|
Post by ugopher on Nov 19, 2012 11:52:26 GMT -5
This is interesting. Did not know this.
This does make sense, though, as to why they try to keep teams fairly close together. As the Guru mentions later, if you were a low ranked team on the East coast who didn't have their travel expenses reimbursed and had to face Stanford in the first round, would you spend the money to travel across the country?
|
|
|
Post by redbeard2008 on Nov 19, 2012 12:10:42 GMT -5
Well, we'll see which consideration applies in the case of PSU: the "one bad loss" or the "one fluke loss".
|
|
|
Post by dorothymantooth on Nov 19, 2012 12:22:05 GMT -5
Well, we'll see which consideration applies in the case of PSU: the "one bad loss" or the "one fluke loss". Well, its both a bad loss and a fluke loss isnt it? It counts on their record, it hurts them, but not a loss that suggests anything other than, on any given night. I nkow it isnt a loss that makes me have any serious concerns about this team. It is what it is.
|
|
|
Post by redbeard2008 on Nov 20, 2012 0:43:58 GMT -5
Okay, just out of curiosity, I took the current AVCA, Pablo, and RPI rankings, summed them, and then divided them by three, and came up with these brackets:
Omaha: 1. PSU, 8. Louisville, 9. Nebraska, 16. Kansas
Berkeley: 2. Stanford, 7. Washington, 10. Minnesota, 15. Ohio State
Austin: 3. Texas, 6. Oregon, 11. FSU, 14. Hawaii
West Lafayette: 4. UCLA, 5. USC, 12. BYU, 13. Florida
* Ties were: UCLA/USC, Louisville/Nebraska/Minnesota, FSU/BYU, and Kansas/Ohio State. In each case, I used RPI as the tie-breaker, except with Kansas/Ohio State, which I reversed, to avoid three seeded Big Ten teams in the same bracket.
Good, bad, indifferent?
Granted, there are still key matches left which could upset the apple cart, particularly: PSU @ MSU, Stanford @ Cal, UCLA @ Ore/vs. USC, etc. For instance, if UCLA beats Oregon, the Ducks and the Huskies could easily switch places, with Oregon going to Berkeley and Washington to Austin.
|
|
|
Post by baywatcher on Nov 20, 2012 2:43:56 GMT -5
I would take more of a 60% RPI, 25% Pablo, 15% AVCA line. The new rules say the Committee can look at other sources, not that they have to look at Pablo and AVCA. Other sources could be a $1,000 bill from a coach wanting to be seeded, I suppose.
|
|
|
Post by passerpearmanbauer on Nov 20, 2012 9:27:37 GMT -5
Wait. Now we're back to "above .500 record" rather than at or above? cal posted that they are now tournament-eligible at 15-14, with one match left. why can't anyone answer this simple question? i can't. that's for sure.
|
|
|
Post by redbeard2008 on Nov 20, 2012 13:57:28 GMT -5
I would take more of a 60% RPI, 25% Pablo, 15% AVCA line. The new rules say the Committee can look at other sources, not that they have to look at Pablo and AVCA. Other sources could be a $1,000 bill from a coach wanting to be seeded, I suppose. Is RPI worth 60%? I'd prefer a specific formula, that let's the chips fall as they may, starting out, rather than some kind of squishy "can look at" policy that might only provide a cover for continuation of the status quo, which tends to involve protecting some teams and exposing others. Now, obviously, where the chips fall might not always be workable. You can't have two regional hosts in the same region (Stanford and Nebraska, for instance). You wouldn't want more than two seeded teams from the same conference in the same region, etc. But those are all very transparent adjustments for very clear reasons. In the example I posted, which accords 1/3 shares to AVCA, Pablo, and RPI, with RPI also as a tie-breaker, it just so happens that only one adjustment was necessary, which was swapping Kansas and Ohio State, to avoid three seeded conference members in the same regional. Change the seeding of Nebraska (a regional host) up or down, however, and it gets a good deal more difficult. Some, of course, might claim that PSU shouldn't have to go to Omaha. But isn't this the kind of "path clearing" that has so much mucked up the process in the past? Making an easier road for one team usually means making a harder road for others.
|
|
|
Post by The Bofa on the Sofa on Nov 20, 2012 14:13:30 GMT -5
I would take more of a 60% RPI, 25% Pablo, 15% AVCA line. The new rules say the Committee can look at other sources, not that they have to look at Pablo and AVCA. Other sources could be a $1,000 bill from a coach wanting to be seeded, I suppose. Is RPI worth 60%? Probably not. It's likely much higher than that. From what has been reported, rpi is still a primary factor, and while the committee is not prevented from using other rankings they are directed to do so either. Given that, there is no reason to expect them to use other rankings to do anything besides tweak the margins.
