|
Post by dancingbear on Dec 7, 2010 11:59:10 GMT -5
I propose a Division I Volleyball Championship Subdivision (VCS), (just like the formerly known Division I-AA in football.
What everyone knows but won't admit is the "mid-majors" in volleyball can not compete 98% of the time with the big conferences by the end of the season.
This will remove the necessity of the conference auto-bids.
|
|
|
Post by pogoball on Dec 7, 2010 12:24:53 GMT -5
I brought the point up in another thread, but it is better here: Choosing which teams deserve to be invited to the tournament is one set of criteria. Choosing how to seed/place those teams in the tournament doesn't have to be and, IMO, shouldn't be the same criteria. We could use RPI for placing teams in the tournament, but we could use a Pablo system for seeding and placing them. Conceivably it would be appropriate to have different or slightly different criteria for selecting the tournament vs seeding the tournament.* But why should RPI be one of them? If it's fundamentally flawed (and it is), get rid of it. Let's not make different criteria just for the sake of having different criteria. * For example (not that I necessarily support the idea) you could use the non-weighted version of Pablo (or an Elo system) for selecting who's invited and the weighted version for determining the seeds. That would be more reasonable than using the RPI for one and Pablo (or an Elo system) for the other. RPI "should" be one of them because the NCAA is invested in it, it gives the NCAA results that it wants and they may be willing to consider this compromise. I also don't think you understand the nature of what I'm saying when you make your above point. To repeat myself: RPI seems to give the NCAA results it wants. If you can change the nature of the debate and concede that it gives them the FIELD they want but possibly not the ARRANGEMENT of the field they want, you might get them to agree on that point and actually "DO something about it".
|
|
|
Post by The Bofa on the Sofa on Dec 7, 2010 12:34:51 GMT -5
I also don't think you understand the nature of what I'm saying when you make your above point. To repeat myself: RPI seems to give the NCAA results it wants. If you can change the nature of the debate and concede that it gives them the FIELD they want but possibly not the ARRANGEMENT of the field they want, you might get them to agree on that point and actually "DO something about it". Agreed 100%, but with the additional caveat: does it really give the NCAA "results it wants" or is it "results that are good enough"? In order to make that determination, we have to establish exactly what it is that the NCAA wants to achieve, and then we can examine to what extent the use of RPI accomplishes that. Given that, we can then explore other possibilities, and compare them to what we get with RPI. If we really want to effect a change, we have to come up with something that does a better job of meeting the NCAA's objectives for the tournament. But to do that, we first have to know what are the NCAA's objectives for the tournament.
|
|
|
Post by karplets on Dec 7, 2010 13:01:48 GMT -5
RPI "should" be one of them because the NCAA is invested in it, it gives the NCAA results that it wants and they may be willing to consider this compromise. I also don't think you understand the nature of what I'm saying... You're taking a pragmatic view and you're right, I didn't see that. But if there's a fundamental problem and you don't fix it, what's the point? It's putting a band-aid on a much bigger problem. If the RPI gives the NCAA the results it wants, as you put it, then what can we do? Convince them that what they want isn't what they want? and that what we're offering them, which they don't want, is what they want? I'm not a snake-oil salesman. Maybe we're missing something here by pretending to read the NCAA's mind that the RPI gives the NCAA what it wants. Organizations like the NCAA don't want to be embarrassed and they value appearances -- they don't want to appear to be unfair. If the RPI can be shown to make the NCAA seem unfair in how they determine the tournament, it (the RPI) becomes a potential liability to them. I don't think the NCAA is so invested in the RPI that it will stick with it if it becomes a potential embarrassment. And that's an opening for change.
