United States Sports Academy
America's Sports University®

The Sport Digest - ISSN: 1558-6448

The Peer Effect in the NFL draft

Submitted by: McDonald Mirabile, University of North Carolina at Chapel Hill

In this paper, I examine more than 300 NFL quarterback draft hopefuls between 1987 and 2003 and their offensive teammates to determine to what extent playing with NFL caliber teammates will have on the draft position and salary of the quarterback prospect.

I employ a selection model wherein the quarterback’s probability of being drafted (in this first stage, I use a probit model and full maximum likelihood estimation) is related to individual characteristics (height, weight, race, passing ability, etc.) and league characteristics (year, number of NFL teams, etc.). In the second stage, we use ordinary least squares estimation to determine the relationship between rookie quarterback draft position and additional individual and league-wide characteristics. In particular, we consider the relationship between the quarterback’s draft position and salary and the number of offensive teammates the quarterback played with during his collegiate career and the total number of offensive and defensive teammates selected during his draft year.

The two variables of interest are DRAFTCAST and DRAFTCLASS. DRAFTCLASS is simply the total number of players on the quarterback’s collegiate team drafted into the NFL during the quarterback’s final (usually senior) year. DRAFTCAST is defined as the number of wide receivers, running backs, and tight ends (with a maximum value of 7) drafted into the NFL within two years of the quarterback’s final year. Although this is a rough measure of the caliber of teammate the quarterback played with during his collegiate career, I believe it generally paints an accurate picture of the quality of offensive teammates at any given quarterback’s school.

I have appropriately controlled for the strength of NCAA division, conference, and school by using dummy variables for division and conference. Additionally, I use measures of the team’s season record and all-time record to account school differences within the conferences. I also control for different personal characteristics (height, weight, all-American status, forty time) to ensure that these peer effects we are seeking remain unbiased.

The point estimate of DRAFTCAST is –5.54 with a standard error of 2.97. The point estimate of DRAFTCLASS is –3.87 with a standard error of 2.65. While DRAFTCLASS is insignificant, DRAFTCAST is significant at the 10 perfect critical level. The interpretation of the DRAFTCAST estimates is straightforward: for each additional NFL drafted (NFL caliber) teammate that quarterback plays with in college, the quarterback will on average be selected about five and a half selections sooner than he would be without the presence of that additional NFL caliber teammate.

The point estimate of –5.54 may seem small until one considers it in monetary terms. Because the pay scale of NFL rookies is tied so tightly to their draft position, I will re-estimate the model using rookie quarterbacks’ salaries as the dependent variable instead of rookie quarterbacks’ draft positions. The salary point estimate of DRAFTCAST is $65,167 with a standard error of $24,550. The salary point estimate of DRAFTCLASS is $49,344 with a standard error of $23,701. Both estimates are significant at the 1% critical level.

These results suggest that NFL franchises draft based not simply on individual player characteristics but also on the caliber of an individual’s teammates. These results are beneficial to both high school quarterbacks looking to make the transition to collegiate athletics and those college quarterbacks considering transferring. While matriculation decisions will ultimately be based on numerous considerations, those athletes with NFL aspirations should carefully consider the tradeoffs between immediate playing time and playing at a university known for producing NFL talent.

Happy feet make happy people.” According to the American Podiatric Medical Association there many unhappy people walking around with painful feet. The APMA states that, “three-quarters of all Americans will experience foot-pain at some time in their life. The APMA is promoting a fitness campaign this season to get everyone walking and its top recommendation is to know your feet and get a good walking shoe. Below is a list of 10 Enemies of the top ailments of the foot: