Wednesday, September 8, 2010

Money for Nothings, Stats for Free

Only in college football can you count the stats against teams that themselves don’t count.

As I was reviewing every game this past weekend for my BlogPoll vote, I was struck by number of FBS teams that played lower division, FCS opponents in week 1. Miami played Florida A&M, FSU played Samford, Penn State played Youngstown State, and so on. It seemed to me to be a greater number of these types of games than I had ever seen before.
After a little research, it turned out I was right.
To start this 2010 college football season a record 39 FBS teams played FCS opponents. That’s right – 1 in 3 teams in college football’s premier division opened with lower division opponents.
Going back to the last 2 years, the trend is definitely to the positive –
FBS teams playing FCS teams for season opener
2009 – 34
2008 – 31
This is a relatively new, and growing, phenomenon. In 2005, for instance, only 14 FBS teams opened their schedule with a lower division opponent. And if we go back to 1997, the first year for the BCS, we find only 4 such games.
1997 FBS (formerly division I-A) openers against FCS (formerly I-AA) –
Idaho at Air Force
Marshall at West Virginia
UAB at Kansas
UCF at Mississippi
(Interestingly, Idaho, Marshall and UCF have since been “promoted”)
Just what in Sam Houston State is going on here?
Well the obvious answer is money. Money, and easy wins.
FCS teams are richly rewarded to throw themselves on the swords of FBS teams, while FBS teams get to sell fans cut-rate games for full price tickets.
But the FBS teams get something more for their money – manipulation of the polls by ignorant and gullible voters.
Not only do FBS teams get a near automatic win (absent a rarity, like Ole Miss), they get a chance to pad their stats significantly. Take a look at the following –
Top 20 Scoring Offenses Year-to-date 2010
1          Oregon 72
2          Houston           68
3          Air Force         65
3          Oklahoma St.   65
5          Florida St.        59
5          South Fla.         59
7          Georgia            55
8          Arizona St.       54
9          Wake Forest    53
10        Auburn 52
10        California          52
10        Stanford           52
13        Indiana 51
14        Tennessee        50
15        Mississippi St.  49
15        Nebraska         49
15        Nevada            49
15        Southern California       49
19        Alabama           48
19        Mississippi        48
19        North Carolina St.        48
19        Texas A&M     48
Of this list only 7 teams – Oregon, Oklahoma State, Georgia, Mississippi State, Nebraska, USC and Alabama even played FBS opponents in week 1.
The offical NCAA statistics make no distinctions for the division of a FBS team’s opponents, so Florida State’s 59 points against Samford (as well as their yardage, and all other stats) will count towards their official season stats. In a 12 game regular season schedule, the score in the Samford game will add 5 points per game to FSU’s scoring offense, a significant difference.
Just to point out how absurd this really is, take a look at one of the teams tied for 19th – Ole Miss – who scored 48 points in a losing effort to Jacksonville State. Notably Jacksonville State would rank the same on this list as the group at 15th with 49 points, were they eligible for it. Only in college football can you count the stats against teams that themselves don’t count.
Not only do these lay-up games pad the stats, they promote an early image about the team, as the numerous articles touting the big win in Jimbo Fisher’s first game can attest to.
It’s pretty clear by the accelerating trend of FBS teams to schedule lower divisional competition that the schools doing so see very little downside, and certainly no impediment as it relates to their ability to compete for the mythical national title.
The downside is, of course, what was formerly the unthinkable in the pre-Appalachian State days – a FBS team losing to a FCS team. Ole Miss and Kansas were similar victims to start this year. The irony of Mississippi and Kansas losing to FCS teams is almost too delicious to contemplate, as both teams are among the most serial offenders in scheduling lower division teams (see the 1997 example above). Kansas, in fact, has had a least one team to open its season for every year I researched.
As far as FBS teams losing to lower division opponents, with the trend in place it’s going to become routine rather than shocking. It’s a matter of simple math. With only an assumed 1% chance of victory for your typical FCS team against a FBS team, 39 games would equal a 39% chance of such a win occurring, or in the case of this year, a 15% chance of two FBS teams losing. As more FCS teams get added look for on average one such upset to start each season.
As far as downside goes, I would say Ole Miss’ and Kansas’ seasons are all but over.
And that, and maybe only that, will act to discourage this otherwise odious trend.


Jason said...

Good analysis, as always. It's deplorable, no doubt it. Florida does good to not do so for the opening, mostly, but then we invalidate that effort by scheduling I-AA teams later in the year. I'm not asking for use to go the game route as LSU and Alabama has with their OOC kickoff game - the pragmatist in me understands it won't happen unless there is good money behind it - but there is no excuse for featuring games vs teams that might qualify as glorified scrimmages, the random Jacksonville St upset not withstanding.

Clark said...

FCS teams win far more than 1% of their games against FBS teams. This year they are 2/39=5%. Last year they won 5 of 91 (Villanova/Temple, William&Mary/Virginia, Richmond/Duke, New Hampshire/Ball State, Central Arkansas/Western Kentucky). Based on this data, 5% looks like a much better estimate.

I can't figure out why it isn't obvious that games against teams in lower divisions shouldn't count. How did this even start?

shawn said...

A short comment on statistics.

If as you assumed there is a 1% chance of FBS beating an FCS team then the cahnces that the FCS wins 39 games is - .99^39 or 67.5%. That means that the lesser schools will win at least one game 32.5% of the time.

The chances of the big schools wining 38 times is 26.5%. Adding that to the 67.5% chance of 38 victories leaves the FBS schools with a mere 6% chance of achieving 2 or more wins in 39 tries.

Mergz said...

Shawn - You are exactly right, it was a rushed/lazy error on my part.

Clark - I was using 1% more as an example/placeholder than something based on actual evidence. Your number is evidence based and probably much more accurate as far as reality goes.

As far as not counting the games, I think it is absurd that teams that don't even play in the same subdivision count for anything. None of those 39 teams is eligible for the BCS or any bowl. Were it up to me teams would be unable to choose who they play - schedules would be conference mandated. It's the logical equivilant of allowing NFL teams to schedule a college team them count the win towards getting in the playoffs.