|
|
|
Post by The Bofa on the Sofa on Dec 7, 2010 13:23:16 GMT -5
If the RPI gives the NCAA the results it wants, as you put it, then what can we do? See my comment above: convince them that some alternate process does an even better job of "giving them what they want." It all depends on what they want. Unfortunately, when it comes to the outcomes, I don't know what that is. (I've been saying that for years, that I don't know the objective of RPI)
|
|
|
Post by karplets on Dec 7, 2010 13:56:47 GMT -5
See my comment above: convince them that some alternate process does an even better job of "giving them what they want." It all depends on what they want. Unfortunately, when it comes to the outcomes, I don't know what that is. (I've been saying that for years, that I don't know the objective of RPI) Yes, excellent point which I didn't mean to ignore. Since we can't read minds, I don't think it hurts to act as if they're not that different from us. Even if they may not care as much as we do about the fairness of the seedings, as I said neither do they want to give out the appearance they don't care. Why have they stuck with the RPI in the first place? Guesses and observations: 1) Inertia 2) It was created by the NCAA. (Alternatives suffer from the "not made here" bias) 3) It was made back in the day before computers were as powerful as they are now. It doesn't require the sort of computing power that "real" statistical methods require. (But why stick with it? Inertia again? Sentimental value?) 4) It is transparent (to a point and then the NCAA doesn't want it to be) -- for example anyone who's willing to go to the bother can duplicate or simulate the RPI up to the secret bonus adjustments. That's what Rich Kern does, right? Compared to the algorithms of other rating systems, it isn't a "black box" 5) They just haven't given it much thought. They don't know what the alternatives are or any idea how they work. (It's possible none of them, for instance, have ever been chessplayers and had a chess rating)
|
|
|
Post by The Bofa on the Sofa on Dec 7, 2010 14:08:15 GMT -5
See my comment above: convince them that some alternate process does an even better job of "giving them what they want." It all depends on what they want. Unfortunately, when it comes to the outcomes, I don't know what that is. (I've been saying that for years, that I don't know the objective of RPI) Yes, excellent point which I didn't mean to ignore. Since we can't read minds, I don't think it hurts to act as if they're not that different from us. I disagree. I think it is a huge mistake to assume that they have the same objectives as us. My evidence? The fact that they continue to use the same process, year after year, despite the fact that it is never what "we" want it to be. Moreover, they use the same process in every other minor sport, too, despite the fact that fans of those sports complain just as much. So either the NCAA is so clueless as to not be able to come up with a better solution for achieving the goal, or they have a different goal altogether. Considering that they HAVE created a better solution in men's basketball, we know that they are capable. Hence, it must be that they don't want to.
|
|
|
Post by karplets on Dec 7, 2010 14:54:31 GMT -5
I disagree. I think it is a huge mistake to assume that they have the same objectives as us. My evidence? The fact that they continue to use the same process, year after year, despite the fact that it is never what "we" want it to be. Moreover, they use the same process in every other minor sport, too, despite the fact that fans of those sports complain just as much. So either the NCAA is so clueless as to not be able to come up with a better solution for achieving the goal, or they have a different goal altogether. Considering that they HAVE created a better solution in men's basketball, we know that they are capable. Hence, it must be that they don't want to. I won't disagree with that assessment. My wording was sloppy at best. All I'm saying is it's possible our objectives can converge with theirs and our best bet is to gun for what we want. The more RPI looks like a potential embarrassment for the NCAA, the quicker they'll get rid of it. (by the way, what is their solution in men's basketball? They use Sagarin and other Elo-type computer ratings? in conjunction with ____?) *** add thought Or they were pressured into it for basketball. There's more money involved in it and if corporate sponsors weren't happy with the old system, that would create pressure for change.
|
|
|
Post by The Bofa on the Sofa on Dec 7, 2010 15:04:22 GMT -5
I disagree. I think it is a huge mistake to assume that they have the same objectives as us. My evidence? The fact that they continue to use the same process, year after year, despite the fact that it is never what "we" want it to be. Moreover, they use the same process in every other minor sport, too, despite the fact that fans of those sports complain just as much. So either the NCAA is so clueless as to not be able to come up with a better solution for achieving the goal, or they have a different goal altogether. Considering that they HAVE created a better solution in men's basketball, we know that they are capable. Hence, it must be that they don't want to. I won't disagree with that assessment. My wording was sloppy at best. All I'm saying is it's possible our objectives can converge with theirs and our best bet is to gun for what we want. The more RPI looks like a potential embarrassment for the NCAA, the quicker they'll get rid of it. (by the way, what is their solution in men's basketball? They use Sagarin and other Elo-type computer ratings? in conjunction with ____?) They don't tell them to use anything specific. They tell the committee to 1) pick the best teams possible, and 2) use whatever resources they can to help inform their decisions. They are allowed to use polls or even other computer rankings besides RPI (they are also given RPI). These are the explicit instructions. They are far, far different from the directions given in other sports. Then again, the key with the basketball committee is that it is sent out to research teams all season long, and presumably, the NCAA allows access to resources during that time. Given the prestige and PR importance of the job, the committee members dedicate great time and effort to it. Volleyball committee members don't have that luxury.
