October 29, 201411 yr But that represents a huge game on the schedule for each of those five ACC schools. One that then can't be filled with another out of conference opponent while still playing eight league games. Makes no sense. Because those 5 games in the ACC every year make ND/ACC almost as insular as you're whining about the SEC being. If the ACC and ND are both terrible, but only play each other, how do you know?
October 29, 201411 yr Getslow is laying the smackdown in this thread. Who else wants some of Getslow????? Go ahead!!! Bring it!!!!!!!!!!!!! GIFSoup
October 29, 201411 yr My point is that NONE of them are tested but the SEC benefits from it at a higher rate than most other leagues. The Big Ten also has an historical bias that works in its favor as well. Michigan's program has been collapsing like a poorly-mixed cake for several years now but they still got votes in the preseason polls. What but historical bias could cause that sort of nonsense? It is really tough to remove biases from the equation. We have no clue exactly how much better one league is than another because -- for the most part -- we only have bowl wins to use as data points (points which are extremely unreliable and too few to really be valuable). Removing polling and rankings from the equation as much as possible will serve to make crowning a champion easier and (probably) more reliable. So, remove polling and rankings, and use something like a computer? Like, say.. Sagarin? Sagarin: Alabama Auburn Miss MS St UGA Oklahoma Baylor TCU LSU Oregon
October 29, 201411 yr Because those 5 games in the ACC every year make ND/ACC almost as insular as you're whining about the SEC being. If the ACC and ND are both terrible, but only play each other, how do you know? We don't. We don't have any idea.
October 29, 201411 yr So, remove polling and rankings, and use something like a computer? Like, say.. Sagarin? Sagarin: Alabama Auburn Miss MS St UGA Oklahoma Baylor TCU LSU Oregon Don't particularly trust Sagarin either as, for good reason, he won't tell anyone his methodology. I get it, neither would I. We're getting somewhere with the Sagarin ratings but again, I want to know more and I certainly don't want his rating system with it's unknown weights and algorithms messing with who gets crowned champion. I've read that Sagarin based his own system on the Elo ratings for international chess players. I don't know much about that either, but if it's roughly as accurate as the system for FIFA World Rankings of soccer teams, then it's hardly worth the paper it's printed on.
October 29, 201411 yr We don't. We don't have any idea. Great, so we agree that ACC vs ND games are just as insular as any other in conference game, and therefore don't get counted on the "quality OOC games" list.
October 29, 201411 yr Great, so we agree that ACC vs ND games are just as insular as any other in conference game, and therefore don't get counted on the "quality OOC games" list. We do not. They present a scheduling obstacle that exists outside the eight-game conference slate. That's an out-of-conference game. Plus, Notre Dame plays more non-ACC teams than they do ACC teams, so insularity isn't really the question there. Notre Dame is actually a really interesting case every year because their LACK of insularity means we can start evaluating them against several leagues. They play more than one team from the ACC, Big Ten, and Pac 12 this season. They are, in fact, among the least insular teams in all of college football.
October 29, 201411 yr Don't particularly trust Sagarin either So pick any other computer ranking. I picked Sagarin because it is well known. masseyratings.com Miss ST Miss Bama Auburn FSU UGA Oregon TCU LSU Mich St colleyrankings.com Miss St Auburn Miss FSU Bama Oregon ND Nebraska LSU UGA
October 29, 201411 yr Bottom line, all these computer rankings have the SEC teams ranked HIGHER than the committee. Committee - 3 of top 4, 4 of top 10 Sagarin - 4 of top 4, 6 of top 10 Massey - 4 of top 4, 6 of top 10 Colley - 3 of top 4, 6 of top 10 To me that says the committee is already paranoid about being berated for "SEC bias" so LSU and UGA are being ranked artificially low.
