- My Forums
- Tiger Rant
- LSU Recruiting
- SEC Rant
- Saints Talk
- Pelicans Talk
- More Sports Board
- Fantasy Sports
- Golf Board
- Soccer Board
- O-T Lounge
- Tech Board
- Home/Garden Board
- Outdoor Board
- Health/Fitness Board
- Movie/TV Board
- Book Board
- Music Board
- Political Talk
- Money Talk
- Fark Board
- Gaming Board
- Travel Board
- Food/Drink Board
- Ticket Exchange
- TD Help Board
Customize My Forums- View All Forums
- Show Left Links
- Topic Sort Options
- Trending Topics
- Recent Topics
- Active Topics
Started By
Message
re: BCS: If you had to pick either polls or computers...?
Posted on 8/21/10 at 4:41 pm to baytiger
Posted on 8/21/10 at 4:41 pm to baytiger
quote:+1
quote:
perhaps one representative from each BCS conference, should be able to perform a sanity check to affirm the results or tweak by consensus, two thirds vote, etc.
that undermines the entire purpose of unbiased polls.
The math in a computer formula IS a sanity check. If you crunch the numbers and do away with votes, any team who ever gets "screwed" will always know exactly why because it can be numerically explained.
0
Posted on 8/21/10 at 5:01 pm to ULL Cool J
quote:Fair enough. I definitely respect your opinion...but I definitely want to make it hard for you to defend it.
But the way the Sooners lost in the Big 12 championship game should have taken them out of the BCS championship, imo.
So here goes....
Your case here is based on a single game (out of xiii at that point) that Oklahoma lost to a barely-ranked team 35-7.
Record:
Oklahoma 12-1
LSU 12-1
USC 11-1
Opponents' Record:
Oklahoma 88-80
LSU 76-77
USC 75-77
Points scored divided by Points Allowed:
Oklahoma 3.04
LSU 3.24
USC 2.25
Now, again, I definitely respect your opinion and know that you respect mine...but I would submit that our opinions should not matter when so much mathematical evidence suggests that Oklahoma had clearly had the best season of the three (and LSU second).
Oklahoma's season wasn't even subjectively better--it was mathematically better.
Southern California played a strong schedule, and they beat their opponents with a high frequency and by a wide margin.
Oklahoma played a stronger schedule, and they beat their opponents with a higher frequency and by a wider margin.
Basically, what I'm hearing from you and those who agree with you on this issue is that a game played on December 5 is more important than is a game played on September 27. To that, I say that all regular season games count equally.
Now, if you want to look at all that on-field evidence and place so much more importance on the game played in Kansas City on December 5, 2001, than on all the other games played, that is your choice and your prerogative, and I can respect it. (I guess I just don't get it.)
I...just...can't...let...this...go...ever...help...me...*cough*
This post was edited on 8/21/10 at 5:04 pm
Posted on 8/21/10 at 5:29 pm to xiv
LOL! I get your point. But, again imo, the FBS division has too disparate a group of programs to go to pure computer selection.
Suppose one year the computer selected undefeated WAC champion Louisiana Tech to play undefeated MAC champion Toledo, when, to the real world, neither deserved to be in the top ten. Further, to make this scenario work, the real bigtime programs don't have one or two teams clearly ahead of the pack (like Texas and USC in 2005) but are bunched up, such as was the case in 2007. Would you be okay with going through the charade of treating Latech vs. Toledo as the national championship game, knowing there were 8 to 10 two-loss teams that would handily beat either?
Oh and on the OU argument, they didn't win their conference.
But on something we both agree: Latech is afraid of ULM.
Suppose one year the computer selected undefeated WAC champion Louisiana Tech to play undefeated MAC champion Toledo, when, to the real world, neither deserved to be in the top ten. Further, to make this scenario work, the real bigtime programs don't have one or two teams clearly ahead of the pack (like Texas and USC in 2005) but are bunched up, such as was the case in 2007. Would you be okay with going through the charade of treating Latech vs. Toledo as the national championship game, knowing there were 8 to 10 two-loss teams that would handily beat either?
Oh and on the OU argument, they didn't win their conference.
But on something we both agree: Latech is afraid of ULM.
This post was edited on 8/21/10 at 5:31 pm
Posted on 8/21/10 at 5:44 pm to xiv
quote:
Opponents' Record:
Oklahoma 88-80
LSU 76-77
USC 75-77
The problem with the whole strength of schedule piece of the formula is that it assumes everything is equal.
