Big Ten lays bowl egg

[Derek]

Guess that's why we need the computers as one-third of the BCS Rankings.

Guess that's why we need the human element as the whole of the BCS Rankings.

HOGWASH!!

Who do you think put them there?????????????

That makes no sense.

[donovan]

Guess that's why we need the computers as one-third of the BCS Rankings.

Guess that's why we need the human element as the whole of the BCS Rankings.

HOGWASH!!

Who do you think put them there?????????????

That makes no sense.

Hogwash?.... Not sure who them refers to....Must remind those that live in this generation, that football championships, games....absolutely same discussions were held long before computers were invented.........

Heck...do not play the games....just let Nitendo decide....

[Derek]

It's not Nintendo. It's 6 different computers written by different people that get the first and last results thrown out. It is a VERY well designed system.

And you brought up what happened before the computers...I submit that without the current system Boise would not have been given the chance to play Oklahoma. The good ole boy Network, consisting of (ND, Big 10) would play for the championship every year.

Do you really think the computers would have ranked ND high, without the support of the human polls. They then get into a bowl game against a real team, and get clobbered. Just like USC and Michigan did to them.

How you cant see the problems that the biased human polls bring, I dont know. And you are definately in the minority on this board.

In your world, Notre Dame would be in a BCS game every year, deserving or not.

[donovan]

It's not Nintendo. It's 6 different computers written by different people that get the first and last results thrown out. It is a VERY well designed system.

And you brought up what happened before the computers...I submit that without the current system Boise would not have been given the chance to play Oklahoma. The good ole boy Network, consisting of (ND, Big 10) would play for the championship every year.

Do you really think the computers would have ranked ND high, without the support of the human polls. They then get into a bowl game against a real team, and get clobbered. Just like USC and Michigan did to them.

How you cant see the problems that the biased human polls bring, I dont know. And you are definately in the minority on this board.

In your world, Notre Dame would be in a BCS game every year, deserving or not.

1.You assume that I championed the cause of Boise playing in a BCS bowl....I did not....do I think they deserved to...absolutely.

2. No time did I ever say the computer systems were not well designed...I just happen to believe that on this subject, human evaluation has worked for years. At least the cantankerous discussion is about the same.

3. I am disinterested in Notre Dame. Do I think humans ranked them higher than computers...well...maybe...but under your system they got there...and mine...they may have not. (Reality is...people like that money...so does not matter..human...computer...they are going)

4. I wonder why you can not see the problems in computer rankings. e.g. The computer ranking on this site..Congrove...supposedly the most accurate in prediction...by their statements....ranks the teams totally different than your celestial group. So whose computer is right?

5. Being a minority on this site is some kind of sin?...I think it may well be a virtue!

[Howdy]

It's not Nintendo. It's 6 different computers written by different people that get the first and last results thrown out. It is a VERY well designed system.

And you brought up what happened before the computers...I submit that without the current system Boise would not have been given the chance to play Oklahoma. The good ole boy Network, consisting of (ND, Big 10) would play for the championship every year.

Do you really think the computers would have ranked ND high, without the support of the human polls. They then get into a bowl game against a real team, and get clobbered. Just like USC and Michigan did to them.

How you cant see the problems that the biased human polls bring, I dont know. And you are definately in the minority on this board.

In your world, Notre Dame would be in a BCS game every year, deserving or not.

____________________________________________________________

61 people in the AP vote but only 6 from the computer files vote.
Not an even balance if you ask me.

[billybud]

All marketing ballyhoo I take with a grain of salt....The Following are NCAA PT Computer Rankings Awards.

2005 Computer Ranking Winners....By Category.

SMALLEST DEVIATION FROM ACTUAL GAME SCORES (Entire Season)

Winner: The Vegas line
2005 Winner: Vegas Line
2004 Winner: Vegas Line
2003 Winner: Vegas Line
2002 Winner: Vegas Line
2001 Winner: Vegas Line
2000 Winner: Vegas Line
1999 Winner: Vegas Line


MOST ACCURATE PREDICTOR (Entire Season)
Winner: The Vegas line

2005 Winner: The Vegas Line
2004 Winner: The Vegas Line
2003 Winner: The Vegas Line
2002 Winner: The Vegas Line
2001 Winner: The Vegas Line
2000 Winner: The Vegas Line
1999 Winner: The Vegas Line

MOST RETRODICTIVE WINS - Small System
Among the small systems the best retrodictive winning percentage was a
tie between Jeff Sagarin and Warren Claassen, both were 551-112, 83.11%.

Several of the BCS ratings were the runner ups, Sagarin, Wolfe,
Colley and Massey were all within 3 games of the lead.
The record was 86.02% set by logistic regression in 2004.