October 29, 201411 yr Bottom line, all these computer rankings have the SEC teams ranked HIGHER than the committee. Committee - 3 of top 4, 4 of top 10 Sagarin - 4 of top 4, 6 of top 10 Massey - 4 of top 4, 6 of top 10 Colley - 3 of top 4, 6 of top 10 To me that says the committee is already paranoid about being berated for "SEC bias" so LSU and UGA are being ranked artificially low. I think there's a possibility that's true, but I also think that there are biases inherent in a computer system as well. They can't help it. One rating may favor statistic A at the expense of another statistic B while another may do it backwards. Computer rankings, poll voters, committees -- I want all of them as far away as is effectively possible from determining a champion in a sport I like. The relative strength of one conference is extremely difficult to establish in a sport where every team has (at most) 13 times in which to establish data. Too much variance and not enough interaction between all the members.
October 29, 201411 yr I think there's a possibility that's true, but I also think that there are biases inherent in a computer system as well. They can't help it. One rating may favor statistic A at the expense of another statistic B while another may do it backwards. Computer rankings, poll voters, committees -- I want all of them as far away as is effectively possible from determining a champion in a sport I like. The relative strength of one conference is extremely difficult to establish in a sport where every team has (at most) 13 times in which to establish data. Too much variance and not enough interaction between all the members. So you're entire argument is that ANY playoff that doesn't include EVERY team is faulty?
October 29, 201411 yr So you're entire argument is that ANY playoff that doesn't include EVERY team is faulty? Well, to be completely honest, any system that doesn't determine a champion by having every single team play against each other is going to be flawed in one way or another. My wish is to eliminate as much of the flukiness as possible from the system. In the case of college football, I think that means expanding the playoff to at least eight teams. I'd personally like it to be more so that we can be sure we're not placing too much emphasis on one game from the regular season possibly defining an entire season. When we eliminate one loss teams out of hand from title contention, we're allowing WAY too much for random chance to interfere with crowning a winner. Notre Dame, for example, was unbeaten in 2012 and so was allowed to play for a title at the expense of teams like Oregon. The Ducks played two teams all season that were even within 20 points of them, they happened to lose one... to Stanford... by three... in overtime. Notre Dame made the title game because the system favored being unbeaten. The Irish had nearly lost a couple times and had at least five games decided by a touchdown or less... some of which were not to to quality teams. But because they won, and the computers and voters like wins, they played in the title game despite the fact that other teams were likely better than them but didn't catch the same breaks. I want those breaks to not affect the crowning of a champion to the greatest degree possible.
October 29, 201411 yr What's in it for ESPN to have SEC teams dominate the final 4? Are there a bunch of fans in Los Angeles and New York and Chicago that will want to watch MissSt or Ole Miss play in the championship game instead of an Oregon or Michigan State or Notre Dame?
October 29, 201411 yr Well, to be completely honest, any system that doesn't determine a champion by having every single team play against each other is going to be flawed in one way or another. My wish is to eliminate as much of the flukiness as possible from the system. In the case of college football, I think that means expanding the playoff to at least eight teams. I'd personally like it to be more so that we can be sure we're not placing too much emphasis on one game from the regular season possibly defining an entire season. When we eliminate one loss teams out of hand from title contention, we're allowing WAY too much for random chance to interfere with crowning a winner. The bigger you make the playoff the more you invite that random chance to happen in the playoff instead of the regular season. So you want to go the basketball route? 68 teams? You think UK was the 2nd best team in college basketball last year? You think UConn was the best?
October 29, 201411 yr The bigger you make the playoff the more you invite that random chance to happen in the playoff instead of the regular season. So you want to go the basketball route? 68 teams? You think UK was the 2nd best team in college basketball last year? You think UConn was the best? Likely no, but we've decided that this is the system for us, so given a massive league (NCAA Division I), we have to do the best with what we have. There are 128 schools playing in the FBS this year. It is extremely difficult to pin down the best of them while eliminating historical bias and recency bias, accounting for injuries, accounting for the role of chance, and trying to figure out who's best between three times that might have four degrees of separation between each of them. Like I said, no way to determine a champion accurately without everyone playing each other. This is what we have and a playoff in which what we imagine are the best teams playing against one another on the field is the best way to determine that, not a computer ranking or coaches poll or media survey.
Archived
This topic is now archived and is closed to further replies.