Question: Does the success of a college football player’s team impact their opportunities in the NFL?

Businesses may attribute traits to future workers based on where they went to school, for example a law firm is likely to find a graduate of Harvard Law as an attractive candidate, because of the stereotypical associations with going to a prestigious school like Harvard. These institutions have high barriers of entry and rigorous coursework and so for the applicant to have gotten into Harvard, much less successfully completed education there, it generally suggests that they are competitively intelligent and hardworking. Therefore a business like a law firm might find that it is good practice to focus their recruiting efforts on schools like Harvard if they want to cut through unintelligent and poor-work-ethic laborers.

Data shows that NFL teams typically draft college players from successful college programs higher in the draft then players from smaller schools. Because financial compensation in the NFL is heavily dependent on draft order for rookies, and the difference of a few picks can be the difference of a few million dollars, this shows that NFL teams are projecting future production of employees based where they attended college and by valuing these workers more they are subsequently compensating them higher, regardless of if they have similar or identical draft measurables to a player from a smaller school.

However, because the NFL subjects incoming talent to screening methods of their abilities, such as college pro days and the NFL combine where teams can evaluate a player's athletic measurables and even conduct interviews to determine disposition, they have a larger pool of information to evaluate workers beyond their “institutional affiliation.” Most businesses outside of professional sports do not pre-screen or test prospective employees for abilities that may correlate to success on the job, much less teams of career talent evaluators who spend thousands of hours scouting potential future workers, so evidence of college-based discrimination in the NFL is arguably of a slightly different nature in sports than in it in other fields.

NFL teams acquire talent for their business in a unique way compared to other businesses because professional football is a monopsony, meaning there is only one league that will hire you to be a professional football player (CFL isn't real). To ensure that bargaining wars for talent do not happen among teams, incoming workers are selected from a draft pool one by one with teams picking in a set order. This ensures that teams will not have to compete with each other in luring talent to their franchise, therefore lowering costs on labor. Because much of sports entertainment is derived from a competitive balance within the league, the draft order is set so that the worst teams get a chance to select talent first and the best teams get a chance to select last. This theoretically allows bad teams a chance to acquire better talent, which in theory makes the league more competitive, which in theory increases the value of the NFL product (entertainment).

NFL collective bargaining agreement dictates that player salaries are determined by scale based on draft order, and if NFL teams are guilty of statistical discrimination based on collegiate team success then players are partially powerless in determining their draft order (and therefore salary) beyond how they convey themselves athletically and intellectually. A team will select a player based upon their organizational projection of how that player will produce, and evidence shows these projections can sometimes be multifaceted and consider external factors beyond the individual.

For example, in the 1997 NFL Draft the Tampa Bay Buccaneers management revealed that they had equal draft grades of running backs Warrick Dunn of Florida State and Tiki Barber of University of Virginia, but ultimately elected to go with Dunn because “with everything else being equal we went with the Florida State guy.”

If a draft eligible player from a successful college football program is more likely to be drafted higher than a player of equal measurables but from a less successful program, then perhaps teams assign intangible attributes to players from more successful programs. Not quite in the way that a Harvard law graduate might demand a unanimous perception of being intelligent, because college sports teams are generally more eclectic in the talent they attract and produce, but rather NFL teams may think that if a player was part of a winning college team than that player may possess a dispositional attitude that is conducive to winning.

The interesting part of this study comes from the fact that despite these projections of future productivity, (which again are concluded by large teams of professional full-time talent evaluators, a luxury that most businesses do not have) players drafted from smaller schools are on average no less-likely to succeed than their big time counterparts. In fact, and although this may be attributed to larger sample size, players from bigger schools have a higher “bust rate”, or chance of not living up to pre-draft projections of productivity, than their small-time counterparts.

Further, while this statistical discrimination may have an effect on how a player is initially valued and compensated on their rookie contract, it has virtually no bearing on a player's long-term success rate. Players of lower draft status may contribute to team success just as equally, and even sometimes surpassingly so, than those of higher draft status. This form of discrimination expires as the rookie contract does, as successful players will negotiate their second NFL contract based on their production in the NFL regardless of their initial evaluation.

