To real baseball fans, statistics are indispensable, and inextricably tied to understanding and enjoying the game. But how useful are ordinary baseball stats as tools for evaluating a player, choosing a strategy, or predicting a winner?
In this lively and thought-provoking look at the numbers and the game, Jim Albert and Jay Bennett examine just what we learn, and just what we think we learn, from baseball statistics. The authors consider the key questions every serious fan obsesses about: What is the best way to rate a great hitter? Is there really a fair way to name an MVP? Does anyone have a reasonably accurate way to predict the outcome of a game? How likely is it that some of the game's milestone achievements (e.g., Mark McGwire's single-season home run record) will be broken?
By incorporating the seldom-used statistical techniques of probability, the authors come to some original and surprising conclusions: It turns out, for example, that the phenomenon of "streakiness" (a hot hand, a hot bat) is measurable and can serve as a very useful predictor of performance. Conversely, they find that a lot of situational statistics (home versus away games, play on artificial turf versus grass) are, statistically speaking, little more than "noise." And, in news that will bring consolation to Cubs and Red Sox fans, they declare that it's not always the best team that wins the World Series.
Keeping the mathematics at an accessible level, Albert and Bennett show that statistics is not just a powerful tool of analysis and prediction, but a pleasurable and informative pastime in its own right.
About the Author
Jim Albert is Professor of Mathematics and Statistics at Bowling Green State University. He currently chairs the Sports Section of the American Statistical Association.
Jay Bennett is a Senior Scientist with Telcordia Technologies and editor of Statistics in Sport (1999). His views on baseball statistics have appeared in USA Today, Time, and Omni.