It seems to me that the combination of BYU being unusually good this year and USU playing georgetown accounts a great deal for the expected RPI of 20. I don't mention the gawdy number of wins because as we know, 23-28 wins/per year is has become pretty much the norm. Even with a very bad WAC this year, our RPI is expected to be much higher than the "usual" 30s where we have seemed to settle in years past. I expect the RPI and difficulty of play in the new WAC to do nothing but fall in years to come. With a top 20 RPI and 25+ wins, at-large seems almost a given. Could it be as simple as scheduling a couple top 20 teams each year and a cakewalk through the new-WAC conference to guarantee an at-large?

A team's overall RPI ranking is at best a secondary consideration when they select at-large teams. A team's record against top-25/top-50/top-100 opponents is far more important. If we manage to occasionally win a few OOC games against tough opponents, then sure, this strategy is great. But I really wouldn't want to have an RPI of 20 and 0 top-100 wins on Selection Sunday. An RPI of 21 wasn't good enough for Missouri State in 2006, and they had far more quality wins than this year's USU team is projected to have.

Stanford... is this true? I just looked up the Wikipedia entry that I read about snubs on and they have updated it to say that Arkansas had an RPI of 19 and was left out of the tournament in 2002.

The Missouri State snub was bad, but an RPI of 19 should get you in IMO.

http://en.wikipedia.org/wiki/NCAA_baske ... on_process

I don't know the make up of past committee's. This year only 4 of the 10 members are from the so called BCS conferences. That in combination with Scott Barnes on the committee has to help us out tremendously. The 5 other non BCS representatives can logically present testimony of the scheduling struggles they face. I'm hoping this plays a big factor...not only if we need an at-large but in our seeding as well.

aggiesdotcom wrote:I don't know the make up of past committee's. This year only 4 of the 10 members are from the so called BCS conferences. That in combination with Scott Barnes on the committee has to help us out tremendously. The 5 other non BCS representatives can logically present testimony of the scheduling struggles they face. I'm hoping this plays a big factor...not only if we need an at-large but in our seeding as well.

Good point. Maybe they'll start giving more emphasis to the NCAA's stated factors such as RPI, various computer rankings, wins, road/neutral record, conference record, polls , etc., and won't be as hung up on "wins vs. top 50" opponents, which clearly favors teams from power conferences who have 10 times as many opportunities to collect such wins, including multiple opportunities to collect such wins on their home floors.

http://www.ncaa.com/sites/default/files ... 0-5-10.pdf

Rabidchild wrote:

aggiesdotcom wrote:I don't know the make up of past committee's. This year only 4 of the 10 members are from the so called BCS conferences. That in combination with Scott Barnes on the committee has to help us out tremendously. The 5 other non BCS representatives can logically present testimony of the scheduling struggles they face. I'm hoping this plays a big factor...not only if we need an at-large but in our seeding as well.

Good point. Maybe they'll start giving more emphasis to the NCAA's stated factors such as RPI, various computer rankings, wins, road/neutral record, conference record, polls , etc., and won't be as hung up on "wins vs. top 50" opponents, which clearly favors teams from power conferences who have 10 times as many opportunity to collect such wins, including multiple opportunities to collect such wins on their home floors.

What is your alternative, then? You really want college basketball to rely on polls like the BCS? Remember how Utah got a BCS bid over Boise a few years earlier mainly because Utah started the season ranked higher than Boise did? Should USU be denied an at-large bid because we weren't ranked high enough in the preseason polls? And if you think that USU has a hard time getting decent OOC games now, I can only imagine how it would be if polls determined who received tournament bids the BCS schools would only play OOC games against cupcakes, since avoiding losses would be far more important than increasing one's SOS.