The methodology in determining the data in the study I keep alluding was to take the players college teams ranking, (Much like there are large deliberations of men who spend thousands of hours evaluating collegiate football talent at the individual level, there are also groups of men who use algorithms to analyze whole team success and produce a dynamic week-by-week ranking systems of the top 25 college football programs) regressed upon where in the draft the player was selected with respect to the individual's projected draft value determined by scouts. This projected value is determined by factoring in position, physical attributes (height, weight, speed, etc.), and perceived mental attributes like intelligence and toughness. "Grittiness."

This data was then judged against whether the player had a successful NFL career. Because the study was not conducted with respect to position, individual NFL stats were not used, nor were team wins because wins are a team statistic. The determinants of the players NFL success were based on length of career (with milestones at 3 years and 6 years), number of games played, and number of games started.

To be eligible to be included in the data set, a player had to have appeared in the NFL Combine, played Division 1 NCAA football, and have been ranked by third party scouting agencies.

This data found that when all of those factors were accounted for, which qualified 1,225 out of the 3,570 players drafted from 1999 to 2012, for every 1 increment higher that a player's college team was rated, the individual player was drafted on average .385 sports higher. At first glance, this confirms evidence of statistical discrimination.

However, despite the marginal sensitivity of a player's college team rating in respect to draft order, when examining players of equal measurables (of everything except for college ranking) it was found that for every 1 round higher a player was drafted, they only had a slightly over 2% increased likelihood of still being in the NFL 6 years after the draft. This proves that long term, playing at a highly ranked college team does not have much long term bearing on a player's opportunity for success beyond the finances of the initial rookie contract. Rookie contracts typically do not involve much guaranteed money outside of the top of the first round anyways, and so if a player is bad a team can cut them without much financial penalty and if a player is good the team can re-sign them to an appropriate deal. The market corrects itself.

In the popular media, the fact that big name schools produce many draft picks is common knowledge to the average football fan and even serves as pseudo-conventional wisdom for armchair talent scouts who follow the NFL draft. It is addressed almost as a given in most sports media circles, with school-centric stereotypes like “Oh, he’s a Michigan guy so you know that he’s (tough/well-disciplined/insert favorite cliche here).” An NFL.com article titled “Seven schools that produce the best draft picks since 2000” reinforces the stereotype that there are figuratively “superior ponds to fish in” when it comes to NFL talent, and not-so-ironically all 7 of the schools are perennial big names in the NCAA (Alabama, Miami, Michigan, Wisconsin, USC, Oklahoma, Tennessee)

Some of the best players in the league have come from colleges that some people might not even know exist, like Antonio Brown from Central Michigan University or Khalil Mack coming out of Buffalo. And on the flip side, every year there is a "blue chip" prospect from an SEC school that gets drafted in the first round and is out of the league in 5 years because they are addicted to lean.

The draft is a science but its also very much Russian roulette. It's kind of like, "Yeah I know this guy who's really good with dogs. I've seen him play with tons of dogs. I bet you any dog would love him" and then you introduce that guy to a crazy difficult dog and if the dog loves him you'll be like "I knew it! I'm a genius" but if the dog hates him you go "Woah, I wiffed there. Back to the drawing board."

This statistical discrimination based on institutional affiliation can be applied to the real world outside of sports in that if a firm is only recruiting talent from certain universities or organizations, they could be missing out by limiting their options of talent and over-looking eligible and competent workers (who they may even be able to sign at a discounted rate, ala undrafted free agents like Antonio Brown) who could prove to be equally or more productive than their big-school counterparts.



The study that I keep alluding to in the article for you nerds out there:

Carl T. Kitchens “ARE WINNERS PROMOTED TOO OFTEN? EVIDENCE FROM THE NFL DRAFT 1999–2012” Economic Inquiry Volume 53, Issue 2, April 2015 Pages 1317–1330 Web 8/17/2017 (http://onlinelibrary.wiley.com.proxy-um.researchport.umd.edu/doi/10.1111/ecin.12165/full)