Quote: Originally Posted By schroedess26 on 1/06/2010RPI was never been meant as a end all be all to everything. I think people forget this sometimes and consider it a media ranking poll. The RPI basically decides tournament seeding in HD...how else does a 12-16 team get to be a #1 overall seed in the PI?

While it does come in handy for bracketology and a quick look no one claims that RPI rates the whole NCAA from best to worst, it just categorizes the teams. You can assume 1-15, 16-30, 31-45, 46-75, 75-150, 150+ is a good grouping of teams where you can get a feel.

Personally a computer model can never fully get it correct. The ones i have experience with are good enough to get by, but there is a reason they use the coaches and media polls when they talk about rankings and not some hair brained computer scheme. They talk about human polls more than computer polls because that's what the average person can wrap his head around. I don't have a lot of mathematical knowledge as to the hows and whys of many computer pollsters, but I do commend and respect them because they're more objective than human polls and have a better understanding/incorporation of SOS. Human polls are more like air traffic control boards, you win you move up, you lose you move down, they're heavily dependent on preseason rankings.

Just look at what the BCS has done trying to put the top 2 teams every season in the championship game, ruined the chances of the playoff system for the better part of 10 years now. Their postseason format is a joke (1 v. 2 "NC" Game as opposed to an 8 team playoff) but the rankings are solid. The only 2 things I'd change with the rankings is make the humans and computers 50-50 instead of 67-33 and don't throw out any of the computer scores (they currently throw out the highs and lows which I don't like).

Quote: Originally Posted By schroedess26 on 1/06/20101 thing is that you gotta drop the 12-16 team, they had a solid background. In real life, they'd never be a #1 overall seed in the NIT...that would just never fly, regardless of RPI, they had a losing record. Their 48 RPI got them the #1 in HD...RPI is like 35th out of 41 or so in the ranking correlation, but its practically the sole method WIS uses to seed its teams...by that alone I think its time for a change. I showed you an arizona team that made the NT that had the same strength of wins with a couple of the 150 rpi + wins because of the lack of schedule length. Who cares there are much bigger travesties that have occurred bring them up when you find them and you MIGHT have a case.

Next were talking about polls here not HD, its off topic. The reason they throw out some of the computer polls is mellow the results of any one poll. But if the computer system is good enough to be apart of the BCS in the first place (I once talked to Richard Billingsley and he said to even have been considered for the BCS, you had to show them at least 10 years worth of rankings), then why throw out any results? I say include all rather than picking and choosing which don't belong.

You obviously have a lot of faith in a computer to decide the fate of rankings. I would rather have an analyst pick the top 10 teams in the country any day over a computer. How is it a 12-1 Florida team that just lost a game is ranked #2 in a computer poll? They lost and at the time 5 other undefeated teams had won their last game.

You complain about a 12-16 team but you don't find the travesty in having a now 14-0 Boise State team behind an Oregon team that is now 10-3 but 10-2 earlier? Two losses compared to zero! How is that even close no matter your competition they still beat them. Boise State v. Oregon was one game. Boise State played 12 other regular season games, Oregon played 11 other regular season games. Boise State's BPI SOS was 110th in the RS...Oregon's was 4th...in a 120 team league...that's HUUUUUGE. Even with all wins rating higher than all losses, Oregon was able to jump Boise State. You complain because someone uses a strong schedule to their advantage but than defend it when it serves the computer rankings. Makes no sense to me.You're comparing apples to oranges here. Quincy played a great schedule, but a lot of their wins were against the dreck of said schedule...9 of the 12 wins were over RPIs over 100, including 3 over 200 (6 sims) in a 276 team league. And unlike the BSU-ORE scenario, there is more than a 2.5 game W-L difference in reference to Quincy's peers.

Quote: Originally Posted By schroedess26 on 1/06/2010
Also gotta love that you have West Virgina who just lost to Purdue, ahead of Purdue who just beat them and is undefeated! Oh and also it was by 15 points not just a buzzer beater. Its one game out of what 12-15? One game does not a season make...I'll try to get you some BPI SOS data on that, I haven't tabbed it yet. Also, point margin is 3rd in my ranking hierarchy, remember.

Right there is front and center of why computer polls can't be trusted.

Quote: Originally Posted By zhawks on 1/06/2010Who cares about the #1 seed in the PI? Seriously

Quote: Originally posted by zhawks on 1/06/2010Who cares about the #1 seed in the PI? Seriously.

Quote: Originally Posted By coach_billyg on 1/06/2010

Quote: Originally posted by zhawks on 1/06/2010 Who cares about the #1 seed in the PI? Seriously.

this is almost certainly the majority opinion. like i said before, nobody is going to accept a ranking system being broken because the 65th team was seeded wrong. the distinction at that level is so small, its fairly arbitrary if you put the 55 or 75 team there