Why neither Cal nor UW got screwed (long)
Nobody has complained that anybody in the Pac-12 was screwed besides Cal (getting a 12 seed and a First Four game) and UW (so-called regular season champion left out of the tourney). But given how the selection is done, I donít think anybody got screwed.
Many of those who argue Cal got screwed in the seeding point out we were placed at 45 (which, if seeding doesnít have to get adjusted for bracketing purposes, would mean the best of the 12 seeds), yet had an RPI of 37 (which, if there were no adjustements in seeding for bracketing purposes would mean the best of the 10 seeds). Many of those who argue UW got screwed say that UWís RPI of 71 should be irrelevant, even if it would be close to historically the worst RPI to get in ever.
I realize these sets of people arenít the same, but still, we canít have it both ways, either RPI is relevant, or it is not. Additionally, history has shown that RPI is more relevant to getting in the field than it is to seeding, so RPI is likely to screw UW out of a spot in the field, RPI is not likely to help Cal avoid being ranked 45. I.e., both arguments are probably wrong, Calís 37 RPI does not mean we merit a 10 seed, and UWís 71 RPI means, barring REALLY compelling reasons to overcome RPI, UW should be out.
Letís look at Cal for the moment. The committee has made clear over and over again that wins over good teams matter a lot. Calís best wins from an RPI standpoint were the wins over Oregon, 63 RPI. There were 9 teams with worse RPIís than Cal who were seeded ahead of Cal (the only team with a better RPI than Cal that flipped behind us was 46 seed/35 RPI Harvard). With the exception of 42 seed North Carolina St., all of them had more than 1 win over top 50ís, and except for NC St. and 44 seed Colorado, all had at least 1 top 25 win. North Carolina St, just had the 1 top 50 over 50 RPI Texas, so it was close, but they also had a total of 4 RPI wins better than Cal, the Texas win, a win over Virginia at a neutral site, and sweeping 60 RPI Miami, home and road. Colorado got the 44 seed, in spite of a 62 RPI and a much worse conference record than Cal, but they did finish strong (finishing strong seems even more important in seeding than in getting in), their two 2 50 winís were both over Cal, and the game at a neutral site was over Cal. You could argue we should flip flop with Colorado, but we'd still be in the last 4 at larges and a First Four game, and since we'd have to match up with South Florida, probably still a 12 seed. 39 seed Virginia (53 RPI) is the only other team that flipped over Cal with fewer than 3 top 50 RPI wins, they only had 2, but one was on the road (v. 49 UNC) and the other was at home over #13 RPI Michigan. Given the emphasis on quality wins, hard to say Cal got screwed with a 45 seed.
Only 2 teams got at large bids with zero wins over top 50 teams, 45 seed/37 RPI Cal, and 50 seed/40 RPI Iona. Assuming that the team rankings werenít affected by how many were getting in, if they still had a field of 65, Cal would have been the last team in, Iona wouldnít have made it. In history (starting with 1994 when the basic RPI formula was used)l, including this year, only 10 teams have gotten an at large bid with zero wins over top 50 RPI teams. Only 3 of them were without a regular season conference title (Iona had a conference title), and 2 of those (Cal this year and Clemson last year) have gotten in since the field was expanded to 68. Iona was the last at large team in the tourney, and it likely helped that they had a regular season title that counted (combined with their 40 RPI, and Iíll say why some regular season titles donít count in discussing UW).
Note that among the first 6 teams left out (the committee didnít release order, but the six were Drexel, Mississippi, Seton Hall, Miami (Fl.), Nevada, and Oral Roberts), only Nevada failed to have any top 50 RPI wins.
I just donít see Cal has having gotten screwed here. Like RPI or not, we got in on the strength of RPI, and based on the committee's emphasis on good wins, and given who we beat (nobody), we probably got seeded in a reasonable spot.
As for UW, keep in mind that the regular season title was seriously discounted by the committee, not because we are the Pac-12, but because we donít play a true home and home round robin. Jerry Palm attended the mock media selection a couple of years ago (I donít remember if he was on the selection committee or just observed that year), and later mentioned that conference position, even conference titles, get ignored if the committee thinks there is a decent chance it was affected by not playing the same schedule as a team close behind it.
Monty was quoted yesterday regarding the regular season title: ďI value it very highly because itís the same for everybody.Ē WRONG!!!! Sorry, Monty, youíre living in the past. With the new Pac-12, it ISNíT the same for everybody. It isn't like the Metro Atlantic, where Iona's regular season title by 2 games over Loyola (Maryland) in a true double round robin was in a schedule that was the same for everyone (which, combined with a 40 RPI got Iona in the tourney).
Letís say that the Pac-12 had played a true home and home round robin. UW would have had to add the toughest road trip in the conference, Cal and Stanford, plus adding a fairly easy home set, Utah and Colorado. Assuming they get swept in the Bay and sweep at home, they go 2-2. Cal would have had to add the Washingtons at home and a road trip to Arizona. Assuming they sweep at home, lose to Arizona on the road but beat crappy ASU on the road, they go 3-1. So if the favorites win all the games, we now have a tie for the conference title at 16-6 between Cal and UW. Should UW somehow get an advantage over Cal because it got an easier schedule than Cal? Sure, UW gets the Pac-12 championship, but that is the Pac-12ís decision to make. The NCAA committee cares about who the best teams are, and did UW really have a better regular season than Cal, because it played an easier schedule? The committee says no.
UW had zero top 50 wins. They were only 4-3 against RPI 51-100 (as compared to Calís 8-4), so the total record top 100 for UW was 4-8, as opposed to Calís 8-7.
RPI does matter in picking at large teams. Typically, the top 40-45 are in, period, barring a totally thin profile (and Calís lack of top 50 wins put Cal pretty close to the edge, but fortunately, with 68 teams in now, the line is closer to 45 than 40). Typically, sub-60 teams are out barring a really strong profile. Since the RPI formula was changed in 2005 to weight road and home victories, 63 has been the worst RPI to get in (before the formula was changed, in 2004 a 70 got in and in 1999 a 74 got in, the only 70ís or worse to get in).
This year was not pretty typical in terms of how RPI mattered, no real outliers. The best RPI left out of an at large bid was Marshall at 47, a team not in anybodyís bubble lists at the end (and not in the top six left out). The worst RPI to get an at large bid was 57 West Virginia . The first six left out had RPIís ranging from 51 to 68.
UW had an RPI outside of the range that, especially since the tweak in the formula in 2005, has any chance of getting in. Even if they had a chance, they didnít have a profile to get in. Zero top 50ís? Sorry, no chance with an RPI of 71. Conference title? Carries little weight, especially when it is built on having an easier schedule than the team that finished one spot behind.
If UW had gotten in, the committee would have been throwing everything it says out the window, and saying that winning an outright regular season conference championship in a Big 6 conference by virtue of playing an unfair schedule matters more than everything the committee has always talked about.
I might agree that the rules are screwed up. I want a true double round robin where the winner gets the conferenceís automatic bid, and no conference tourney (in which case, maybe Cal ties for the conference title this year and gets an automatic bid by virtue of sweeping UW). But for money reasons, it will never happen again. By the rules that exist, UW didnít get screwed.
And I donít think Cal got screwed, either.
GO BEARS!!! BEAT THE BULLS!!
Lengthy but a good read putting in the numbers helps. In truth, not surprised where Cal wound up, wouldn't have been surprised if we had been left out either. No quality wins hurts and blow-out losses don't help perception either. USF won't be an easy game either. Going cross-country on short notice the wrong way is never an easy trip to begin with.
True Blue Golden Bear
I think we maybe got slightly screwed, in that the committee overvalued the very recent performances of some other bubble teams (like Colorado and NC State) and pushed us too far down. But it wasn't a huge adjustment; I think our record merited a 10 or 11 and we got a 12. It's not like it was some great injustice.
