, which after three months of research produced the charts used in this issue. There was more to the job than feeding data into a machine. "The computer verified our numbers on relief innings pitched, ERAs and wins and saves," says project director Steve Hirdt, "but then we had to create a formula for runners per nine innings and look up specific situations—what a guy did against the first batter he faced, what the situation was when he came in. That was the most time-consuming part of the study."
It was also the most interesting. "I kept discovering guys who gave up hits with men in scoring position and were never charged with the runs," says statistician Bob Rosen. "That showed me how valueless the usual box score line for relievers is."
Our system, of course, hasn't been adopted by the major leagues, but we have our hopes. In 1959 Jerome Holtzman, then of the Chicago Sun-Times, created the save statistic and began publishing it in The Sporting News. Baseball officially recognized the save in 1969. "The sport has taken special notice of relief pitching for only 32 years, since Jim Konstanty helped the 1950 Phillies win the National League pennant," says Seymour Siwoff, president of Elias. "Now relief pitching is one of the most important ingredients in building a winning team. We feel that SPORTS ILLUSTRATED has created something meaningful for understanding the phenomenon. This is a beginning."