Building a better Sun Belt

How much more impressive is the level of competition in the Sun Belt than it was between 2001 and 2004?

There are two main arguments that are advanced by North Texas supporters who are (rightfully) trying to convince frustrated fans to be more patient with Todd Dodge's rebuilding process.

1) Former head coach Darrell Dickey left the program bereft of athletic talent, particularly on offense.

2) The Sun Belt has improved dramatically since Dickey led the team to a 26 game conference winning streak.

I wholeheartedly agree that Todd Dodge deserves more time to try and succeed, though I qualify that statement by saying that deserving time does not make him immune to legitimate criticism. But I'd still like to address those two arguments to hopefully spark some intelligent discussion about how true the premises are.

The first one I'll address later in the week. In this post, I want to look into the development of the Sun Belt as a football conference.

What's the best way to measure how strong a conference is? There isn't really an established metric for this sort of thing. And to try and measure a difficult-to-quantify concept across time is an even more intellectually challenging task.

Here's the best objective measure I could come up with: The strength of the Sun Belt can best be measured by the conference members' aggregate performance against Division 1 teams from other conferences. Conference games are a zero-sum proposition, and winning games against lower division teams (in my opinion) doesn't seem like an appropriate measure of how Sun Belt teams are or aren't improving relative to the rest of Division 1-A.

Note: I've been using and will continue to use terminology from the recent past. A BCS school refers to members of the 6 auto-bid conferences: Pac 10, Big 12, Big 10, ACC, SEC, and Big East. A non-BCS school would be a FBS (formerly known as Division 1-A) program that is not a member of an auto-bid conference: MWC, WAC, C-USA, MAC and Independents. Teams that I refer to as being 1-AA would be those currently classified by the NCAA as FCS teams.

I went through every schedule for every team in the Sun Belt from every season since 2001 and tallied up the wins and losses against non-Sun Belt Division 1-A teams. The analysis starts with 2001 because it marks the first year that North Texas joined the conference and the start of the 26 game streak. Though I employed several measures to avoid making any mistakes, it is distinctly possible that I may have overlooked one or two games out of the roughly 800 games that I scoured through to come up with these figures. If you think I've made an error, please let me know via email or in the comments section of this post so I can look into it and make any needed corrections.

Here are the numbers:

2001: Combined D-1A OOC record: 3-26
vs. BCS: 1-16; vs. non-BCS: 2-10

2002: Combined D-1A OOC record: 9-29
vs. BCS: 1-22; vs. non-BCS: 8-7

2003: Combined D-1A OOC record: 1-30
vs. BCS: 1-18; vs. non-BCS: 0-12

2004: Combined D-1A OOC record: 4-29
vs. BCS: 1-18; vs. non-BCS: 3-11

2005: Combined D-1A OOC record: 1-25
vs. BCS: 1-17; vs. non-BCS: 0-8

2006: Combined D-1A OOC record: 6-31
vs. BCS: 0-24; vs. non-BCS: 6-7

2007: Combined D-1A OOC record: 7-28
vs. BCS: 3-22; vs. non-BCS: 4-6

2008: Combined D-1A OOC record (as of 10/1/08): 5-21
vs. BCS: 2-15; vs. non-BCS: 3-6.
5 remaining OOC games vs. BCS: Louisville vs. MTSU, Mississippi State vs. MTSU, Alabama vs. Arkansas State, Ole Miss vs. ULM, LSU vs. Troy
1 remaining OOC game vs. non-BCS: UTEP vs. ULL

--------

By this measure, the strongest year the Sun Belt has ever had was in 2002. This was also the only year that the Belt has had an aggregate record against non-BCS teams that was over .500. This feat is already mathematically impossible in 2008.

2007 ranks a reasonably close second, and it has one subjective advantage that I'll address shortly. Interestingly, the worst year for the Belt was the year that the streak was snapped- 2005. And the only season that the Belt didn't win any games against BCS teams was in 2006.

In terms of wins and losses, the Sun Belt isn't performing significantly better (or, compared to 2002, better at all) than it ever has before. The only subjective argument to the objective numbers is the value of "signature wins"; wins over BCS schools.

In 2007, Belt teams tallied three wins over BCS teams. So far this year, they have two. There is a very compelling argument that these "signature wins" carry premium weight. While this may very well be true, there is one other factor at work that may help explain the recent uptick in signature wins: Location.

In 2007 and so far through 2008, Sun Belt teams have played BCS programs at home (meaning in the Belt school's home stadium) five times, tallying a record of 2-3.

In the five seasons from 2002 and 2006, Belt teams only hosted BCS programs in a total of five games. Their combined record in those five games was also 2-3. This count excludes "designated home games;" games where the Sun Belt school was designated as the home team at a supposedly neutral site, such as Louisiana-Monroe's 2004 home game against Arkansas in Little Rock, Arkansas or ASU's 2005 home game against Missouri in Kansas City, Missouri.

Given the identical record in home games during both periods of time, it seems that the recent increase in signature wins may be properly attributed to an increased ability to secure contracts that guarantee true home games for Belt teams.

So, is the Belt significantly better overall than it was during the years of North Texas supremacy? Looking at the numbers, it does not appear that it is. It may only appear that way to North Texas fans because they overestimate other teams' current success relative to the success of the Mean Green from 2001-2004.

A brief digression: I am of the opinion that the quality of the bottom teams in the conference has improved significantly since we bade farewell to some current WAC teams. Since 2004, members of the Sun Belt have lost to Division 1-AA teams only on very rare occasions. Prior to that year, Division 1-AA losses were not uncommon for the lowest ranked teams in the conference. So even though my opinion based on this analysis is that the ceiling hasn't risen much since the North Texas glory years, I do believe that the floor has gotten higher.

There is one tangent of the "much better Belt" theory that this analysis doesn't cover- the notion that even in losses, Belt teams have been more competitive recently than they were in years past. I would like to address this in an objective way, but I can't come up with a consistent and valid way to measure the idea. If anyone else can, I would love to see their analysis. But know in advance that because of the game-to-game factors that can influence scores (willingness to run up a score, annual increases in total scoring, etc.) I don't think mean or median margins of defeat qualify as valid measures of improvement or relative strength.

Personally, I don't think the 'more competitive lately' notion is true, and I don't think that a margin of defeat comparison from year to year would support the argument of people who believe in the 'more competitive' concept anyway. If anyone does look into this, please share your observations with us.


IN BRIEF:

The Belt hasn't gotten significantly better. The Belt overall is almost exactly the same as it has always been since 2001. The only difference is that North Texas isn't on top of the mountain anymore and Belt teams are hosting more BCS teams instead of playing them exclusively on the road.