Washington didn't get screwed at all. They had no Top-50 wins and a lousy record against the Top 100. They just didn't have enough good wins and needed a deeper run in the P12 Tournament to get their bubble case to a reasonable level. Losing to Oregon State killed them.
Consistency and cold hard logic .... works every time
Nice write up ..... I wonder what the mangy mutts would think about this ..... on second thought, who really cares????
Nice analysis. My take isn't as detailed but simply Cal got what they deserved as a bubble team, i.e., didn't win the tourney, didn't win the conference, suspect conference.
Frankly UW might have gotten screwed but you know what, I'm 100% okay with that because something else was at work - Curse of Tosh.
As I indicated, getting pushed behind Colorado probably didn't matter, because we still would have been in the last 4 at larges, matched against USF, needing to get the same seed as USF, and most likely being bumped down to a 12 (BYU got it bad, bumped from a 12 to a 14 along with Iona bumped from a 13 to a 14, mostly to avoid making them travel too far for the second round).
Originally Posted by sycasey
If we were ahead of NC St., however, we avoid a First Four game. These days, recent performance means less in selection than it used to, but it still pretty meaningful in seeding. Cal losing 3 of the last 4 didn't likely help. Even if you think there is one at large team ahead of Cal like a NC St. that should have been behind Cal, it is hard to say we were screwed, especially if it isn't a clear cut case. I think NC St. is the only at large team ahead of Cal you can really argue could have replaced Cal in the First Four, but I think you can argue either way with NC St. Cal has the better RPI but the crappier finish and NC St has slightly better best wins, it isn't clear enough to me to say there was a screw whichever way the committee put them.
To me, it is more clear that Texas should have been seeded ahead of NC St. instead of one spot behind, but since both got 11 seeds, it is hard to say Texas got screwed.
True Blue Golden Bear
Well, this is why I said we got maybe slightly screwed, not hugely. However, there are a few things I think the committee overvalues:
Originally Posted by Cal8285
1. Recent performance (only way to explain how NC State or Colorado get higher seeds than us).
2. Wins against Top 50 rather than RECORD against Top 50.
Cal had no wins against the Top 50, but on the other hand had only 3 chances to play against Top 50 teams, and NONE at home. Our above-.500 record against the Top 100 as a whole indicates that we probably could have picked up a win or two if we'd gotten some chances at Haas. Texas had 4 Top-50 wins, but all were at home, and it was against 8 losses. Is that really more impressive? The committee seems to think so, since it vaulted Texas ahead of Cal despite Cal having a 36-to-51 difference in overall RPI. West Virginia is a similar case. They had 4 wins against Top 50 teams (2 at home, 2 neutral court), but it was also against 8 losses. So they get seeded higher than Cal despite only a 19-11 record and an RPI number of 57. Virginia had only 2 wins against Top 50 teams (and 4 losses) and an RPI number of 53.
I can make an argument that Cal should have been seeded better than any of those other teams (much higher RPI, better raw W-L record, not significantly worse performance against Top 100, and lack of Top 50 for Cal was just an opportunity problem). Getting seeded higher than any one of them would have gotten us out of the First Four, so I think we were slightly screwed.
I think there is a fair argument that the committee over/under values certain things in seeding, and to that extent, you can argue Cal got screwed. Once we take as a given the screwy way the committee does things, then I think it is hard to argue Cal was very screwed. I.e., if Cal was screwed, it was screwed by the system, not by some badly screwed up misapplication of the system.
Originally Posted by sycasey
True Blue Golden Bear
I think it's more that the criteria are not always applied consistently. Cal had a strong RPI number and overall record; these are supposed to be important criteria as well. Yet we got seeded behind a whole bunch of teams with worse overall records and much worse RPIs because they each had a couple of Top 50 wins (and if you want to talk recent performance, Texas, WVU, and Virginia didn't finish particularly strongly either). Really? Cal wasn't better than any of them? Not sure I buy that.
