BCS poll

#1
grimvol

grimvol

Just a dude
Joined
Oct 15, 2008
Messages
1,008
Likes
17
#1
Brad Edwards (ESPN's BCS GERU) just said on ESPN radio he has seen the BCS computers..and if the BCS came out today,,,Boise State is #1 Oregon is #2 TCU is #3
Bama is 10 and Ohio State is 11.
 
#3
#3
The BCS standings can't be correctly calculated until a certain date. Or at least the computer portion of them can't be.
 
#4
#4
i was about to say that's probably why they aren't released until midseason

just for kicks, there should be a segment on gameday or something that keeps track of the bcs rankings(as messed up as they would be) early on in the season. you'd probably get some funny reactions.
 
Last edited:
#5
#5
I think the BCS comes out next week. Ohio State will be #1 if they win.
 
#6
#6
I'm just reporting what the dude said on ESPN..he did say something about the computers don't look at the games ahead...only the past games.
and this is ONLY the BCS computers..no other polls were involved.
 
#7
#7
duckman398686 said:
how would ohio st be 11?? they're undefeated and #2 right now in the polls.
Click to expand...

well really Ohio State is #1 in the polls but they have played only 1 top 25 team....like I said..the dude says the BCS coumputers only look at the games played so far. and Ohio State hasn't played anybody really.
 
Last edited:
#8
#8
The BCS computers (and just them) are very different from a human one. There's a reason that 2/3rd's of the poll is the Coaches' and media poll. And since they took out SOS weighting from the BCS formula, it doesn't matter as much if you haven't played anyone.
 
#10
#10
ESPN formulated the BCS standings without one of the six computers, in addition to the Harris and USA Today polls. So it's an estimation, but a close one at this point. IIRC it looked something like this:

1. Boise State
2. Oregon
3. Texas Christian
4. Oklahoma
5. Ohio State
6. Louisiana State
7. Nebraska
8. Auburn
9. Michigan State
10. Alabama
 
#11
#11
Also, their "BCS expert" said the aggregate computer ranking had Ohio State at 10th and he made a near-guarantee that the computers will end up costing Boise State a shot at the title game due to SOS by seasons end. That is, of course, unless something crazy happens like all the AQ conferences producing two loss teams.
 
#12
#12
grimvol said:
well really Ohio State is #1 in the polls but they have played only 1 top 25 team....like I said..the dude says the BCS coumputers only look at the games played so far. and Ohio State hasn't played anybody really.
Click to expand...

Now that Miami has lost, AGAIN, they really haven't played anyone. Miami will be overhyped again next year to be sure though.....
 
#13
#13
duckman398686 said:
SOS should still be in the BCS IMO.
Click to expand...

i think so too rather than just that BS "some of the computers use it" percentage it has been lowered to

but that cost USC, so clearly anything in the system that impairs one of the big selling 5 or 6 can't be kept (rolling eyes)
 
#14
#14
IPorange said:
The BCS computers (and just them) are very different from a human one. There's a reason that 2/3rd's of the poll is the Coaches' and media poll. And since they took out SOS weighting from the BCS formula, it doesn't matter as much if you haven't played anyone.
Click to expand...

they still look at SOS..just not style points.
Boise could beat Nevada by 10 or 100..the computers look at it as a win...now the coaches poll will look at it another way, but that is human nature.
 
#15
#15
Nebraska is better than 7th and will continue to go through the Big12 like a buzz saw.
 
#19
#19
grimvol said:
they still look at SOS..just not style points.
Boise could beat Nevada by 10 or 100..the computers look at it as a win...now the coaches poll will look at it another way, but that is human nature.
Click to expand...

the computers do, but it doesn't carry as much weight as it used to pre whenever LSU played Oklahoma. but you're right, the "quality win" component was removed

of course the human polls are fickle as hell
 
#20
#20
starting in 98:

The BCS formula calculated the top 25 teams in poll format. After combining a number of factors, a final point total was created and the teams that received the 25 lowest scores were ranked in descending order. The factors were:

Poll average: Both the AP and ESPN-USA Today coaches polls were averaged to make a number which is the poll average.

Computer average: An average of the rankings of a team in three different computer polls were gathered (Jeff Sagarin/USA Today, Anderson-Hester/Seattle Times, and New York Times), with a 50% adjusted maximum deviation factor. (For instance, if the computers had ranked a team third, fifth, and twelfth, the poll which ranked the team twelfth would be adjusted to rank the team sixth.)

Strength of Schedule: This was the team's NCAA rank in strength of schedule divided by 25. A team's strength of schedule was calculated by win/loss record of opponents (66.6%) and cumulative win/loss record of team's opponents' opponents (33.3%). The team who played the toughest schedule was given .04 points, second toughest .08 points, and so on.

