Regional projections from earlier today

[raginsaints]

Two consecutive sweeps are doable with our opponents.


igeaux.mobi

[LifeTimeCajun]

Two consecutive sweeps are doable with our opponents.

They are doable, but can we stay mentally prepared to do it for 6 strait games?
igeaux.mobi

igeaux.mobi

[LSUalum1]

I have a question for Brian.

Why do we even have RPI Rankings if we have at large teams in the 30's, 40's, and 50's - in RPI rankings - getting in before teams in the 20's?

I'm not saying LSU deserves to be in the NCAA tournament because they don't this year - unless they sweep the rest of their games (which they won't). But, if the reason for having RPI rankings is to determine who is the better team, why don't they use it to put those teams in the tournament?

I'm I missing something?

[raginsaints]

RPI rankings are important but they're not the end all be all for things like this. LSU has a good RPI but a lot of that is due to OWP and OOWP, Brian can correct me if I'm wrong about that. If they don't make their own conference tournament they shouldn't make the big tourney, in my opinion.

[Cajun90]

I have a question for Brian.

Why do we even have RPI Rankings if we have at large teams in the 30's, 40's, and 50's - in RPI rankings - getting in before teams in the 20's?

I'm not saying LSU deserves to be in the NCAA tournament because they don't this year - unless they sweep the rest of their games (which they won't). But, if the reason for having RPI rankings is to determine who is the better team, why don't they use it to put those teams in the tournament?

I'm I missing something?

Well I can probably give you his answer in short form. It is one tool of many that the NCAA uses. They will obviously look at overall record and then record against top 25, 50, & 100. They will also look at last 10 games played and they will consider where you finished in your conference.

We all know that the RPI is flawed. The big flaw is the weight given to your OWP. As such it is possible for a team to play a bunch of games against teams with very good winning percentages and lose those games but still end up with a high RPI. LSU fits that bill pretty well.

I think the RPI is more of a tool to eliminate teams rather than to put them in. Another words the committee probably chops off any team below an RPI of around 75. They know none of these teams are getting in. Now they can start the process of narrowing down the at large field from this group of top 75 RPI teams.

[GoneGolfin]

I have a question for Brian.

Why do we even have RPI Rankings if we have at large teams in the 30's, 40's, and 50's - in RPI rankings - getting in before teams in the 20's?

I'm not saying LSU deserves to be in the NCAA tournament because they don't this year - unless they sweep the rest of their games (which they won't). But, if the reason for having RPI rankings is to determine who is the better team, why don't they use it to put those teams in the tournament?

I'm I missing something? _

I bolded your key statement above. Even the NCAA realizes that such a goal is well beyond the capabilities of the RPI formula. It is well beyond the capabilities of any formula ... but especially the RPI.

And if you recognize the RPI problems that nobody on the NCAA selection committee fully understands (except for the statisticians that aid the process ... whom are probably residing in a dark basement somewhere), it is obvious that using the RPI exclusively to determine the best teams would be a cruel pathetic joke. The RPI is so full of gaping holes, that it is not hard to "game" or "rig" it in your favor (I have written several detailed posts in the last year or two pertaining to exactly how this can be accomplished). And even for those not smart enough to "play the RPI in their favor", if you happen to play in an elite conference not in the west, a nice portion of the potential "gaming" or "rigging" is already done for you by default.

While the NCAA relies on the RPI much more than it should, it is simply one of the tools they use ... but it is the principal tool and the tool they utilize to justify their selections when they must.

Brian

[GoneGolfin]

_ RPI rankings are important but they're not the end all be all for things like this. LSU has a good RPI but a lot of that is due to OWP and OOWP, Brian can correct me if I'm wrong about that. If they don't make their own conference tournament they shouldn't make the big tourney, in my opinion. _

Yes ... but I would pare this down to being all about OWP. OWP is half the formula and can be used to reasonable extent to manipulate your overall RPI. OOWP is much less controllable and is worth no more than the winning percentage component of the formula. LSU's RPI is as high as it is because of their OWP, not their OOWP.

If you want an egregious example, look at Georgia. Georgia is 25-23, but has a Base RPI of 18 (Adjusted RPI is 21). Why is this? Quite simply, their opponents win more than 63% of their games. Thus, you can win half of your games and sport a Top 20 RPI simply by playing the right schedule. And this does not take into account some additional tricks that could be employed to spruce their OWP even more (while not making the schedule any more difficult ... and possibly making it easier).

At 25-22, Auburn (#22 Base RPI, #23 Adjusted RPI) is another prime example as their OWP is .6268.

Brian

[LSUalum1]

Thanks for the quick reply. As I said, LSU doesn't deserve to go to the tournament because of their poor SEC play because their overall record is not that bad and we all know their RPI is good enough.

My complaint is if we use the RPI just to eliminate the "over 75 RPI's" - what's really the use.

I always believed that if you are going to spend all this time with this RPI formula, it should be used as the BCS uses their formula.

[raginsaints]

_ Yes ... but I would pare this down to being all about OWP. OWP is half the formula and can be used to reasonable extent to manipulate your overall RPI. OOWP is much less controllable and is worth no more than the winning percentage component of the formula. LSU's RPI is as high as it is because of their OWP, not their OOWP.

If you want an egregious example, look at Georgia. Georgia is 25-23, but has a Base RPI of 18 (Adjusted RPI is 21). Why is this? Quite simply, their opponents win more than 63% of their games. Thus, you can win half of your games and sport a Top 20 RPI simply by playing the right schedule. And this does not take into account some additional tricks that could be employed to spruce their OWP even more (while not making the schedule any more difficult ... and possibly making it easier).

At 25-22, Auburn (#22 Base RPI, #23 Adjusted RPI) is another prime example as their OWP is .6268.

Brian _

That's ridiculous.

[Cajun90]

it should be used as the BCS uses their formula.

Well that should get this thread to at least 20 pages.

While I did not get into the details I certainly believe the RPI is used beyond determining a cut off point. I just think that is one of the big purposes. You can eliminate 225 teams right off the top. Now you can spend time looking closer at those 75 and yes the RPI will still have weight in that decision tree.

[Ned Clampett]

it should be used as the BCS uses their formula. _

LSU, Brian just told you why it should NEVER be used as the BCS uses theirs.

[GoneGolfin]

My complaint is if we use the RPI just to eliminate the "over 75 RPI's" - what's really the use.

This is certainly not the case. The NCAA selection board for at-large teams does in fact begin with at-large candidates in the Top 75 RPI plus conference regular season champions only. But the RPI is also used extensively throughout the selection process ... everything from national seed selection before at-large bids are awarded to selecting the last few at-large teams. I remember one year where the national seeds were the RPI teams ranked 1->8 save one team that was #9.

I always believed that if you are going to spend all this time with this RPI formula, it should be used as the BCS uses their formula. _

RPI needs to be taken out and shot.

Brian