Data Involving Historical Preseason Polls Sees USC as No. 2 in “Overratedness”


Kirby Lee/Image of Sport-USA TODAY Sports

In last year’s preseason AP Poll, the USC Trojans were ranked No. 1 overall. By November, they found themselves unranked and at one point had fewer total votes than five other California schools, including all three Cal State schools playing at the FBS level.

Naturally, they were labeled as ‘overrated’, with some calling the 2012 Trojans the worst preseason No. 1 ever.

All things considered, it’s hard to argue against that. They had sky-high expectations and fell flat on their face during the second half of the year and never lived up to their preseason marks.

But according to, the ‘overrated’ trend isn’t exactly new to USC. Dating back to 1989, they’ve determined that only Oklahoma has a worse cumulative “overratedness”.

By taking the yearly plus-minus differential of the consensus of preseason polls and the consensus of final polls, STASSEN is able to create a number that determines how a team has lived up to preseason rankings, thus allowing for “overratedness” and “underratedness” to exist.

For example, a team that was ranked No. 2 in the preseason and No. 5 in the final poll would earn a score of negative-3 for that year. Add these numbers up each year since 1989, and a trend can be formed with a cumulative count.

As of now, USC scores a -97, just ahead of Oklahoma at -97.5. As for the opposite end of the spectrum, teams like Oregon(+80), Boise State(+79) and Kansas State(53) have shown the greatest ability to out-perform their preseason rankings.

Despite registering a plus-17 mark in 2011, the Trojans scored a negative-25 in 2012, topping 2009 and 1996, in which they concluded the season ranked 18 spots lower than they started in the preseason.

What’s it really mean? Well, on the surface is means that USC and Oklahoma have a knack for letting down pollsters. But the reality is that the data says more about the ideology of polling in college football and the fatal flaws of the preseason poll in general.

The teams at the bottom of the cumulative score include major powers USC, Oklahoma, Texas, Florida State, Nebraska, Michigan and Notre Dame. Minus the absence of Alabama, that list reads like a Family Fued board of the most popular college football teams.

Not only have all of those schools won a national championship since 1997, they’re the schools that are typically given the benefit of the doubt in preseason polls based on how good they’ve been on a long term basis, which in turn makes them far more likely to underachieve than overachieve, given the annual lofty expectations put on them.

Inversely, teams like Oregon, Boise State and Kansas State are considered to have the most “underratedness” due to the fact that they are continuously getting more production from less talent. That variance in talent level gives them the perception of being hard to gauge, which typically winds up in lesser expectations that are more capable of being surpassed.

With STASSEN’s data, and taking into account both a bit of Ewing Theory and elitism, the continuous differential of preseason rankings goes far in showing just how inaccurate preseason polls are, as predictors and evaluators of the past, future and present.

Just because USC and Oklahoma have historically had the worst differential over the last 24 years doesn’t mean that they’re truly overrated programs in relation to Boise State. It means that the rationale of preseason polls are incredibly flawed, as is the notion that they hold some sort of weight in gauging a good team from a bad team.