I do think the current system favors schools in bigger conferences, but I don't think it's as bad as some people think. Generally speaking, a school that has a few wins at home against teams ranked 51-100 (and lots of road losses to these teams, plus some home losses against higher ranked teams) doesn't do well on Selection Sunday. To get an at-large bid, you either need to beat a few very good (ie top-25 RPI) teams, or have a favorable record against teams in the top 100. (I think the Dance Card formula uses the number of wins against top-25 teams along with a teams win-loss record against schools ranked 26-50 and 51-100.) And road/neutral wins are weighted much more heavily than home wins. It's not perfect, but I don't think you can have a perfect system as long as some conferences are stronger than others. In the mean time, I really like the idea that you get an NCAA bid by winning games against tough opponents rather than by impressing the poll voters or by padding your computer ranking. Any alternative would be even worse for the smaller schools IMHO.

StanfordAggie wrote:And if you think that USU has a hard time getting decent OOC games now, I can only imagine how it would be if polls determined who received tournament bids the BCS schools would only play OOC games against cupcakes, since avoiding losses would be far more important than increasing one's SOS.

Isn't that pretty much what they do now anyway? Nearly all of which are on their home courts?

For example: Syracuse.

They played a total of ZERO non-conference road games this year. Who did they play at home?

Kutztown
LeMoyne
Northern Iowa (decent, RPI high 80's)
Canisius
Detroit (tournament hosted by Syracuse)
William & Mary (tournament hosted by Syracuse)
Cornell
North Carolina State (decent, but sub 100 rpi)
Colgate
Iona
Morgan State
Drexel

StanfordAggie wrote: You really want college basketball to rely on polls like the BCS?

Yes, I would like it very much if college basketball relied on a poll like the BCS, because USU would in the field every year based on its RPI and other computer rankings and polls.

Also, as pointed out by treesap, power conference schools already avoid playing any decent competition pre-season, and rarely if ever play away from home because they know they should just stockpile wins and that they'll get their chances at Top 50 or Top 100 wins during conference play. See Cincinatti's schedule this year, and they are supposedly on the bubble now.

The system has many flaws, but I will continue to posit that as long as Top 25 wins, Top 50 wins, and Top 100 wins are the key factor, the system is tilted severely in favor of BCS conference teams, who get many opportunities for such wins, and against mid-major teams, who get relatively few opportunities.

And to get back to silentmaster's original question, I guess the answer will depend on what happens this year; i.e. whether or not USU gets an at-large if it needs one. I think a midmajor who wins virtually all of the games it should win should be rewarded every bit as much as a BCS team that wins 10-20% of the games it plays against good teams.

The NCAA, as usual, needs to grow a pair and mandate a minimum number of road non-conference games for ALL D1 schools. All of them. Everybody is sick of Krzyzewski camping at Cameron until January.

MahlerFan1 wrote:The NCAA, as usual, needs to grow a pair and mandate a minimum number of road non-conference games for ALL D1 schools. All of them. Everybody is sick of Krzyzewski camping at Cameron until January.

Amen.

StanfordAggie wrote:
What is your alternative, then? You really want college basketball to rely on polls like the BCS?

Yes, I do. As long as there is a playoff to determine the champion, I don't have any problem whatsoever with the bids and seeding being determined by computers and polls. I don't think that most people's beef with the BCS is that it is determined by computers and polls. I think the biggest beef is that it is not played out on the field.

Rabidchild wrote:

StanfordAggie wrote: You really want college basketball to rely on polls like the BCS?

Yes, I would like it very much if college basketball relied on a poll like the BCS, because USU would in the field every year based on its RPI and other computer rankings and polls.

Also, as pointed out by treesap, power conference schools already avoid playing any decent competition pre-season, and rarely if ever play away from home because they know they should just stockpile wins and that they'll get their chances at Top 50 or Top 100 wins during conference play. See Cincinatti's schedule this year, and they are supposedly on the bubble now.

The system has many flaws, but I will continue to posit that as long as Top 25 wins, Top 50 wins, and Top 100 wins are the key factor, the system is tilted severely in favor of BCS conference teams, who get many opportunities for such wins, and against mid-major teams, who get relatively few opportunities.