Originally Posted by Cal8285
But as I've noted we're talking the difference between a 12 and an 11 (and maaaaybe a 10 if you stretch), so it's not like we were horribly underseeded.
My main critique of the NCAA (not this great analysis) is that Cal, through absolutely no fault of its own, had only THREE chances to PLAY top 50 RPI teams. You can't look at the OCC schedule and say "well my, Monty has developed an appetite for cupcakes"). Georgia has been a quality opponent in past years. GW has played in the past quality BB. Santa Barbara is a fairly decent program. Since these are set a couple of years in advance it is a total roll of dice. And we know from the KU home-and-home that "power teams" have limited interest in coming west - both because of the travel and because of the smallish size of Haas. I would agree with Monty - not worth it if that means we have to move the game to Oakland.
So what you are penalizing the Bears for are THREE OOC losses to A) A team that got totally screwed out of a number 1 seed - and arguably should be the overall number one in the tournament; B) Road losses (one close) to the number 1 and 2 in the MWC - which are seeded as 6s. Yes. If we want to be a contender and not a pretender we need to blow SDSU out and stay competitive with UNLV and Mizzu. We stink, suck, and deseve the bed we made. But wouldn't it have been NICE if Washington hadn't stunk it up against South Dakota State or if UCLA could have f'ing beat Loyola. What you are doing is penalizing the Bears for missteps of others and not even TRYING to apply the eyeball test.
To see how truly strange the math gets look at Michigan State - who has the SAME record in league as the Bears, lost to Duke and NC to open the season and then had the "fortune" to play minor conference teams like Leigh (who knew!) that did well. They then get into conference and because the Big-10 (in large part because of geography) gets some OOC against the OVC, A10, CUSA their RPI is nice and, presto, good chances for top 50 games.
Now I am NOT arguing that Cal deserved a 1. MSU won its conference tournie and we got beat by an ordinary Buffs team. But I think the stress on the "wins against top 50" RPI is a screwy measure that really penalizes conferences that have YOUTH AND the west coast in particular since there is ONE mid-major (MWC) in our geography. And note that the MWC is imploding and thus we are likely to find itEVEN harder to get OOCs that yeild high RPIs in the future.
My bigger concern (and the call out for Cal8285 to do some more math) is that this is starting to be a trend. I went back and did some work on trying to figure out the highest Pac-12 seeds in recent years.
2012 - 11
2011 - 5
2010 - 8
2009 - a 4
2008 - UCLA (a 1)
You have to go back FIVE years to 2007 when the conference go a 2 and two 3s and 6 of 10 teams total before we got the kind of haul that the Big-10 and ACC do on a regular basis
I am of the opinion that Scott needs to address this. The math is pretty compelling that given the importance of the computer and the way the committee uses RPI we HAVE to get our OOC strength better and THAT means finding a way to get Ohio Valley, Conference USA and A10 teams to travel west. We can NOT rely upon beating up SDSU, UNLV, and New Mexico because, unfortunately, those are no longer guaranteed Ws.
Finally, it is telling that the 2007 season was a year that UCLA won at Maui - which I sure immensely helped the RPI for the conference. But interesting you see the same pattern in quick glances at the OOC for the rest of the conference that year - very few A10, CUSA, or eastern "mid majors". Instead SEC schools, lower tierd ACC programs like Virginia and then bottom 100 programs.
We need to start scheduling the OOC like Temple or Zaga - and get EVERYONE ELSE in the conference to do it as well. Would help tremendously and likely make for better conference BB.
Not going to argue that Pac-12 needs Kansas on the schedule every year (and actually they played UCLA and DUKE played Washington). But we have to get the OOC schedule better if we want to break the awful string of poor seeds that the conference currently is battling.
