SEMINARS
|
|
Fall 2006
STATISTICS
COLLOQUIUM
Wednesday, October 4, 2006
3:30-4:00—Refreshments
4:00-5:00—Talk
Yost Hall, Room 101
Jim Albert, PhD
Department of Mathematics & Statistics
Bowling Green State University
Hitting in the Pinch
In baseball, there is a general fascination regarding the tendency of players to perform well in clutch situations. We look at the clutch-hitting phenomena by focusing at four fundamental rates, the rate of obtaining a walk, the rate of striking out, the rate of hitting a home run, and the rate of getting an ``in-play" ball to fall in for a hit.
We fit a random effects model to the rates of players for a particular season and show that rates differ with respect to the proportion of variability that can be attributed to luck. The success of a hitter can depend on the runners and number of outs and we show how the four hitting rates can depend on these variables. A random effects model with a bias component is used to represent hitting data in "clutch" and "non-clutch" situations. Using this model, we look for players who perform unusually well in clutch situations.
|
