longhornsIt’s here.  It’s finally here!  No, I’m not talking about the RNC.  I’m talking about football.  Vanderbit vs. South Carolina officially begins the beauty contest that is the BCS in its current form.  Preseason rankings set the pecking order for the BCS from the start.  So, we’ve conducted an analysis of the preseason AP Top 25 vs. the final AP Top 25 for every season since the AP changed their ranking system to a Top 25 back in 1989.  We started with the information at collegepollarchive.com.  What we were looking for was a pattern of teams being consistently overrated (or underrated).  Our methodology was simple.  For each year, take a team’s final ranking and subtract their preseason ranking, then average that out over the total number of years they showed up in the poll.  We’re only considering teams that have been consistently in the polls (at least half of the seasons in question).  If a team falls completely out of the poll, we used their SRS rating.  SRS takes into account points for, points against, and strength of schedule.  We’ve included a more detailed description of the methodology at the end so that readers with mathematics allergies are not exposed to it.

Since 1989

The first study was looking at the entire period from 1989 to 2011, during which the following teams were the most consistently overrated:

UCLA: -12
Clemson: -9
West Virginia: -9
Texas: -9

No surprise that Texas  is among the leaders here.  Texas, on average, finishes 9 spots lower than where writers picked them in the preseason.  UCLA’s showing is surprising now, but not if you remember back to the early 90’s.  They would be ranked highly because their talent looked good on paper.  Then they would finish in the middle of the Pac-10 (pun intended).  Hope springs eternal for Clemson and West Virginia, regardless of whether hope is warranted or not.

On the opposite end of the spectrum, there was one team that stood out for being consistently underrated:

Kansas State: +2

Keep in mind that we give each team that starts the season outside the Top 25 a rank of #26.  So you can lose a lot more than you can gain, but there’s no way to quantify where an unranked team really stands in the preseason.  Anyway, Bill Snyder can coach some football, but it’s usually well into the season before you can tell how good the Wildcats are due to their flaccid non-conference schedule.

2001 to 2011

For a second look at the numbers, we decided just to look at the last 11 years of the Top 25 (2001-2011).  Again, the team had to be included in the polls at least 6 times during the period to be considered “consistently overrated.”  Here were the results:

Most overrated:

Arizona State: -11
Louisville: -9
Pitt: -8
Nebraska: -7
Tennessee: -7
Texas A&M: -7
Washington: -7
Texas: -7

Again, a Pac-10/12 team sets the pace.  Texas makes another strong showing even with a recent national championship skewing their average.

Most underrated:

Boston College: +7
Alabama: +5
Arkansas: +4

It might seem surprising that Alabama could make a list of underrated teams, but they consistently exceeded expectations in the early Saban years.  More recently, it’s hard to be underrated when you’re always in the top 5.  BC being that consistently underrated is very surprising, until you look at the league they play in (see conference ratings below).

2006 to 2011

So breaking it down to even more recent performance, here’s a look at which teams consistently got more preseason love than they deserved over the last 6 seasons.

Georgia Tech: -13
Louisville: -13
California: -12
North Carolina: -12
Kansas: -7
Illinois: -7
Texas: -7
Arizona State: -7

The Big 10, ACC, Big East, Pac-12, and Big 12 are all represented here.  Georgia Tech was a sexy pick for a few years.  Bobby Petrino set the bar high at Louisville, and then abandoned them (as he has everyone else).

The most underrated teams over the last 6 years:

Utah: +13
Stanford: +10
Cincinnati: +9
Houston: +9
Boston College: +8

Congratulations to these teams.  In almost every case, if your team shows up on this list, your coach got hired away.

Teams noticeably absent from my analysis

You might be wondering, “What about Notre Dame, Michigan and Boise?”  Notre Dame was actually high on the list, but just missed the cut-off for the period 1989-2011.  Their average finish in relation to their starting position in the polls was -7.  But, that was also the same number for Oklahoma, Nebraska, and Miami — other big brand names.  In a recent trend, Michigan has been discounted by pollsters, and that has helped their average.  Historically, they’ve had a tendency to outperform when they get a ranking outside the top 10, which keeps their average down.

Boise State did show up on the underrated teams,  but they burst on the scene so recently, they didn’t have time to get into enough polls to qualify as consistently underrated until the last six years.  However, they have been showing very high in the preseason polls here lately, which makes it hard for them to exceed expectations.  On average, they were +2, though they’ve had a few years where they were unranked to start and ended up near the top.

Get ready for this, SEC-haters

As our last bit of analysis, we compared the combined performance of the teams in a particular conference.  Here’s how the BCS conferences stacked up over the two time periods:


ACC: -6
Big 12: -6
Pac-10/12: -5
Big East: -5
Big Ten: -4
SEC: -3


ACC: -5
Big 12: -4
Big East: -3
Big Ten: 0
Pac-10/12: +1
SEC: +2

As you can see, over the longer period (1989-2011) all of the BCS conferences were overrated.  The preseason poll is always dominated by the power conference schools.  As the folly of that bias becomes evident and winning percentages dip, the WAC, MWC, MAC, and Conference USA teams (sorry, Sun Belt) are invited into the poll, displacing BCS schools.  As a side-note, I didn’t include the old SWC, though it had the highest number (-6 — thank you, Texas).

Surprisingly, even with all the success that the SEC has had, the conference is still being undervalued in the preseason polls over the last 6 seasons.  Seems totally improbable since the league has produced the last 5 national champions.  But as the Ol’ Ball Coach recently quipped:  “It’s easier to win the national championship than the SEC. Ask Nick Saban.”


Methodology (Warning:  Do not read past here if you are narcoleptic or ADD.)

For teams that didn’t begin the season in the polls, we assumed a ranking of 26.  For teams that dropped from the poll completely, we used their SRS (Simple Rating System) rating as provided by sports-reference.com.  SRS takes into account points scored, points allowed, and strength of schedule.  For example, if a team started the year ranked #3 and fell to #20, their score for that year would be -17.  However, if they started out unranked and ended up at #10, they would receive a score of +16 for that year.  For each period we researched, we assumed that the team needed to be included in the poll for at least half of the years during the period being researched.  This kept the numbers from being eye-popping by outliers or flash-in-the-pan team that just happened to have one good senior class, or like South Florida who had one good year to set expectations high, then followed it up with horrible play.

For the conference averages, we simply took the yearly +/- performance for each school in that conference and added it to the total for that conference, then divided by the total number of appearances of schools from that conference in the Top 25 during the time period under analysis.

About Elton Conrady, Technical Editor

Elton Conrady is Technical Editor at Media Rostra. As a network engineer, Elton has worked in traditional telecom, wireless, state government, and DoD environments. While engineers are considered dull or nerdy in some social circles, Elton has attempted to be an interesting person -- he has played on a barnstorming basketball team in Europe, worked as a commercial fisherman off Kodiak Island, and lived in Croatia for 4 years. Now that he is "grown up", Elton spends his spare time fishing, camping, and hiking.