|
|
|
Post by redbeard2008 on Nov 20, 2012 14:31:48 GMT -5
Probably not. It's likely much higher than that. From what has been reported, rpi is still a primary factor, and while the committee is not prevented from using other rankings they are directed to do so either. Given that, there is no reason to expect them to use other rankings to do anything besides tweak the margins. But the "guru" said that RPI was just another factor. Weighting RPI at 60%, or above, says otherwise. Louisville is the clear marker this year - where they fall could reveal a lot about what thumbs are on the scale.
|
|
|
Post by The Bofa on the Sofa on Nov 20, 2012 14:41:12 GMT -5
Probably not. It's likely much higher than that. From what has been reported, rpi is still a primary factor, and while the committee is not prevented from using other rankings they are directed to do so either. Given that, there is no reason to expect them to use other rankings to do anything besides tweak the margins. But the "guru" said that RPI wadd just another factor. Weighting RPI at 60%, or above, says otherwise. Louisville is the clear marker this year - where they fall could reveal a lot about what thumbs are on the scale. Kansaz is even better. Unseeded in Pablo and Avca but top 5 in rpi. If they get a top ten seed then rpi rules. As noted, weigh them equally and KU is more like 16.
|
|
|
Post by redbeard2008 on Nov 20, 2012 15:23:26 GMT -5
But the "guru" said that RPI wadd just another factor. Weighting RPI at 60%, or above, says otherwise. Louisville is the clear marker this year - where they fall could reveal a lot about what thumbs are on the scale. Kansaz is even better. Unseeded in Pablo and Avca but top 5 in rpi. If they get a top ten seed then rpi rules. As noted, weigh them equally and KU is more like 16. Actually, top 6, just to quibble a mite... Who has Kansas beat? #18 Iowa State (split), #21 Creighton, #25 KSU. No "bad" losses, though. The other marker could be Nebraska (#5 in RPI)? Do Michigan and MSU (#33 and #35 in RPI, both within the last ten matches) count as "bad" losses? Or does the "good" win against PSU (also within the last ten matches) erase all that? However, the loss to Ohio State (#20 in RPI) is also within the last ten matches... If weighed equally, Nebraska is more like a #9 seed, putting them in the same regional as the #1 seed, which only works if PSU is seeded #1. To "liberate" PSU from the Omaha regional, you'd have to move Nebraska up to #5 (or leave them at #5, if RPI is the starting point) or down to #12.
|
|
|
Post by karplets on Nov 21, 2012 12:11:50 GMT -5
I'm impressed that Pablo has achieved official status. Bofa and the others who worked to make this happen can be rightfully proud. I have to say, though, it came at a great cost - the inclusion of a coaches poll into the official process. First, we have no idea what weight the Committee gives to the three different rankings. The committee doesn't even know for sure at the start of a season how much weight they will assign to each of the three rankings and there's not much to stop them from changing their minds. Second, it will now be much harder to determine which ranking is to blame if the bracket or seedings appears to be flawed.
There were other surprises in the interview - such as the emphasis placed on the last 10 games and on the "bad loss" (the USC example). But I have to think about that some before I comment more on it.
|
|
|
Post by karplets on Nov 21, 2012 12:55:47 GMT -5
The NCAA should not use the coaches poll. They should use one ranking tool or the other, Pablo or RPI (or an Elo for those who know what I mean by that). They should reserve use for RPI under exceptional circumstances -- for example should serious issues arise regarding integrity of an outside ranking measure (be it Pablo, Massey, Sagarin, etc etc) But combining them into a composite ranking makes no sense.
|
|
|
Post by sirtrojan on Nov 21, 2012 13:27:10 GMT -5
I completely agree karplets. The problem with using three different rankings is they become a vague mish-mash of conflicting information if not approached in a consistent manner by all committee members. The committee could quantify a composite ranking by taking a mathematical mean or average of the three rankings but there is nothing to indicate that is their plan. A possible downside to having three rankings to choose from is committee members can select one of the rankings to confirm their own subjective opinion. I agree with choosing the best available ranking and using that as the primary guideline.
|
|
|
Post by Phaedrus on Nov 21, 2012 13:33:54 GMT -5
But the ranking themselves are an average of what each team have done through out the season. So byu averaging the average, you are smoothing the data but not any worse than what has been done already.
|
|