Finally, this could also be addressed by getting 2 games with a power conference (SEC most obvious candidate) working in probably an additional deal for football. Would benefit both conferences and likely help both schools long term RPI issues. For that to work, however, SEC would likely fight Kentucky, who is pleased as peach with the way things are set up.
Last edited by socaltownie; 03-12-2012 at 12:46 PM.
It is a problem when the conference as a whole sucks, and you have limited opportunities to play top teams.
Originally Posted by socaltownie
Over the years, the committee has made it pretty clear, if your own conference doesn't give you the opportunities to beat good teams, then you need to dominate your conference, and you need to avoid to losing to crappy teams both in and out of conference.
The RPI ranking of the Pac-10 conference from 1994 through 2011 for non-conference games was never worse than 8, and until the last three years, almost always in the top 6. The inaugural Pac-12 season brought a non-conference conference ranking of 11, far and away the worst year in the RPI era for the Pac-10/12.
If the conference is going to have a season like that, then teams that want decent seedings need to really take advantage of every opportunity that comes along. You can't lose to SDSU by a point (could have used Solomon). You can't lose to sub-100 RPI teams even in road conference games. You can't lose to inferior teams in the conference tourney. You can't lose to anybody at home (even if it was one of the worst officiated games I can remember).
If we'd beaten SDSU, gone 16-2 in conference, avoiding losing to WSU and OSU, and lost in the conference tourney finals, we'd have a whole lot better seed. Maybe not high enough, and maybe we'd have been punished for not having enough chances against top 50 teams, and I might feel like the system screwed us. As it is, I feel like we didn't play well enough to merit better than we got.
Agree with your logic 100%. The problem is, in general, that the criteria are all fuzzy (except for RPI, which is a number, albeit flawed) and they are inconsistently applied. For some teams, recent results count the heaviest. Others, strength of schedule. Others overall record; this is especially problematic when you have a number of schools with similar resumes, but different strengths/weaknesses in terms of criteria. Cal v. Colorado, for instance. CU didn't collapse down the stretch as we did; OTOH, they have a much poorer road record (supposedly a crucial criterion) and a poorer RPI. How do you weight those, and why weight them differently in different instances?
I agree, there is nothing clear about how different factors get weighted, and there ends up being inconsistent application. For instance, to me, it is pretty clear, based on the supposed criteria, that Creighton should have gotten better than an 8 seed. Don't know how they ended up there.
Originally Posted by UrsaMajor
Colorado is the one team that kind of bugs me, even if I can understand justification for putting them ahead of us (I wouldn't have them ahead of us). I wouldn't weight the strong finish as much as the committee did, I'd probably have them as a 13 seed. But since we'd still be in a First Four game even if we flipped ahead of Colorado, and it is quite likely that we'd be a 12 seed anyway even if we were a 44 seed instead of 45 on the list of 68, I have trouble being bothered by it.
Once again there are a few things that to me are real headscratchers, not based on KenPom or Sagarin or anything the committee doesn't use, but on committee criteria. But to me, Cal playing as a 12 seed in a First Four game isn't one of them.
To have fun - is it hard to figure out Pac-12's OOC strength if UCLA and UTAH are excluded? Remember, UTAH played ZERO power conference opponents in the preseason and the Ruins, of course, stunk up horrifically against sub tier schools like Loyola.
Originally Posted by Cal8285
BTW - the main problem that I have with the hypothetical is that it isn't clear that another 5 wins in conference (16-2 in conference, 2 wins at LA) really gets Cal off the snide. Maybe a 5. Perhaps a 6. Lot better than a 12.5 but again, people would be saying no top 50 wins.
Well, what about...?
Sorry, I must admit to having just skimmed your post and I usually like to read them thoroughly...
In other words, you may have addressed this, but...
The reason I think Cal got "somewhat screwed" is that they have to do a play-in, even though they are seeded higher than several teams that don't. At large, conference champion, who cares? Why is this not getting screwed?