February 28, 200521 yr http://www.hardballtimes.com/main/article/closer/ By use of real play by play data, you can discover a team's probability to win a game in any situation. Up by 1 at home in the 7th inning with 1 out and a man on first? That team will win 80% of the time. Get a single and walk to get 2 men on, and it jumps to 83%, but ground into a double play it goes down to 75%. Win probabilities are truly some interesting stuff. It was what converted me to the Bill James club of avoiding outs at all costs and adoption of on-base percentage (that is, non-out plays / plate appearances) as a key measurement of a ball player's ability. Anyhow... in this study, Dave of Hardballtimes uses a method developed by Doug Drinen and drawn from win probabilities to determine the level of high-leverage situations of every reliever. From this he finds that it's mostly middle relievers and situational platoon pitchers are brought into the game when their team is at the greatest risk of surrendering runs and letting the game get away. He then goes on to check if this is simply the result of managers' trend to hold off using their closers, lest fear them not be around for the ninth or extra innings. That's not effecting the results, he finds, what is a significant amount of managers are unwilling to use their closer, their best reliever, their relief ace, without holding the lead. Rather than shut down the opposition when the opposition's chance to win is at its highest, managers are asking lesser pitchers to do the job. Even in the ninth inning! (Some of you may be aware of this, but the way it was presented was interesting and easy to read, so I'd thought I'd share. When HardballTimes' 2005 book comes out, I'll post the results of Marlins pitchers in the appropriate forum.)
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.