In your formula, if team A beats an SEC school that finishes 11-1, and team B beats a Sun Belt school that finishes 11-1, there is no difference.
In addition, once you get past a certain point, does it really matter?...ie when you look at the weaker portion of a team's schedule. For instance, should team A's victory over a 1-11 team really vault them over team B who beat a 2-10 squad?
Posted on 8/21/10 at 9:47 pm to ULL Cool J
quote:
LOL! I get your point. But, again imo, the FBS division has too disparate a group of programs to go to pure computer selection.
Suppose one year the computer selected undefeated WAC champion Louisiana Tech to play undefeated MAC champion Toledo, when, to the real world, neither deserved to be in the top ten. Further, to make this scenario work, the real bigtime programs don't have one or two teams clearly ahead of the pack (like Texas and USC in 2005) but are bunched up, such as was the case in 2007. Would you be okay with going through the charade of treating Latech vs. Toledo as the national championship game, knowing there were 8 to 10 two-loss teams that would handily beat either?
Oh and on the OU argument, they didn't win their conference.
But on something we both agree: Latech is afraid of ULM.
You don't know anything. You assume they would beat them. Assumptions are the flaw of FBS college football and why its not a competitive sport.
Posted on 8/21/10 at 9:51 pm to xiv
computers
Posted on 8/21/10 at 11:23 pm to NorthshoreTiger76
The only way I would accept computers is if they only looked at your top 6(or something similiar) opponents to throw away the gimme wins.
I just want to ensure schools with difficult schedules are heavily rewarded. I'd rather a team play 8 top 25 teams have a loss than a team that only played 1 or 2 top 25 teams and go undefeated.
I just want to ensure schools with difficult schedules are heavily rewarded. I'd rather a team play 8 top 25 teams have a loss than a team that only played 1 or 2 top 25 teams and go undefeated.
Posted on 8/23/10 at 9:24 am to Obi-Wan Tiger
quote:My SOS formula is the same as the BCS's old formula. Your opponents winning % times 2, plus your opponents' opponents winning % times 1, divided by 3. OU and LSU had tougher schedules than USC that year based on that formula.
The problem with the whole strength of schedule piece of the formula is that it assumes everything is equal.
In your formula, if team A beats an SEC school that finishes 11-1, and team B beats a Sun Belt school that finishes 11-1, there is no difference.
quote:It matters a lot more than an opinion does.
In addition, once you get past a certain point, does it really matter?
Posted on 8/23/10 at 9:28 am to ULL Cool J
quote:As much information as you've given me in this hypothetical, it still isn't nearly enough for me to give an honest opinion with conviction. Whom did Tech and Toledo play? Whom did their opponents play?
Suppose one year the computer selected undefeated WAC champion Louisiana Tech to play undefeated MAC champion Toledo, when, to the real world, neither deserved to be in the top ten. Further, to make this scenario work, the real bigtime programs don't have one or two teams clearly ahead of the pack (like Texas and USC in 2005) but are bunched up, such as was the case in 2007. Would you be okay with going through the charade of treating Latech vs. Toledo as the national championship game, knowing there were 8 to 10 two-loss teams that would handily beat either?
I will say that, based on what I know about the math used in the six BCS computers (I know all the formulas except for Sagarin--and frankly, I'd rather not know that one
), I trust them enough that if they put Tech and Toledo at #1 and #2, I'd be ok with them in the title game.quote:USC didn't win their state (Fresno State did).
Oh and on the OU argument, they didn't win their conference.
quote:+1
But on something we both agree: Latech is afraid of ULM.
Posted on 8/23/10 at 9:33 am to xiv
quote:
Whom did Tech and Toledo play? Whom did their opponents play?
the wac and the mac...
quote:
I trust them enough that if they put Tech and Toledo at #1 and #2, I'd be ok with them in the title game.
that's why we have human judgement...
Posted on 8/23/10 at 9:40 am to C
quote:All six computers have either a quality win quotient (victories over higher-rated teams glean brownie points) or a function that puts more weight to top wins. I think that is sufficient. There is a hole in your specific suggestion (I'll use extremes in this hypothetical to make the point clear):
The only way I would accept computers is if they only looked at your top 6(or something similiar) opponents to throw away the gimme wins.