Losses: One point was added for every loss the team has suffered during the season. All games are counted, including Kickoff Classics and conference title games.[5]
Click to expand...

Small changes from 98 on to 03:

Before the 1999-2000 season, five more computer rankings were added to the system: Richard Billingsley, Richard Dunkel, Kenneth Massey, Herman Matthews/Scripps Howard, and David Rothman. The lowest ranking was dropped and the remainder averaged.

Beginning in 2001, The Peter Wolfe and Wes Colley/Atlanta Journal-Constitution computer rankings were used in place of the NYT and Dunkel rankings. The change was made because the BCS wanted computer rankings that did not depend heavily on margin of victory.[6] The highest and lowest rankings were discarded, and the remainder averaged. A team's poll average, computer average, strength of schedule points, and losses were added to create a subtotal.

Also in 2001, a quality win component was added. If a team beat a team which was in the top 15 in the BCS standings, a range of 1.5 to .1 points was subtracted from their total. Beating the #1 ranked team resulted in a subtraction of 1.5 point, beating the #2 team resulted in a deduction of 1.4 points, and so on. Beating the #15 ranked team would have resulted in a deduction of .1 points. A team would only be awarded for a quality win once if it beat a Top 10 team more than once (such as in the regular season and a conference championship game), and quality wins were determined using a team's current subtotal, not the ranking when the game was played. The subtotal ranks were used to determine quality win deductions to create a team's final score.

The BCS continued to purge ranking systems which included margin of victory, causing the removal of the Matthews and Rothman ratings before the 2002 season. Sagarin provided a BCS-specific formula that did not include margin of victory, and the New York Times index returned in a form without margin of victory considerations. In addition, a new computer ranking, the Wesley Colley Matrix, was added.[7] The lowest ranking was dropped and the remaining six averaged. Also in 2002, the quality win component was modified such that the deduction for beating the #1 team in the BCS would be 1.0, declining by 0.1 increments until beating the 10th ranked team at 0.1. Teams on probation were not included in the BCS standings, but quality win points were given to teams who beat teams on probation as if they were ranked accordingly in the BCS.
Click to expand...

since 2004 on:

In response to the controversy created by the voters in the AP poll naming USC as the No. 1 ranked team at the end of the year,[8] the formula was rewritten. Supporters of USC and the media in general criticized the fact that human polls were not weighted more heavily than computer rankings and this criticism led to the new algorithm.

AP Poll: A team's AP Poll number is the percentage of the possible points it could receive in the poll. As an example, in the final regular-season poll of 2003, LSU received a total of 1,580 out of a possible 1,625 points from the voters, giving them an AP Poll percentage of 97.2.

Coaches' Poll: This is calculated in the same manner as the AP Poll number. For LSU, their final regular-season number in this poll would have been 99.4 (1,516 out of 1,525 possible points).

Computer Average: The BCS used six ranking systems: Jeff Sagarin, Anderson/Hester, Richard Billingsley, Wes Colley, Kenneth Massey, and Peter Wolfe. In the calculation, the highest and lowest ranking for each team are dropped. Then, it will give a team 25 points for a Number 1 ranking in an individual system, 24 points for Number 2, and so on down to 1 point. Each team's set of numbers is then added, conveniently making the number compatible with the percentages from the two polls. To address concerns about loss of the schedule strength factor, the description of the computer rankings explicitly included schedule strength as a consideration.


For USC, dropping their highest and lowest computer rankings would have left them with four third-place finishes, worth 23 points each for a total of 92, while LSU would have had four second-place finishes for a total of 96. The BCS averaged the three numbers obtained above, divided the result by 100, and converted it to a decimal fraction. This system placed twice as much emphasis on human polls than computer rankings (since there were two human polls and an average of six computer polls), and made it highly unlikely that the top team in both human polls would be denied a place in the title game, as it happened in 2003–04.
Click to expand...

and of course, a short while later the AP poll dropped out and was replaced with the Harris Poll

Bowl Championship Series - Wikipedia, the free encyclopedia
 
Last edited:
#21
#21
The ideal way to do the BCS is to simply remove the human polls entirely. Humans are prone to corruption, subconscious biases, and intentional biases. If there's one thing I've learned from tinkering with computers as it relates to any type of stat analysis, it's that it's dang near impossible to intentionally alter formulae and algorithms to achieve a particular desired outcome without really screwing up a bunch of other stuff along the way.

Also, removing the strength of schedule component to pacify the crybabies at USC was ridiculous in every way.
 
New posts
You must log in or register to reply here.

VN Store



Back
Top