And to get back to silentmaster's original question, I guess the answer will depend on what happens this year; i.e. whether or not USU gets an at-large if it needs one. I think a midmajor who wins virtually all of the games it should win should be rewarded every bit as much as a BCS team that wins 10-20% of the games it plays against good teams.

As soon as the NCAA starts using computer rankings and polls to choose at-large teams rather than quality wins, then teams will design their schedules to maximize their poll/computer ranking. All the BCS schools will play schedules like Cal did last year: Teams that are good enough that it won't kill their RPI, but not so good that they have any risk of losing. All the BCS schools will have all the top spots in the RPI, since the non-BCS schools will have their RPI's killed in conference play. And polls are even worse. We lost our third game of the year and almost dropped out of the top 25. At the point where polls become a criteria for getting an at-large bid, USU will never get a bid again unless we go undefeated. Using polls as a selection criterion sounds great when we're ranked, but I'm sure glad it wasn't a criteria for selection in 2006 or 2010. (And I have a feeling that we'll drop out of the top 25 if we lose in the WAC tournament, so that criteria would probably ultimately hurt us this year as well.)

But as I said, to me, it's mostly philosophical. Do you want a system where you make the tournament by gaming your computer ranking and whoring yourself out to pollsters? Or do you want a system where you make the tournament by winning games? USU got screwed this year because all of our opponents (both in out of the WAC) turned out to be lousy this year, but in general I think the system is much better for us than the alternatives.

StanfordAggie wrote:
As soon as the NCAA starts using computer rankings and polls to choose at-large teams rather than quality wins, then teams will design their schedules to maximize their poll/computer ranking. All the BCS schools will play schedules like Cal did last year: Teams that are good enough that it won't kill their RPI, but not so good that they have any risk of losing. All the BCS schools will have all the top spots in the RPI, since the non-BCS schools will have their RPI's killed in conference play. And polls are even worse. We lost our third game of the year and almost dropped out of the top 25. At the point where polls become a criteria for getting an at-large bid, USU will never get a bid again unless we go undefeated. Using polls as a selection criterion sounds great when we're ranked, but I'm sure glad it wasn't a criteria for selection in 2006 or 2010. (And I have a feeling that we'll drop out of the top 25 if we lose in the WAC tournament, so that criteria would probably ultimately hurt us this year as well.)

But as I said, to me, it's mostly philosophical. Do you want a system where you make the tournament by gaming your computer ranking and whoring yourself out to pollsters? Or do you want a system where you make the tournament by winning games? USU got screwed this year because all of our opponents (both in out of the WAC) turned out to be lousy this year, but in general I think the system is much better for us than the alternatives.

You seem to be saying that (in this alternate world we are discussing) you would have to have a top 25 ranking or a top 25 RPI to get an at-large bid. Last I checked, there are 30+ at large bids, and many of the automatic bids will also go to teams currently in the top 25. So, even if the aggies or other midmajor teams lost a late season game and dropped out of the top 25, it wouldn't be that drastic because at large bids could come from the 15-45 (polls/rpi) range of teams. That is an area that the aggies normally do well in.

Like others have said, most middle ground (and high level) BCS teams already are scheduling a heavy home OOC schedule with mostly cupcakes to "pad" their wins or prepare them for conference play like the aggies do. Under the current system, they still have lots of easy wins in OOC, and then rely on their 10 chances to beat top notch teams including hosting these teams at home. I agree that the current system favors "BCS" schools than would a system that relies more heavily on RPI. If a great team pads their OOC record, fine, they'll also do well in the conference and their RPI will be through the roof. If a mediocre BCS team pads its OOC record, they'll still lose a lot of in conference games, but their RPI will still end up in bad shape because their own record will suffer - and that is the highest contributor to the RPI. If they happen to pull off a couple of marquee wins (which they often do) the overall conference record (and RPI) would hold them back rather than saying (as is the case now) "well they have a 78 RPI, but they won a home game against UCONN that one time, they have to be in!" Well, if USU had all the Big East teams come to the Spectrum, I think we could pick up a couple big name wins to justify us getting in, but we don't have that luxury.

