Week Two of the BCS standings brings even more turmoil that Week One. With two undefeated teams going down last weekend, the seven remaining are scrambling to remain in contention for the crystal ball. With only two of the still-undefeated teams playing each other, the strength of schedule becomes more and more important, as the year goes on. That strength of schedule, of course, is based primarily on conference play, and, thus, on how strong your conference is.

Now, I know that I said that I wouldn't rehash the method of calculation again, but my methods changed, sorry. Using the suggestions from last week's comments, I have come up with three new formulas (and datasets) to base the scores on, then use them to create the final score.

The average conference points is, simply put, the same conference points from last week, simply divided by the number of schools in the conference, then divided by two (this was to keep the potential points on target with the rest of the components). This calculation shows how strong the average team in the conference is, in the voters minds. The parity is a calculation of the in-conference record (even is worth 10, +/- one is worth 9, etc.), averaged out. This calculation shows the average strength of a team in a conference vs. another team in the same conference, thus showing how well balanced the conference is.

There is then the OCC score, which is the out of conference record (the w-l is the total points a team gets) averaged out, then multiplied by 2 (to even out the points). This calculation shows how well the average team in one conference does against the average team in all the other conferences (including FCS). Finally, there is the final score, which is simply an average of the three previous scores. This score should show which conference is the best overall, and when you take an average team from one conference, how it would do against an average team from another.

Okay, now, hopefully, I will never have to repeat those again...

Continue reading this post »

When trying to decide who has the best conference in college football, some will argue that is it in the past, where previous success dictates current positioning. Is it the number of BCS championships within the last 5 years (SEC has that one in the bag)? Maybe the most BCS NCG appearances (Hello Big 12)? Or maybe it is simply the most BCS bowl games (Big Ten)?

Others will argue that it is the current season that matters. Is it the Pac-10 where the best team is averaging .9 points a minute? Maybe it is the MWC, who has two members in the Top-10 (add one for next year)? Or, is it the Big 12 who has the most undefeated teams (for the rest of this week, that is)?

When you dare to start a debate over the best conference in college football, everybody comes out of the woodwork, and they start to argue any position conceivable, provided it supports their chosen conference. No matter what they argue, however, one thing is clear – everything is based on a biased and highly chosen variable. People will often pick the factor which leads either their team, or their conference, to the top. They will then sit their in their self-indulgence, knowing that their conference is the best in the world.

An intro to my system:

With that in mind, I started to consider various 'fair' ways to determine the proper positioning of the conferences. Working closely with Hilary last year, I came up with four variables, which most people would agree with, which show which conference is the best in the nation. Using BCS polling (arguments of BCS bias aside), we can determine the best conference on the following factors: the number of schools ranked; the ranking points for those schools ranked; the parity of the conference; and the 'final' score (a combination of all three).

As this is the first post in this weekly series, I think that it will be helpful if I explain how each of these factors works. The number of schools ranked is, well, somewhat obvious. The conference with the most schools in the Top 25 BCS poll will earn this spot. The conference points is also easy to figure out. Take all the teams ranked, assign them points based on their position (25 points for the 1^st, 24 for the 2^nd, etc.), and then add them up (there are partial points awarded for just dropped out teams). The parity is slightly harder to understand, but it is still a clear factor for deciding a winner. To find the parity, take the total schools in the conference, and find the percentage that is ranked; this allows smaller conferences with one or two ranked schools to be comparable to the larger 12-school 'super' conferences.

Finally, there is the 'final' score, which combines all of the previous three scores – thus making it the most complete measurement. To find the final score, you take the parity and multiply it by the conference points. Doing so penalizes conferences that are "top-heavy," those with two or three great teams and a lot of crap below.  Then, take that number, and divide it by the max possible score for that conference (this is found by assuming that all schools are ranked 1-n, where n is the number of schools in the conference), which shows how close to perfect the conference really is. This number is turned into a percent, and, simply put, shows you the actual value of a conference, with all other variable accounted for.

Now that that is out of the way (don't worry, I won't bother explaining it again in the future), let's see what this means for this week...

Continue reading this post »