If it was up to the BCS computers only....

[Derek]

I am in favor of the Computers having more weight than the human polls. I would be happy with 2/3rds computer.

Of course, it is humans that are setting up the computers, as to how the results should be obtained.

The formulas as set. They can't be tinkered with to get the results they want.

[Spence]

I am in favor of the Computers having more weight than the human polls. I would be happy with 2/3rds computer.

Of course, it is humans that are setting up the computers, as to how the results should be obtained.

The formulas as set. They can't be tinkered with to get the results they want.

All the BCS computer rankings have been changed to satisfy changes the BCS wanted or they would not be used. They went all in. It is the reason CCR didn't take part.

[silverfox]

I rest my case. I have testified as a forensic witness in the past concerning the reliability of computers and can tell you, without a doubt, computers can always be suspect due to variables that are used and the bias implemented by the creators, the programmers. I think most participants in this forum agree.

[billybud]

The difference is that although computer algorithmn may have a "set" bias, it will be set throughout the season and will be applied equally across the spectrum.

Humans have a bias that is not applied equally across the spectrum and can be pretty whimsical. I have been reviewing the votes of AP voters and the bias and whimsy are easy to see.

[silverfox]

The difference is that although computer algorithmn may have a "set" bias, it will be set throughout the season and will be applied equally across the spectrum.

Humans have a bias that is not applied equally across the spectrum and can be pretty whimsical. I have been reviewing the votes of AP voters and the bias and whimsy are easy to see.

Actually I agree with you...

[donovan]

The difference is that although computer algorithmn may have a "set" bias, it will be set throughout the season and will be applied equally across the spectrum.

Humans have a bias that is not applied equally across the spectrum and can be pretty whimsical. I have been reviewing the votes of AP voters and the bias and whimsy are easy to see.

Whimsy was one of my best subjects in school, right after lunch and recess.

[silverfox]

The difference is that although computer algorithmn may have a "set" bias, it will be set throughout the season and will be applied equally across the spectrum.

Humans have a bias that is not applied equally across the spectrum and can be pretty whimsical. I have been reviewing the votes of AP voters and the bias and whimsy are easy to see.

Whimsy was one of my best subjects in school, right after lunch and recess.

LOL

[RazorHawk]

I am in favor of the Computers having more weight than the human polls. I would be happy with 2/3rds computer.

Of course, it is humans that are setting up the computers, as to how the results should be obtained.

The formulas as set. They can't be tinkered with to get the results they want.

Who set those formulas? I would suspect humans.

[billybud]

The formulas at least have internal consistancy.

The formulas are cut and dried..how they treat the "value" of a win, how they rate power. They don't see that it was a monsoon and the passing team was at a disadvantage, they don't see that a back up QB had to fill in for a game, they don't see how a lucky bounce or a bad call allowed the other team to win.

Us humans are much better at qualifying a win or loss...but we also may allow our bias to play when doing that qualifying. I think that my boy is handsomer than he probably is. I see him with a father's warm eye.

[Spence]

The formulas at least have internal consistancy.

The formulas are cut and dried..how they treat the "value" of a win, how they rate power. They don't see that it was a monsoon and the passing team was at a disadvantage, they don't see that a back up QB had to fill in for a game, they don't see how a lucky bounce or a bad call allowed the other team to win.

Us humans are much better at qualifying a win or loss...but we also may allow our bias to play when doing that qualifying. I think that my boy is handsomer than he probably is. I see him with a father's warm eye.

You can't take bias out of human polling. As long as people can put aside their negative bias - the poll will be fine. Positive bias is weeded out from voter to voter. Negative bias, though, has a huge impact.

[Derek]

I am in favor of the Computers having more weight than the human polls. I would be happy with 2/3rds computer.

Of course, it is humans that are setting up the computers, as to how the results should be obtained.

The formulas as set. They can't be tinkered with to get the results they want.

All the BCS computer rankings have been changed to satisfy changes the BCS wanted or they would not be used. They went all in. It is the reason CCR didn't take part.

Yes, but never in the middle of the year.

I understand the point, but I still believe that they are more honest than a sportswriter or coach. See 1989 and FSU.

[Spence]

The computers can be biased to a style or play or a certain criteria that favors one brand of football after another, but they can't be biased against teams. All humans have bias, but as long as it isn't a negative bias or they try to manipulate the total, it isn't going to affect the overall results. Provided you have a large enough sample.

[Derek]

The computers can be biased to a style or play or a certain criteria that favors one brand of football after another, but they can't be biased against teams. All humans have bias, but as long as it isn't a negative bias or they try to manipulate the total, it isn't going to affect the overall results. Provided you have a large enough sample.

Yeah, and they have these stories of coaches voting other teams way down, just like the Heisman. It's crooked.

[WoVeU]

Certain Engineers are supposed to give me a Sequence of Operations that I then turn into a program. This can often include the way to formulate something (the formula, perhaps approach, or the frame for an algorithm.) One of the fun parts is when they watch the unit run and the program...and say man "this or that works well." And I say, "no, I threw that out..." and "I programmed it to give you the operation and result you wanted instead of the method you asked for." *They don't give me sequences anymore!

So what I don't get...if the BCS requires this or that to be in the formula..then why isn't there a BCS Program rather than programs??? If there are multiple programs, then I'd have to conclude there is differentiation between them or time and effort is being wasted (and money too I would assume) OR there are other factors not stipulated by the BCS?

[WoVeU]

The computers can be biased to a style or play or a certain criteria that favors one brand of football after another, but they can't be biased against teams. All humans have bias, but as long as it isn't a negative bias or they try to manipulate the total, it isn't going to affect the overall results. Provided you have a large enough sample.

Yeah, and they have these stories of coaches voting other teams way down, just like the Heisman. It's crooked.

To be sure I voted consistently for schools like Notre Dame and Michigan...and USC...and Miami...and Chokelahoma, which I do not care for, I had to use a formula for them. I average their AP and Coaches rankings then I raise that to the zero power, divide by 1, then multiply by 115! I have to retain my integrity.