|
|
|
Post by karplets on Dec 7, 2010 15:18:16 GMT -5
They don't tell them to use anything specific. They tell the committee to 1) pick the best teams possible, and 2) use whatever resources they can to help inform their decisions. They are allowed to use polls or even other computer rankings besides RPI (they are also given RPI). These are the explicit instructions. They are far, far different from the directions given in other sports. Then again, the key with the basketball committee is that it is sent out to research teams all season long, and presumably, the NCAA allows access to resources during that time. Given the prestige and PR importance of the job, the committee members dedicate great time and effort to it. Volleyball committee members don't have that luxury. Ah. Well, we can't expect to see that happen for volleyball, soccer, or any of these other non-revenue sports soon. I think the best, most cost-efficient lesson they can take from basketball under the circumstances is get rid of the RPI -- or, since they didn't entirely get rid of it, allow the use of Pablo or an Elo rating system and encourage its use as the primary objective measure in the Committee's procedure.
|
|
|
Post by The Bofa on the Sofa on Dec 7, 2010 15:29:17 GMT -5
They don't tell them to use anything specific. They tell the committee to 1) pick the best teams possible, and 2) use whatever resources they can to help inform their decisions. They are allowed to use polls or even other computer rankings besides RPI (they are also given RPI). These are the explicit instructions. They are far, far different from the directions given in other sports. Then again, the key with the basketball committee is that it is sent out to research teams all season long, and presumably, the NCAA allows access to resources during that time. Given the prestige and PR importance of the job, the committee members dedicate great time and effort to it. Volleyball committee members don't have that luxury. Ah. Well, we can't expect to see that happen for volleyball, soccer, or any of these other non-revenue sports soon. I think the best, most cost-efficient lesson they can take from basketball under the circumstances is get rid of the RPI -- or, since they didn't entirely get rid of it, allow the use of Pablo or an Elo rating system and encourage its use as the primary objective measure in the Committee's procedure. Then again, who's to say that an ELO approach is any better? i've pointed out before that I have found that when it comes to predictive ability, for example, a simple ELO approach doesn't perform any better than RPI. That would only be worse if you take out the time component. Yes, full blown Pablo outdoes RPI in every way, but I can't say the same thing about the approach you describe (as you note, we should not be too optimistic about them accepting Pablo; I wouldn't, if I were the NCAA) Let me ask you this: if it were found that RPI performs better in terms of predicting match outcomes (in whatever dataset you decide to test), would you agree that they should use RPI instead? The key to this approach is to first define standard and what you are trying to achieve, and then search for the best way to do it without making any a priori value judgments.
|
|
|
Post by redbeard2008 on Dec 7, 2010 15:30:35 GMT -5
The NCAA uses the same criteria and goals for all its minor sports, I believe, including an unmodified RPI system, so I think radical and complex changes would need to apply to all those sports and not significantly increase costs. In many of these sports, however, there just isn't the kind of fan/public feedback we have in volleyball. The NCAA isn't likely to make changes just for volleyball, however. That's why I think more evolutionary changes (going to the modified RPI system used the major sports and allowing human judgment, rather than slavish adherence to RPI, to adjust seedings to better reflect comparative team strengths, while still meeting other NCAA goals), rather than revolutionary ones, might have a better chance of getting made.