The RPI is designed to try to create an objective standard to judge teams off of. I have to say, StanfordAggie, I am surprised that a mathmatical guy like you is not as on board with the idea of having a computer formula to objectively compare teams have more weight in the selection process. Is it the principle of it, or do you just think that the RPI formula is not the right one?

Sure it may get gamed by a team or two each year, but I think the in conference records would weed out the big conference pretenders that are trying to game the system, because they'll still lose a bunch of games in the conference, thus lowering their RPI/poll rankings significantly. If anything, it may have the effect of midmajors trying to do what Stew is doing, and be more conscious of their own final records rather than the elusive top 25 road win. Rather than prostituting themselves out to major conferences for 1 and dones (and likely losses), they might play more midmajors like us. That would maybe send BCS teams scrambling to fill their schedules, and even force them out on the road every once in a while.

Rabidchild wrote:

MahlerFan1 wrote:The NCAA, as usual, needs to grow a pair and mandate a minimum number of road non-conference games for ALL D1 schools. All of them. Everybody is sick of Krzyzewski camping at Cameron until January.

Amen.

This wouldn't necessarily guarantee the Arkansas-Pine Bluffs of the world any nonconference D1 home games. But at the very least, it would force the bluebloods to go on the road before conference season. And that would be a start. I'm always amazed at the impotence of the NCAA in the face of caste systems and scheduling injustices. I love how the professional leagues guarantee parity in scheduling. Even the Yankees and the Lakers have to play as many road as home games. The NCAA can't quite do that, but they CAN take baby steps. They're too useless to do so, though. We're talking about an organization the farmed out its most profitable postseason to a self-interested cartel and refuses to do anything to take it back...

noods wrote:

StanfordAggie wrote:
As soon as the NCAA starts using computer rankings and polls to choose at-large teams rather than quality wins, then teams will design their schedules to maximize their poll/computer ranking. All the BCS schools will play schedules like Cal did last year: Teams that are good enough that it won't kill their RPI, but not so good that they have any risk of losing. All the BCS schools will have all the top spots in the RPI, since the non-BCS schools will have their RPI's killed in conference play. And polls are even worse. We lost our third game of the year and almost dropped out of the top 25. At the point where polls become a criteria for getting an at-large bid, USU will never get a bid again unless we go undefeated. Using polls as a selection criterion sounds great when we're ranked, but I'm sure glad it wasn't a criteria for selection in 2006 or 2010. (And I have a feeling that we'll drop out of the top 25 if we lose in the WAC tournament, so that criteria would probably ultimately hurt us this year as well.)

But as I said, to me, it's mostly philosophical. Do you want a system where you make the tournament by gaming your computer ranking and whoring yourself out to pollsters? Or do you want a system where you make the tournament by winning games? USU got screwed this year because all of our opponents (both in out of the WAC) turned out to be lousy this year, but in general I think the system is much better for us than the alternatives.

You seem to be saying that (in this alternate world we are discussing) you would have to have a top 25 ranking or a top 25 RPI to get an at-large bid. Last I checked, there are 30+ at large bids, and many of the automatic bids will also go to teams currently in the top 25. So, even if the aggies or other midmajor teams lost a late season game and dropped out of the top 25, it wouldn't be that drastic because at large bids could come from the 15-45 (polls/rpi) range of teams. That is an area that the aggies normally do well in.

Like others have said, most middle ground (and high level) BCS teams already are scheduling a heavy home OOC schedule with mostly cupcakes to "pad" their wins or prepare them for conference play like the aggies do. Under the current system, they still have lots of easy wins in OOC, and then rely on their 10 chances to beat top notch teams including hosting these teams at home. I agree that the current system favors "BCS" schools than would a system that relies more heavily on RPI. If a great team pads their OOC record, fine, they'll also do well in the conference and their RPI will be through the roof. If a mediocre BCS team pads its OOC record, they'll still lose a lot of in conference games, but their RPI will still end up in bad shape because their own record will suffer - and that is the highest contributor to the RPI. If they happen to pull off a couple of marquee wins (which they often do) the overall conference record (and RPI) would hold them back rather than saying (as is the case now) "well they have a 78 RPI, but they won a home game against UCONN that one time, they have to be in!" Well, if USU had all the Big East teams come to the Spectrum, I think we could pick up a couple big name wins to justify us getting in, but we don't have that luxury.

