FanPost

"Jeremy's Method" for determining who should play for BCS NC

Some of you might recall my fanpost about this last year. Here is is for a refresher.

Brief synopsis:

I am using Sagarin rankings because they include all FBS and FCS teams in one composite ranking system.

There are two components, "raw Sagarin" ranking (this is SUBTRACTED when a team loses a game) and "reverse Sagarin" ranking (this is ADDED when a team wins a game). Since there are 246 teams in FBS and FCS, reverse Sagarin is simply a value of 247 - raw Sagarin (example, prior to today's games, Alabama is #1 in Sagarin rankings, so if a team loses to Alabama, 1 is subtracted from their score, and if a team beats Alabama, 246 is added to their score).

Keep in mind: I am anxious to see these numbers, so I am using the Sagarin rankings at 1:19 am on Sunday 11/11/2012 (these are the rankings PRIOR to the games played on 11/10/2012). I am also only evaluating KSU, Oregon, Notre Dame, Bama, and Georgia (top 5 teams prior to games played on 11/10/2012). I will update this as the weeks progress and we get closer to the BCS show on 12/2/2012 just to see if the BCS "gets it right".

Here goes

KSU: raw score through 10 games 1937, average 193.7 // if today's rankings are used and assume KSU wins out raw score is 2375, average 197.9

Oregon: raw score through 10 games 1757, average 175.7 // if today's rankings are used and assume UO wins out raw score is 2230, average 185.8 (Pac 12 champ game not included)

ND: raw score through 10 games 1985, average 198.5 // if today's rankings are used and assume ND wins out raw score is 2351, average 195.9

Bama: raw score through 10 games 1769, average 176.9 // if today's rankings are used and assume Bama wins out raw score is 1982, average 165.2 (SEC champ game not included)

Georgia: raw score through 10 games 1576, average 157.6 // if today's rankings are used and assume UGA wins out raw score is 1911, average 159.3 (SEC champ game not included)

According to this if season ended on 11/10/2010, the NC game would be ND (#1) vs KSU (#2). If the season plays out with these 5 teams winning out (using Sagarin rankings of 11/10/2012), NC game would still be ND (#2) vs KSU (#1).

Surprisingly (or not really) this confirms the CBS "blind resume test". Based on Saturday's results, the above numbers will change slightly as Sagarin rankings are adjusted. I guess I will start doing this on Monday from here on out, but as I mentioned I was anxious to see what this method would produce after the Bama upset.

I still don't have a good way to incorporate margin of victory into this method. If any of you have a suggestion, please let me know and I will see what I can do to improve the calculation.

UPDATE 11/11/2012:

Here are the updated results based on the most recent Sagarin rankings. It should be mentioned that I am using the composite Sagarin rankings, which is a composite of the ELO_CHESS and Predictor rankings. The Predictor rankings already take MOV into account, so after some further consideration, I am not going to try incorporating my own MOV component into these calculations.

I decided to add Florida into my comparisons, and the results were quite shocking.

KSU: raw score through 10 games 1933, average 193.3 // if today's rankings are used and assume KSU wins out raw score is 2384, average 198.7

Oregon: raw score through 10 games 1797, average 179.7 // if today's rankings are used and assume UO wins out raw score is 2268, average 189.0 (Pac 12 champ game not included)

ND: raw score through 10 games 1981, average 198.1 // if today's rankings are used and assume ND wins out raw score is 2340, average 195.0

Bama: raw score through 10 games 1778, average 177.8 // if today's rankings are used and assume Bama wins out raw score is 1979, average 164.9 (SEC champ game not included)

Georgia: raw score through 10 games 1591, average 159.1 // if today's rankings are used and assume UGA wins out raw score is 1937, average 161.4 (SEC champ game not included)

Florida: raw score through 10 games 1820, average 182.0 // if today's rankings are used and assume UF wins out raw score is 2149, average 179.1 (Florida will not be playing in the SEC champ game)

According to this if season ended on 11/10/2010, the NC game would be ND (#1) vs KSU (#2). If the season plays out with these 5 teams winning out (using Sagarin rankings of 11/11/2012), NC game would still be ND (#2) vs KSU (#1). The updated Sagarin rankings did not change the overall results.

As I mentioned above, I was very surprised with respect to Florida. Prior to the games yesterday, the general perception (based on BCS standings anyway) was that UGA was the best of the one-loss teams. Using this method, we actually see that if the season ended on 11/10/2012, Florida is not only the best of the one-loss teams, they are even better (considerably) than an undefeated Oregon.

With the exception of ND being more deserving than Oregon (and I really don't think you will find that much unrest on a national level about this), this method actually falls in line with what most people feel in their gut about who the deserving teams are (when compared to blind resume tests and among people on BOTC, anyway).

As a refresher, this method showed that Okie State was more deserving than Bama to play in the MNC game last year.

All comments, FanPosts and FanShots reflect only the view of the user creating them.

X
Log In Sign Up

forgot?
Log In Sign Up

Forgot password?

We'll email you a reset link.

If you signed up using a 3rd party account like Facebook or Twitter, please login with it instead.

Forgot password?

Try another email?

Almost done,

Join Bring On The Cats

You must be a member of Bring On The Cats to participate.

We have our own Community Guidelines at Bring On The Cats. You should read them.

Join Bring On The Cats

You must be a member of Bring On The Cats to participate.

We have our own Community Guidelines at Bring On The Cats. You should read them.

Spinner.vc97ec6e

Authenticating

Great!

Choose an available username to complete sign up.

In order to provide our users with a better overall experience, we ask for more information from Facebook when using it to login so that we can learn more about our audience and provide you with the best possible experience. We do not store specific user data and the sharing of it is not required to login with Facebook.

tracking_pixel_9347_tracker