Teaching Statistics Using Baseball / Edition 1 available in Paperback
Teaching Statistics Using Baseball is a collection of case studies and exercises applying statistical and probabilistic thinking to the game of baseball. Baseball is the most statistical of all sports, since players are identified and evaluated by their corresponding hitting and pitching statistics. There is an active effort by people in the baseball community to learn more about baseball performance and strategy by the use of statistics. This book illustrates basic methods of data analysis and probability models by means of baseball statistics collected on players and teams. Students often have difficulty learning statistics ideas since they are explained using examples that are foreign to the students. The idea of the book is to describe statistical thinking in a context (that is, baseball) that will be familiar and interesting to students.
The book is organized using a same structure as most introductory statistics texts. There are chapters on the analysis on a single batch of data, followed with chapters on comparing batches of data and relationships. There are chapters on probability models and on statistical inference. The book can be used as the framework for a one-semester introductory statistics class focused on baseball or sports. This type of class has been taught at Bowling Green State University. It may be very suitable for a statistics class for students with sports-related majors, such as sports management or sports medicine. Alternately, the book can be used as a resource for instructors who wish to infuse their present course in probability or statistics with applications from baseball.
Read an Excerpt
Over twenty years, this author has been in the enterprise of teaching introductory statistics to an audience that is taking the class to satisfy their mathematics requirement. This is a challenging endeavor because the students have little prior knowledge about the discipline of statistics and many of them are anxious about mathematics and computation. Statistical concepts and examples are usually presented in a particular context. However, one obstacle in teaching this introductory class is that we often describe the statistical concepts in a text, such as medicine, law, or agriculture that is completely foreign to the undergraduate student. The student has a much better chance of understanding concepts in probability and statistics if they are described in a familiar context.
Many students are familiar with sports either as a participant or a spectator. They know of the popular athletes, such as Tiger Woods or Barry Bonds, and they are generally knowledgeable with the rules of the major sports, such as baseball, football, and basketball. For many students sports is a familiar context in which an instructor can describe statistical thinking.
The goal of this book is to provide a collection of examples and
exercises applying probability and statistics to the sport of baseball. Why
baseball instead of other sports?
· Baseball it the great "American game." Baseball is great in that it has a rich history of teams and players, and many people are familiar with the basic rules of the game. The popularity of baseball is reflected by the large numbers of movies that have been produced about baseball teams and players.
· Baseball is the most statistical of all sports. Hitters and pitchers
are identified by their corresponding hitting and pitching statistics. For
example, Babe Ruth is forever identified by the statistics 60, which was the
number of home runs hit in his 1927 season. Bob Gibson is famous for his record
low earned run average during the 1968 season. A flood of different statistical
measures are used to rate players and salaries of players are determined in part
by these statistics. There is an active effort among baseball writers to learn
more about baseball issues by using statistics.
· A wealth of baseball data is currently available over the Internet. Player and team hitting and pitching statistics can be easily found. Comparisons between players of different eras can be made using a downloadable dataset that gives hitting and pitching data for all players who have ever played professional baseball.
This book is organized using the same basic organization structure presented in most introductory statistics texts. After an introductory chapter, there is a chapter on the analysis on a single batch of data, followed by chapters on the comparison of batches, and the analysis of relationships. There are chapters on introductory and more advanced topics in probability followed by topics in statistical inference. Each chapter contains a number of "essays" or "case studies" that describe the analysis of statistical or probabilistic methods to particular baseball data sets. After the collection of case studies in each chapter, there is a set of activities and exercises that suggest further exploration of baseball datasets similar to the analysis presented in the case studies.
How can this book be used in teaching probability or statistics? We
suggest several uses of this material.
· This book can be used as the framework for a one-semester introductory statistics class that is focused on baseball. Such a class has been taught at the author's home institution. This course covers he basic topics of a beginning statistics course (data analysis, introductory probability, and concepts of inference) using baseball as the primary source of applications. This course is suitable for students who are interested or curious about the game of baseball. It is also suitable for students with sports-related majors, such as sports management or sports medicine.
· This book can also be used as a resource for instructors who wish to infu8se their present course in probability or statistics with applications from baseball. The material in the book has been presented at different levels to make it useable for introductory and more advanced courses. The case studies can be used by the instructor to present the particular topic within a baseball context and then the associated exercises and activities can be used for homework, The case studies can serve as useful springboards for undergraduate students who wish to do additional explorations on baseball data.
Table of Contents
1. An Introduction to Baseball Statistics
2. Exploring a Single Batch of Baseball Data
Case Study 2-1: Looking at Teams' Offensive Statistics
Case Study 2-2: A Tribute to Cal Ripken
Case Study 2-3: A Tribute to Roger Clemens
Case Study 2-4: Analyzing Baseball Attendance
Case Study 2-5: Manager Statistics: the Use of Sacrifice Bunts
3. Comparing Batches and Standardization
Case Study 3-1: Barry Bonds and Junior Griffey
Case Study 3-2: Robin Roberts and Whitey Ford
Case Study 3-3: Home Runs- A Comparison of 1927,1961, 1998, and 2001
Case Study 3-4: Slugging Percentages Are Normal
Case Study 3-5: Great Batting Averages
4. Relationships Between Measurement Variables
Case Study 4-1: Relationships in team Offensive Statistics
Case Study 4-2: Runs and Offensive Statistics
Case Study 4-3: Most Valuable Hitting Statistics
Case Study 4-4: Creating a New Measure of Offensive Performance Using Multiple Regression
Case Study 4-5: How Important is a Run?
Case Study 4-6: Baseball Players Regress to the Mean
Case Study 4-7: The 2000 Dinger Drop-Off
5. Introduction to Probability Using Tabletop Games
Case Study 5-1: What is Barry Bonds's Home Run Probability?
Case Study 5-2: Big League Baseball
Case Study 5-3: All Star Baseball
Case Study 5-4: Strat-O-Matic Baseball
6. Probability Distributions and Baseball
Case Study 6-1: The Binomial Distribution and Hits Per Game
Case Study 6-2: Modeling Runs Scored: Getting on Base
Case Study 6-3: Modeling Runs Scored: Advancing he Runners to Home
7. Introduction to Statistical Inference
Case Study 7-1: Ability and Performance
Case Study 7-2: Simulating a Batter's Performance if His Ability is Known
Case Study 7-3: Learning about a Batter's Ability
Case Study 7-4: Interval Estimates for Ability
Case Study 7-5: Comparing Wade Boggs and Tony Gwynn
8. Topics in Statistical Inference
Case Study 8-1: Situational Hitting Statistics for Todd Helton
Case Study 8-2: Observed Situational Effects for Many Players
Case Study 8-3: Modeling Batting Averages for Many Players
Case Study 8-4: Models for Situational Effects
Case Study 8-5: Is John Olerud Streaky?
Case Study 8-6: A Streaky Die
9. Modeling Baseball Using a Markov Chain
Case Study 9-1: Introduction to a Markov Chain
Case Study 9-2: A Half-Inning of Baseball as a Markov Chain
Case Study 9-3: Useful Markov Chain Calculations
Case Study 9-4: The Value of Different On-base Events
Case Study 9-5: Answering Questions About Baseball Strategy
A. An Introduction to Baseball
B. Datasets Used in the Book and Acquiring Baseball Data over the Internet