The Repository @ St. Cloud State

Open Access Knowledge and Scholarship

Date of Award


Culminating Project Type


Degree Name

Applied Economics: M.S.




School of Public Affairs

First Advisor

David Switzer

Third Advisor

David Robinson

Keywords and Subject Headings

Baseball, Salary, Runs, Performance, Modeling, Major League


There are numerous statistics, such as Batting Average, On-Base Percentage, and Slugging Percentage, which attempt to measure the value of Major League Baseball players. However, the statistics commonly used today each have problems associated with it, which means they are not a good predictor of runs scored.

In an attempt to improve on these statistics, numerous models have been developed by Lindsey, Pankin, James, Bennet and Flueck, Thom, Palmer, and Gershman, and Furtado that estimate runs by assigning values to various statistics. There have also been numerous models, such as those by Hakes and Sauer, Brown and Jepsen, and Stone and Pantuosco, that attempt to model a player's salary based on his statistics.

Attempting to improve upon models developed previously, this study uses data from 1955 through 2010 to build a model, which we call Linear Runs Estimated (LRE), that predicts team runs based on numerous statistics. In order to look at league effects, we also develop separate models for both the American and National League. We find that these models are a better predictor of runs than the statistics currently used today. We then apply these models to individual players so we can measure their performance.

Next, we use salary data from the 2012 season to test various statistics, including our model, to see which one is the best predictor of a player's salary. Using data from the previous four seasons, we find that, along with models by James and Furtado, our model is the best predictor. Based on this salary model, and a player's performance over this time period, we measure which players are over and under-paid.

Included in

Economics Commons