Why a win explains so little
'Win' statistic no longer useful toward evaluating a pitcher's overall performance
More than a century ago, it's true the starting pitcher had more impact on whether or not his team won a game, and that may be why the "win" statistic for a pitcher was invented in the first place. When my good friend Old Hoss Radbourn started 73 games and completed all of them in 1884, perhaps his total of 59 wins meant something, since he threw every pitch, took nearly 300 turns at the plate, and might have even slipped something into the other teams' Gatorade.
The next season, John Clarkson of the Chicago White Stockings -- all 5-foot-10 and 155 pounds of him -- started 72 games, completed 68 of them, and managed a 53-16 record. In essence, he was on the field for the entirety of 96 percent of his starts, and like Radbourn, he had about a half-season of at-bats and out-hit the team's primary catcher. But even for Clarkson and Old Hoss, the pitcher win statistic was incomplete, even if it had more meaning at the time.
Now, of course, the situation has become more extreme. The pitcher win has outlived any usefulness it might have had at some point in the past.
Now, of course, the situation has become more extreme. The pitcher win has outlived any usefulness it might have had at some point in the past, and it's time to let the statistic die, even if we have to strangle it with our bare hands.
To see Keith's full breakdown on the Pie of Responsibility, and what percentage of a win a pitcher actually controls, you must be an ESPN Insider.