LSU (13-0, 8-0)
McNeese State (FCS)
Louisiana-Monroe (2-10)
Louisiana Tech (2-10)
Miami-Ohio (2-10)
Vanderbilt (2-10)
Kentucky (2-10)
Mississippi State (2-10)
Florida (9-3)
Ole Miss (9-3)
Alabama (9-3)
Auburn (9-3)
Arkansas (9-3)
Florida (SECCG) (9-3)
USC (13-0, 9-0)
Nevada (5-7)
Notre Dame (5-7)
San Jose State (5-7)
Washington (5-7)
Oregon (8-4)
Oregon State (8-4)
Colorado (8-4)
California (8-4)
Stanford (8-4)
UCLA (8-4)
Arizona (8-4)
Arizona State (8-4)
Utah (P12CG) (8-4)
Your suggestion would give LSU (1 FCS team, 6 2-10 teams, 6 9-3 teams) a higher SOS, but I'd say that USC's schedule (4 5-7 teams, 9 8-4 teams) is definitely tougher.
This post was edited on 8/23/10 at 9:44 am
Posted on 8/23/10 at 9:42 am to chalmetteowl
quote:And THAT'S why we need computers. "Louisiana Tech" and "Toledo" don't sound right to you, and there's virtually nothing that either team could possibly do to eliminate that bias in your head...and that goes for us all; we all have that bias to some extent. But math doesn't. Math is 100% consistent, and numbers are fair to everyone. Allowing people to vote on these things does nothing but throw bullshite in there.
quote:
I trust them enough that if they put Tech and Toledo at #1 and #2, I'd be ok with them in the title game.
that's why we have human judgement...
Posted on 8/23/10 at 9:52 am to C
quote:
I just want to ensure schools with difficult schedules are heavily rewarded. I'd rather a team play 8 top 25 teams have a loss than a team that only played 1 or 2 top 25 teams and go undefeated.
I kige this.
That's why the BCS used to have a quality win bonus. It's better to beat #1,2,3,4,5,6,115,116,117,118,119, and 120 than it is to beat #55,56,57,58,59,60,61,62,63,64,65, and 66.
But they ditched that because their precious little media darling got punished for a weak schedule.
Posted on 8/23/10 at 10:21 am to Bestbank Tiger
quote:All the computers have SOS factored in; the BCS's independent SOS factor arguably gave too much weight to it. SOS wasn't eliminated--it's just been left to the computers. I'm fine with that.
But they ditched that because their precious little media darling got punished TWICE for a weak schedule.
Posted on 8/23/10 at 10:42 am to Gmorgan4982
quote:
The "computers" are opinions, though. It's math formulas. There is not even one agreed-upon "best" math formula. One guy thinks his math formula is more accurate than another guy.
agree
in defense of the computer polls I would say this tho - the pollsters (AP, coaches, w/e) who put out a weekly opinion poll will find it inevitably biased by their prior week's opinions... and that is a pretty big problem (the pollsters become more worked up about defending prior opinions than trying to form objective results based opinions)
Posted on 8/23/10 at 10:53 am to xiv
quote:
Your suggestion would give LSU (1 FCS team, 6 2-10 teams, 6 9-3 teams) a higher SOS, but I'd say that USC's schedule (4 5-7 teams, 9 8-4 teams) is definitely tougher.
Let's take the words LSU and USC out of this as much as we can... so we argue about the theory instead of LSU/USC or their conferences
For different types of teams different schedules pose different challenges.
For a bad BCS team (say Vanderbilt) the second schedule would be much harder because there are very few teams there of equal talent. But for a higher quality team (say OSU, Bama, Texas) the first schedule would be more difficult - because the more highly talented teams you face... the more teams with a puncher's chance you have to get past
Posted on 8/28/10 at 11:10 pm to shel311
quote:
So, NO teams would ever have any incentive to play tough OOC games.
Strength of schedule is listed as a tiebreaker. That, and playing tough OOC opponents gives teams a chance to knock someone out of contention and take their place. The incentive to play tough OOC games would be about the same as it is now.
Posted on 8/28/10 at 11:14 pm to xiv
quote:
A conference season is only 2/3 or 3/4 of the entire season. Teams should be judged on their whole seasons without any outsider giving any special designation to a fraction of their season.
A conference championship is only a prerequisite, not an automatic bid. You can't have 6 teams play in 1 championship game. It only prevents those who could not even win their own conference from being able to hold a national title for the same year that they do not have a conference title.
Page 3 of 3
Back to top
Follow TigerDroppings for LSU Football News