2005 Winner: Jeff Sagarin, Warren Claassen
2004 Winner: Logistic Regression
2003 Winner: Anderson/Hester
2002 Winner: Logistic Regression
2001 WInner: System Average

Among the small systems the best retrodictive winning percentage was a
tie between Jeff Sagarin and Warren Claassen, both were 551-112, 83.11%.


MOST STRAIGHT UP WINNERS (Second Half of Season)
2005 Winner: Dunkel Index
2004 Winner: Jeff Sagain, Wesley Colley
2003 Winner: Born Power Index & Edward Kambour
2002 Winner: Born Power Index
2001 Winner: Chris Montgomery
2000 Winner: Geoff Freeze

MOST ACCURATE PREDICTOR (Second Half of Season)
Winner: Vegas Line

Predictably, the Vegas line gets the award for most accurate predictor over the second half of the season. The computer adjusted line and the opening line came in second and third. The best manned system was the Dunkel Index.

As good as the line was this year it still had a mean square error higher
than Ashby Accuratings had last season.

2005 Winner: Vegas Line
2004 Winner: Ashby Accuratings
2003 Winner: Vegas Line
2002 Winner: Vegas Line
2001 Winner: Vegas Line
2000 Winner: Vegas Line

SMALLEST RETRODICTIVE MEAN ERROR - Large System

Winner: Least Square Regression with team HFA

Least squares regression with individula team HFA finished 0.60 points
better than the second best system for the second year in a row.
Being a 'large' system did not give much of an advantage this year as the
2nd, 3rd, and 4th place systems were all simple systems.
The winning mean error this year was 9.3118, which is a little worse that
the record of 9.206 set last season.


2005 Winner: Least Square Regression with team HFA
2004 Winner: Least Square Regression with team HFA
2003 Winner: The Sports Report - SLOTS
2002 Winner: The average of all systems.
2001 Winner: Edward Kambour Football Ratings
2000 Winner: CPA Rankings

SMALLEST RETRODICTIVE MEAN ERROR - Small SystemWinner: Jeff Sagarin

The winner of the small class ratings is Jeff Sagarin's points or predictive ratings, with a mean error of 10.0129. This breaks least square regressions two year reign and sets a new record.

2005 Winner: Jeff Sagarin Predictive
2004 Winner: Least Squares Regression
2003 Winner: Least Squares Regression

[donovan]

There is, I think, a very basic difference between the Vegas Line and the rest....and I believe it substantiates my position. Sagarin is strict computer analysis; but the Vegas line is a reflection of betting continual adjusted to balance the books. Betting is done by humans......

this link explains the Vegas Line and the player bias used.

http://www.sportswire.com/phl/faq/sport ... asline.asp

[billybud]

As someone who has familiarity with Vegas lines as well as testing for validity, I can tell you one thing...

1...validity doesn't care how it is derived...it is a value that is proven by results..and the Vegas line is, beyond a doubt, about the most valid college football predictor you have.

But the beauty of the Vegas line is that it is self adjusting..they aren't trying to identify a team that will make the point spread, the main objective is that their clients get equal action on both sides. By identifying the approximate handicap a team has while playing another team, they’re trying to make it a coin flip, a tough decision (for the bettor). And while doing that, the system is the most valid predictor of who will win a college football game.

[donovan]

As someone who has familiarity with Vegas lines as well as testing for validity, I can tell you one thing...

1...validity doesn't care how it is derived...it is a value that is proven by results..and the Vegas line is, beyond a doubt, about the most valid college football predictor you have.

But the beauty of the Vegas line is that it is self adjusting..they aren't trying to identify a team that will make the point spread, the main objective is that their clients get equal action on both sides. By identifying the approximate handicap a team has while playing another team, they’re trying to make it a coin flip, a tough decision (for the bettor). And while doing that, the system is the most valid predictor of who will win a college football game.

I certainly do not disagree with any of this. My point was only this. Wide spread betting...i.e. a lot of human imput....is an important ingredient of making a valid prediction. These people move the odds to entice people to bet on certain teams and have a balanced booking.
It is the same idea that the wider the group of opinions the closer you come to truth. Not always...but most often. In some circle we call that democracy.
I just get amused at people that argue computers have the ability to correctly predict. Computers compute. Imput data is the basis and as the Vegas Line explains...they have wide imput. Others may also...

[billybud]

Donovan...Oh, I agree...the human betters are using their carbon based computers to figure at what point they will make a bet...and the tens of thousands of such bettors may "adjust" the betting line...which results, in turn, a valid predictor of who will win the game and an approximation of the point spread in the game.

The Vegas line, of course, can't be used to compare two teams that don't play each other and that is why computers are used...they offer a simulation of sorts.

That is why I like a computer system that has algorithmns that rate teams fairly close to the Vegas line (when they play each other, the Vegas line can be checked against the computer rating)...

