Stats Pay for Performance

Only available on StudyMode
  • Topic: Regression analysis, Simple linear regression, Linear regression
  • Pages : 8 (1687 words )
  • Download(s) : 38
  • Published : December 14, 2011
Open Document
Text Preview
Statistics Project

Pay for Performance in the NFL


Pay for performance is a common theme throughout almost all organizations. Merit increases, performance bonuses for executives, and commissions for real estate salespeople are common examples of this concept.

Even teachers’ pay in some states is linked to performance of their students. According to the Washington Post, the state of Florida instituted a policy that individual teacher’s raises and performance starting in 2007 will be tied directly to student’s scores on standardized tests. This pay for performance concept has generally been accepted by the new Obama administration and may make its way into more common usage across the United States.

In corporate America, examples of pay for performance are quite common, especially for top executives. Most year end bonuses are based on individuals meeting certain criteria established by the board of directors. These bonuses can be quite substantial. According to the Proxy Statement for Meredith Corporation, the total executive bonuses for the year 2007 exceeded $2.5 million dollars.

While pay for performance seems a reasonable concept in general, it is not without its critics. In education, there are a number of critics that question the fairness of the standardized test score results as a measure of teacher performance. They worry about teaching towards the exam at the expense of the overall education of the student. The criticism from Congress and much of the population of the United States over the bonuses paid to AIG executives questions how performance is actually measured.

This paper will attempt to partially address the issue of pay for performance in professional sport, specifically in the National Football League. Many different positions in football are difficult to obtain good performance measures. Offensive lineman, special teams players especially do not have good measures of individual performance that are tracked. This analysis will focuses on two groups of NFL players, quarterbacks and running backs where individual performance measures are readily available.

Analytical Technique

A correlation study will be done on a variety of performance measures and the salaries of both NFL quarterbacks and running backs to see which of the individual performance measures are most closely related to the individuals salaries. The assumption will be that the current salary is based on last year’s performance. In addition to the correlation study, a multiple regression model with the best performance measures will be used to explain the relationship between the measures and salaries. This could be potentially used as a basis of predicting next year’s salary for those players that are in contract discussions or are entering the market as free agents.

The data for the study will be obtained from two primary sources, which tracks player performance measures for a number of years, and for player salaries.
Professional football players are compensated in a number of ways, base salary, signing bonus, and other bonuses. This study will be using base salary as the pay in the pay for performance analysis.

Performance measures for quarterbacks will include: completion percentage, total passing yardage, touchdown completions, interceptions, and finally QB rating.
Performance measures for running backs will include: total yards, yards per game, touchdowns, and fumbles lost. While other measures are collected it is felt that these are the most appropriate performance measures to use for both categories of NFL players.

A sample of 22 NFL quarterbacks from the 2007 season was selected while a sample of 13 NFL running backs from 2007 was used.


NFL quarterbacks:

Pearson’s correlation coefficients for all variables in the study were run and are presented in the table below:

|  |PCT |YDS |TD |INT...
tracking img