|
|
|
Post by The Bofa on the Sofa on Dec 7, 2010 15:43:53 GMT -5
The NCAA uses the same criteria and goals for all its minor sports, I believe, including an unmodified RPI system, so I think radical and complex changes would need to apply to all those sports and not significantly increase costs. In many of these sports, however, there just isn't the kind of fan/public feedback we have in volleyball. The NCAA isn't likely to make changes just for volleyball, however. That's why I think more evolutionary changes (going to the modified RPI system used the major sports Has anyone pointed out yet that the "modified RPI system used [in] the major sports" would have served to HELP UNI and HURT Hawaii? UNI played it's entire non-conference schedule on the road. They won all their matches in conference play, so that is a wash. The only effect it would have had is on those two matches in the conference tournament. However, they were already 5th in RPI before the conference tournament, and that wasn't going to take them down. Meanwhile, Hawaii had all but their three conference tournament matches in the friendly confines of the SSC (is Stan Sheriff still alive? I like to point out in any post involving both UNI and Hawaii that before going to Hawaii and getting the building named after him, Stan was the head football coach at Iowa State Teacher's College, which later got renamed the University of Northern Iowa). Therefore, they would get less credit for their win over UCLA (and other home wins) and more penalty for their loss to USC. Similarly, I don't think that the modified RPI would address the regional bias in volleyball RPI, either. So I'm surprised that so many people support this modified RPI idea, considering that it will hurt Hawaii and help the small schools like UNI and Dayton (Duke played half its non-conference matches at home). Given the resources allotted to the committee, the NCAA most definitely wants to avoid subjective judgments as much as possible, and that is smart. Given all the committees they have to manage, they can't afford to allow them run off uncontrolled. They have to have very clear guidelines for the committees to follow, and will ensure that they are adhered to. They don't want the selection criteria to be dependent on the makeup of the committee, so they give the committee very explicit instructions on how to make the decisions. That works, we have seen it. Despite the continual turnover of members of the committee, they ultimately act the same way all the time. That is by design.
|
|
|
Post by jake on Dec 7, 2010 16:11:56 GMT -5
If you know that the committee relies on RPI so much: teams should schedule so as to acquire the best RPI by seasons end. Makes no sense. After the preliminary games, everyone plays in their conferences. Teams in weak conferences will earn higher RPI's than those in the PAC-10, Big-12, and Big-10.
Wonder,....you hit the problem squarely on the HEAD!
And, that's the issue. Sure, I suppose it is possible to schedule just the right OOC schedule to compensate for your conference play. But, why should that be the criteria that determines if a team is good enough to make the NCAA's?
Suggestion: Create a selection committee guideline, that limits the number a teams that can be selected from any one conference to 50%.OUT!
|
|
|
Post by karplets on Dec 7, 2010 16:15:52 GMT -5
Then again, who's to say that an ELO approach is any better? i've pointed out before that I have found that when it comes to predictive ability, for example, a simple ELO approach doesn't perform any better than RPI. That would only be worse if you take out the time component. Yes, full blown Pablo outdoes RPI in every way, but I can't say the same thing about the approach you describe (as you note, we should not be too optimistic about them accepting Pablo; I wouldn't, if I were the NCAA) Let me ask you this: if it were found that RPI performs better in terms of predicting match outcomes (in whatever dataset you decide to test), would you agree that they should use RPI instead? Good points which I don't have time to address right now. But a couple notes: 1) I would take Elo over RPI as long as they were at least comparable in terms of prediction or "retrodiction" BUT I understand if most people would hesitate to agree. The burden of proof would definitely be on me to make a case for it. I think it can on QUALITATIVE grounds even if strictly QUANTITATIVE ones don't clearly indicate the superiority of one over the other 2) My guess is, based on the work of someone on BigSoccer named cpthomas, that an Elo system would show at least slight measurable superiority when it came to comparing different regions. And regional comparisons is obviously a big big problem in volleyball -- more so even than soccer. 3) It's not surprising that it's hard to distinguish the superiority of one system over another. Well, maybe it is at first glance but when you think of it, you can see why various systems seem roughly comparable in how well they predict the outcome of matches. Heck, I even wonder sometime how far off it would be if one solely relied on won-loss records (maybe I vaguely remember someone coming up with some numbers on that)
|
|