[Swamp Daddy]

So now the Las Vegas line is correct; and, unreproachable? So we now vote (bet) on whether you call the temperature outside warm or cold. And the most votes is what we write in the history book. HMMMM.

Gee, I guess no one has to show up for a game anymore; we can go by the line.

More relevant to the Las Vegas line would be statistics on what teams continually beat the line; and, what teams continually are worse, or much worse than the line (Notre Dame maybe). ESPN poll before UL/WV game had voters in 49 of 50 states predicting a WV win this year. Are these voters any smarter than betters who make up the line? Don't think so.

1) Do I want a national playoff: not sure yet
2) Do I want only biased media voting for bowl game particpants: NO FOR SURE, too much baggage there in where they went, where they write,etc..
3) Do I want computers to do it all: I'd have to see the programming to check for any bias there; but, probably it should not be all computer. Generally speaking computers "average" a whole season. Suppose an injured player returns and a team goes wild. Maybe a key player or two is injured in the last game and a team is thereby not as good as the 'whole season' would indicate.
4) Do I like the current system: Not totally, but it works fairly well.
5) My suggestion for improvements: Hmmm, maybe have lesser conferences as a whole always provide one participant past any from BCS standings.

Am I a 'real' expert. No far from it.

I'm still wondering how to turn all these bucks for posting into ATTABOYS at work.

Swamp Daddy

[donovan]

Hard to be redundant on this. The Vegas line is accurate not because it reflects the best team or who they think will win. Bookmakers do NOT care who wins. Vegas does not care who wins. Odds...the line...are used to entice people to bet on the team where less money has been bet. Bookies would like an equal amount bet on Team A and Team B. They make their money from the vig or vigorish....a commission. That percentage varies from type of game to type of game. For instance..the vig in roulette with double zeros is about 5.25%.

So why do the oddsmakers care about past performance and injuries and all of the statistics...they use that to manipulate the way people bet. That is their sole purpose. Oddmakers never make win predictions. Not their job or purpose.

Polls...predictors..etc...are another thing. You can not mix the two.
Their purpose is to try and be right on the prediction.

The reason people look at the Vegas line and say it is accurate is because the majority of people bet for the team that won; hence the odds favor that team because the enticement was to get people to bet for the losing team in order to balance the books with equal bets. That is why street corner bookies lay off bets to other places....

Every gambling game has the same philosophy. The house..bookie gets the vig and they could careless about payouts, who wins, etc as long as the books are balanced.

So are computers more accurate than humans....when it comes to calculations, especially when complexity and speed come into play.

We all can add one plus one as quickly..or quicker than a computer....but for most of us..it ends there.. When complex algorithms are involved...computers leave humans in the dust....and that ..as in every other industry is what is happening with sport analysis. Is it good to rely just on that....no more than business decisions....there is always a human element....In pro gambling..it is the people that bet..where are they putting their money....they all calculate that into the formula.

I like the mix of the two....I think leaving a good portion of human element is a good thing....I do not want the robotic society.....

Play the game....absolutely.....We like watching the game...if it was only the winner..then read the paper...they print the scores....

For me...back to the old bowl systems........why... I liked it better....seems best for me... And..would Boise played Oklahoma in the Fiesta...NO...and as I have said zillion times...I am ok with that....Let those of the SEC that forsake education for multi million dollar coaches worry who is the best of the best....(Need some controversy.)

[Spence]

Let those of the SEC that forsake education for multi million dollar coaches worry who is the best of the best....(Need some controversy.)

None of the money used to hire coaches is expected to be made up by the general fund of the university. Athletic departments have to find ways, beyond the general fund, to cover expenses.

As far as the SEC eduacation or any major conference for that matter, it just isn't true (anymore) that athletes don't have to be "student athletes" or that you can't get a good education from these schools. The last part was never true. Look on the list of top research universities in the country and you will find the traditional privates, but you would also find state schools that belong to major conferences. People who play sports for these schools get their degrees and become Doctors, Lawyers, Scientists, Broadcasters, Journalists, Business leaders, and Political leaders. The majority will become regular contributors to society in a variety of vocations. A few will become people that no one wants to have their name associated with the school. You will find this at any school in the nation.

It just isn't true, as a general rule, that athletes do not have to go to school or can major in basket weaving. We like to bring up the failures of this school or that school, but there are so many more successes that never get a mention.

[donovan]

OK.....will retract my sarcastic remark about the SEC.....going to go watch reruns of Beverly Hillbillies....or that show with boss hog and the car that flys over every bump in the road....memory fails me...oh yeah University of Dukes Hazzard.

[Spence]

And that proves the point. Buddy Ebsen went to the University of Florida in the SEC. He seem to do pretty well for himself among the socially elite west coast crowd.