WOW! Check this out...

[openSkies]

A man named T. Beck keeps a yearly record of ALL the published computer rankings across the nation.

He then ranks them based off of different information, mainly using the straight-up wins vs losses.

This years final rankings were released today. Here is the outcome:

Out of 56 computer rankingss...

(9) Congrove Computer Rankings 119
- This should actually be 6th place, due to 3 compilation-rankings ahead of the CCR 119.

Now, let's take a look at the 6 rankings used in the BCS...

(39) Sagrin
(40) Richard Billingsly
(48) Anderson & Hester
(51) Colley
(53) Massey
(54) Wolfe

Interesting, huh?

=]

[bama_girl]

i wish i knew what you were talking about

[openSkies]

i wish i knew what you were talking about

OK, so here it is a bit more simple (sorry, I even confused myself with the way I wrote that).

CFP is home of the CCR 119 (Congrove Computer Rankings 119). The rankings are created by a computer (through a bunch of mathematical equations).

There are 55 other computer rankings like that in the country.

The BCS Standings are made up of two human polls, and six computer polls, all averaged together in their crazy method.

Each year, this guy, Mr Beck, takes those polls, and looks at how they performed. He then ranks them based on their performance.

So, of all 56 computer rankings in the nation, the CFP rankings (CCR119) is the 6th best (just this year... it's actually one of the lower rankings we've had).

The BCS computer rankings are all miserably wallowing at the bottom of the pool, from 39th to 54th.

Got it now? I think I made it a bit easier to read.

Sorry, I just think in screwed-up ways.

//

Anyways, it was all just 'food for thought'. I think that if the BCS would do away with the human polls, and fix their equation for inputting the computer rankings, it'd be a lot better off.

Or they could just use the CCR 119, nothing else =]

(Not sure if this is public info or not, but a few years ago, the BCS did ask the CCR 119 to join in. Mr Congrove told them that unless they changed the way they add extra equations into the mix, he wouldn't join. They wouldn't change it, so he didn't join. The reason is that the BCS factors in, I believe, strength-of-schedule. Well, the CCR 119, as do most of the computer rankings, already take this into account. So it's being factored in TWICE.)

[openSkies]

Yeah, it really, really is a joke.

[bama_girl]

oh! that's great!

[..fanatic]

The biggst problem I've seen had to do with this year's title game. ESPN kept saying USC was trying to three-peat, completely dismissing the fact that LSU won the BCS title in 2003 and USC only topped the AP Poll. Prior to the BCS, you could call that a split title. You can't say that now. The BCS title IS THE CHAMPIONSHIP, period. The AP was merely a component at the time and it's not even that now. USC was trying to repeat, not three-peat. LSU was completely dissed by the media.

Congratulations, CFP, for never falling prey to that crap.

Their Rose Bowl preview stated..

"Texas finally got the Oklahoma monkey off of its' back and here they are in the BCS Championship. Now, the Longhorns get the chance to try and stop USC from winning their second consecutive BCS title.

USC was also crowned champions in 2003 by the Associated Press while LSU won the BCS Championship"

Drop the polls, let the computers rule. Better yet, just let CFP take over the whole thing! That way, we'd ownly have one source to throw sticks and stones at.

[mountainman]

That is interesting, Matt.

If my memory serves me correctly, the 6 computer models used by the BCS, along with the BCS method of tossing out the highest and lowest computer ranking and averaging the remaining four, had Texas as #1 for every week the BCS published its standing except for week #1.

Now, I don't know what to make of that or what if any conclusion to draw from it, but it is interesting.

[mountainman]

I can do that. Thanks, Matt.

[Spence]

The CCR poll is a good computer poll. The good and bad thing about the computer pollls is that they don't see the games. Good because they aren't infuenced by the media perceptions of one team over another. Bad because they can't really dissect a loss. If an important player goes down during a game and they lose, but that player comes back and they return to form they do not get credit for that. IE: Georgia losing Shockley. I don't know if a computer can ever factor in something like that because it is subjective, but it is a real influence over evaluating a team.

The AP poll is heavily biased toward certain teams.
The Coaches poll is biased by regional perceptions.
The Harris poll exudes zero independent thought. (it is a joke)
The computer polls are only as good as the people who create the formula for the polls.


Still with all the flaws they seem to get pretty close to "right" most of the time. If there were no preseason polls, at least ones that count, I think they would be better. There is no perfect system.

Congrats to the CFP staff for finishing high in comparison to the ones the BCS uses. Maybe down the road the BCS will continue to tweak the system to including the most accurate polls instead of the ones who are willing to change to conform to their policies.

[colorado_loves_football]

I actually am proud of the fact we are associated with the people who created the Congrove Computer Poll, as we all ought to be.
In effect, the BCS could utilize it as the only poll, since (according to Matt) it had better success in predicting the outcome of games, than any other, with respect to the BCS.
We are obviously in pretty exclusive company, and I for one am humbled by it. That being said, I think it would be a disservice to the people who have given us this 'sounding' board if we didn't all agree to utilize it in some fashion, so I for one would like to 'vote' for the Congrove Computer Poll as the 'best' way to select teams in the 'pre-season' match-ups, 2006. I hope you will all join me in that regard, as they obviously have a valuable (invaluable) resource which ought to be utilized in some fashion.