clock menu more-arrow no yes

Filed under:

Conference Change Impact by the Numbers

New, 39 comments

There's a lot of discussion right now about how the conference change will make it easier for Texas to go undefeated in a season but will hurt both our computer ratings and our media perception, the latter which will affect our human poll rankings.

I can't scientifically address that media perception angle, but let's take a look at the first two hypotheses.

The easier one to look at is the probability of going undefeated in the old conference lineup versus the new. I looked at 2008 and 2009 with their actual schedules and what would probably be the new schedule. For the 2008 real schedule I added the conference championship game against Missouri to Texas' schedule because we would have to have played in it for it to have a real impact on this discussion.

For the revised schedules, in 2008 I removed the road games at UTEP (which we almost certainly wouldn't schedule with only three non-conference games) and Colorado and the supposed conference championship game versus Missouri and added a road games at Iowa State and Kansas State; in 2009 I removed the road game at Wyoming, the home game against Colorado, and the championship game versus Nebraska and added home games against Iowa State and Kansas State.

The result was that our probability of going undefeated in 2008 changed from 16.1% with the real schedule to 20.4% with the revised slate and in 2009 the change was from 16.7% to 19.2%. So the obvious is confirmed, that Texas' chances of going undefeated improve in the new setup if opponents' strengths are held constant. However, the change in probability may not be as much as some had expected.

Now, for the trickier part, the computer rankings. A full analysis of this question would take longer than I'm willing to give it, so I took a couple of shortcuts. I don't have any of the BCS algorithms coded, but I do have my NMV ratings that closely approximate Massey's ratings each season and Colley lets you play with his by adding and changing games.

Colley's matrix rankings are only online for 2009, but for that season if we make the changes above then Texas' ranking with the new schedule (I also removed the bowl games for Alabama/Texas and Florida/Cincinnati) before the bowls stays #2 but the gap between Texas and Cincinnati shrinks from .02886 to .01566. The lead is roughly half what it was. Unfortunately this feature is not available for past seasons so I could only look at 2009 for Colley's ratings.

In the NMV ratings, 2008 Texas would have a 93.73 mark and rank #2 before the bowls if they had beaten Missouri in the championship game instead of Oklahoma. With the new schedule style the Longhorns have a 92.37 rating and a #4 ranking. However, it's not quite that bad. Texas fell behind Florida and Texas Tech, and adjusting the Red Raiders' schedule keeps Texas ahead of them.

That being said, clearly the new schedule will hurt our computer ratings by as much as many are thinking. We aren't as weak as a Big East team but we now have little to no shot at being ahead of an SEC team with the same number of losses and would be in a dogfight with a Pac-10 team instead of having a leg up on them.

So my conclusion? Those that are hailing this as a good move for making the BCS title game more likely are missing the mark. At best it's a wash between the certainly lower computer ratings and the slightly higher probability of going undefeated, but when you add the human polls I think that this is probably a net negative to BCS title hopes.

It seems to me that during the realignment uproar Texas fans have gotten used to thinking of our conference mates in terms of their market appeal and forgotten to think of them in terms of their actual threat level on the field. And while Kansas State, Oklahoma State, and Texas Tech are indeed much lower profile than Texas, they do in fact field teams that can beat us from time to time.