The RPI is designed to try to create an objective standard to judge teams off of. I have to say, StanfordAggie, I am surprised that a mathmatical guy like you is not as on board with the idea of having a computer formula to objectively compare teams have more weight in the selection process. Is it the principle of it, or do you just think that the RPI formula is not the right one?

Sure it may get gamed by a team or two each year, but I think the in conference records would weed out the big conference pretenders that are trying to game the system, because they'll still lose a bunch of games in the conference, thus lowering their RPI/poll rankings significantly. If anything, it may have the effect of midmajors trying to do what Stew is doing, and be more conscious of their own final records rather than the elusive top 25 road win. Rather than prostituting themselves out to major conferences for 1 and dones (and likely losses), they might play more midmajors like us. That would maybe send BCS teams scrambling to fill their schedules, and even force them out on the road every once in a while.

If you were going to do this, the RPI is quite clearly not the right formula. The NCAA would acknowledge as much. It was never intended to be a ranking like the ones Sagarin or Pomeroy produce where #1 is the best team and #2 is the second-best team and so on. It just gives a general idea of the relative strength of teams. So selecting teams with the highest RPI rankings doesn't make sense in the first place. Moreover, it is possible to get a very high RPI ranking despite never beating a decent team all year. (See last year's Cal team, for example... Or as much as I hate to say it, this year's USU team.)

And every other computer ranking is just as bad. If you use a ranking that considers point spread (i.e. Sagarin's predictor), then you give teams an incentive to play the weakest opponents possible and run up the score. If you use a ranking like the Sagarin Elo Chess that is based off win-loss record and SOS, then you give teams an incentive to play as many mediocre teams as possible but no one so good that you risk losing. Under the current system, there is an incentive to play the most challenging opponents possible. That's the incentive system that I would like to see. Even with this system in place, a lot of the top programs only play home games against cupcakes in OOC play. Do you really want to create a system that gives them even less incentive to play challenging games?

And polls... I'm frankly dumbfounded that we're even having this conversation. How many times have we been anywhere near getting ranked at the end of the season? We went from #17 to almost dropping out of the top 25 entirely after one loss, and we almost certainly will drop out if we don't beat St. Mary's this week. If we lose to St. Mary's and then lose in the WAC tournament, I'll be surprised if we get a single vote. And this is in a world where these votes are meaningless. In a world where these votes determine a team's NCAA future, do you really think that many sportswriters are going to pick Utah State over UNC when they fill out the bottom line on their ballot? If polls become a component in the selection process, then we have probably ensured that we will never get another at-large bid unless we win 30 games or so.

Also, it seems like there is a system for choosing postseason participants that is based entirely on polls and computer rankings. It's called the BCS. And everyone knows that it always chooses the most deserving teams, and there are never any instances where a less-deserving team is picked over a better team. For all the issues with the current system, I haven't seen a system that I like better.

Automatic bid to the Conference Tournament winner versus regular season conference winner?

Conf Tourney winner gets bid:
Pros:
-(potentially) generates lots of money
-adds excitement and uncertainty (cinderellas)
-may get more teams in the ncaa tournament this way
Cons:
-other than seeding and at-large bids, negates the regular season
-may wear out players right before the big dance
-may not send the best team to the big dance (less likely to win games in the ncaa tourney?)

This is at the discretion of the conferences, so they've apparently decided that the benefits outweigh the costs...at least for the conference....aside from the Ivy League.

Thoughts?

Conference tournaments are a bane. The ones for low- and mid-major conferences have too much riding on them and the ones for major conferences are essentially meaningless. They should be done away with. I know Whelliston loves them, but the risks outweigh the benefits.