Academic journal article Contemporary Management Research

Measuring Pitchers' Performance Using Data Envelopment Analysis with Advanced Statistics

Academic journal article Contemporary Management Research

Measuring Pitchers' Performance Using Data Envelopment Analysis with Advanced Statistics

Article excerpt

INTRODUCTION

Competitive sports interest a large number of people who watch them because of their uncertainty and unpredictability. Kao (2011) showed that the sports stars and the records in the games are the most important factors to spectators. In addition, professional sports represent a highly competitive battlefield characterized by the survival of the fittest. The games always focus on players' actual strengths and accomplishments. Thus, victory is certainly related to a player's individual performance.

Over time, the game of baseball has become increasingly complex. With the gradual improvement in baseball players' stamina and skill and increasingly varied game strategies, the division of labor in the tournament is becoming more delicately balanced. Specifically, the pitcher plays an important role in baseball. The performance of the pitcher can be crucial to the outcome of a game (Kindall, 1993; Lopez, A., & Kirkgard, J., 1996; Yeh, Lee, & Zhu, 2003).

However, how can we judge the pitcher's performance? Can we determine the pitcher's efficiency and "concretize" the results? The statistics we always hear are the earned run average (ERA) and walks plus hits per inning pitched (WHIP). Henry Chadwick defines ERA as the mean of earned runs given up by a pitcher per nine innings pitched. We determine the ERA by dividing the number of earned runs allowed by the number of innings pitched and multiplying the result by nine. Runs resulting from defensive errors (including pitchers' defensive errors) are recorded as unearned runs and are not used to determine ERA. Meanwhile, the WHIP was invented in 1979 by Daniel Okrent. In baseball statistics, the WHIP is a sabermetric measurement of the number of base runners a pitcher has allowed per inning pitched. Since the WHIP reflects a pitcher's propensity for allowing batters to reach a base, a lower WHIP indicates a better performance. We calculate the WHIP by adding the number of walks and hits allowed and dividing this sum by the number of innings pitched.

Because of the complexity of baseball, traditional statistics cannot provide a comprehensive measure of the pitcher's performance; however, advanced statistics can be credited with such a measure. Bradbury (2007) used multiple regression analysis to analyze the factor of lost points; the results showed that the pitcher's ability contributes about 73% and the defense's ability accounts for about 27% of the game's outcome. Thus, the ERA does not wholly reflect the pitcher's ability.

For example, when a runner is on base, if the relief pitcher cannot help the starting pitcher hold the runner on the base until the end of the inning, the runner will make a run or runs that affect the starting pitcher's ERA rather than the relief pitcher's ERA. This means the ERA is affected by not only the starting pitcher's, but also every other pitcher's ability. Meanwhile, the teammates' defense ability also affects the ERA and WHIP.

Therefore, the selective use of statistics to judge whether or not a pitcher performs well is bound to create bias. This explains the creation of advanced statistical metrics, such as fielding independent pitching (FIP). The 2009 American League Cy Young Award winner Zach Greinke won over the rivals using the FIP, created by Tom Tango in 2008: "That's pretty much how I pitch, to try to keep my FIP as low as possible" (Kenper, 2009). Meanwhile, 1974-2008 MLB data showed that FIP is the metric that most likely reflects the pitcher's real ability (Piette, Braunstein, McShane, & Jensen, 2010). Therefore, we seek to address the weak points of ERA and WHIP through advanced statistics, such as FIP and skill-interactive ERA (SIERA). We introduce these two statistics in the following subsections.

Fielding Independent Pitching

In 1999, Voros McCracken analyzed a record called Defense-Independent Pitching Statistic (DIPS), which is not affected by the team's defense. McCracken outlined a better way to assess a pitcher's talent level by looking at the results a pitcher can control: strikeouts, walks, hit by pitches, and home runs. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.