Well said. I wonder if Stollings will continue to schedule cream puffs moving forward in her new gig.
I suspect she will.
Why not put the blame on the past coaching staff instead of Charlie Creme? Everyone knows the committee uses RPI. Pitino talked about it a couple years ago when he built his schedule and they still went mostly off rpi, he put a bunch of solid mid majors but none they should ever lose to and they had a rpi around 25 at the end of the year and got a 5 seed. Its up to the coaching staffs to figure out what the selection committee uses when building their schedules and exploit it. Instead Stollings crew put together a schedule of 200+ rpi teams that do nothing of value except boost her coaching record. I'm with you that Minnesota is better than Indiana, but Minnesota now has a nonconference sos of 344 while Indiana's is 102. MInnesota's overall sos is 150 while Indiana's is 45th. SOS is the difference and the only person to blame is Stollings for putting that crap together.
Well said, indeed. Thanks for posting, whalenfan, since I agree with you 110%. And you stated the important points quite eloquently.
I actually kept disparaging remarks for Stollings at a minimum in my post - I suppose mainly because I think I've lambasted her sufficiently in prior posts, and I'm now taking a slightly different tack to the problem (see below). I'm glad you filled in and connected the dots that I left out. However, I guess I did actually call her an idiot. ["(Stollings) was absolutely idiotic in picking NC teams to play. They chose NC teams (minus Syracuse and a couple more) that were both puffy and injected full of cream, and creamy frosting to boot."]
I think we all (Lindsay included, see next paragraph) just feel frustrated by the fact that Stollings messed up the SoS, and that messed up the RPI, and there's nothing we can do about it. Except cheer for the team to win the Big Ten championship as well as the two remaining regular-season games as (at this point) practice for the Big Ten championship.
In the recently posted Whalen interview "Coach Whalen Previews Rutgers" in the "Minnesota visits Rutgers" thread, Lindsay answers a question regarding RPI (couldn't hear the question, they really need to mic up the interviewers in these videos). Her answer includes a comment on the order or "There's nothing we can do about the non-conference schedule. It was given to us." This doesn't tell us anything we didn't know, but it does show us that Lindsay is acutely aware of how we got a really raw deal in (using my own allegoric language from this point) this year's SoS/RPI poker game. In essence, Lindsay had to come in mid-poker-game and take over the seat + cards + stake from a prior poker player (e.g., coach Stollings) who didn't even know how to play the game of SoS/RPI poker. Stollings thought the SoS part of the RPI game was there so that she could boost her own win/loss resume as a coach. A totally selfish and self-promoting approach, with no regard for the team.
In reality, a team should play the SoS part of the SoS/RPI poker game in a manner that maximizes the probabilistically expected value of the RPI metric (and we are currently stuck with that metric, unfortunately) at end of season when the Committee is evaluating teams for potential invitations to the NCAA playoffs. Based on the cards that Stollings left us, we totally lost the SoS game this year (or a more accurate description would be that we are the SoS laughing-stock of the Big Ten).
Because the Gophers so badly lost the SoS part of the SoS/RPI poker game this year, we also badly lost the RPI part of the game as well. Unfortunately, neither the Committee nor Charlie Creme seem to look much beyond RPI. RPI is just some arbitrarily-defined type of poker chip that is valued at a linear combination of 0.75 times the number of SoS chips that you won + 0.25 times (roughly) the fraction of games that you won. It puts not enough emphasis on basketball performance, and horribly too much emphasis on who you were able to finagle to play. To make it understandable how bad RPI is as a metric, again using the poker analogy, you could say that although you need to be a pretty good basketball team just to get a decent seat at the poker table, the winners of the poker game are mostly determined by who you were lucky enough to talk into playing you, plus how smart you were in signing up to play teams that are bad enough for you to be able to beat, but good enough to boost your SoS.
In the Mens game they use a different type of arbitrarily-defined poker chips called NET that are equally bad for different reasons. One of the many faults of NET is the opposite of the RPI fault, namely that they place too much emphasis on how much you clobber the other teams by, on average, and not nearly enough (if any) emphasis on how strong were the teams that you clobbered. So playing cream-puffs will tend to jack up the Mens NET ranking.
The part that I think is horribly ironic is that if the Womens game had been using NET this year instead of RPI, then the Lady Gophers would be trumpeted as NCAA leaders - probabily coming in at a NET ranking of about #20 to #30 - and definitely In the playoffs. In the RPI poker game, playing a bunch of horrible teams will kill your SoS and thus kill your RPI. In the NET poker game, playing a bunch of horrible teams is largely a don't-care for NET in terms of how the horrible-ness of your opponents per se impacts NET, but it does inflate your average margin of victory (which is what NET focuses on), and thus boosts your NET score. If next year the Women still have RPI and the Men still have NET, then the best interests of the Lady/Men Gophers would be served if both Whalen and Pitino had a little talk with the ADs of the worst schools that the Women played this year (such as Arkansas Pine Bluff and Coppin State and New Hampshire and Incarnate Word) and make a deal with them that our Women will not play them next year, but we'd be most happy to have our Men play them next year. Thereby the Women lose the SoS boat anchor around their neck that drags down their RPI ranking, whereas the Men gain the helium baloon around their neck that would inflate their NET ranking. If we are truly to do what is best in our own basketball-self-interest, then that's what we should do next year. [On the other hand, do we really want to stoop so low as to make Pitino and the Mens team play Arkansas Pine Bluff and Coppin State and New Hampshire and Incarnate Word, just to "game the NET systems"? Because every other team is gaming the system? And if you don't game the NET system too, then you might not get to go to the NCAA playofs since your NET score is not high enough. So that's one of my points here. Yes, we should perhaps properly game the system in any given year for whatever horrible system that we're given to game. But at the same time, I assert that the higher goal should be to totally revamp both the RPI and NET systems, so that they are useful and accurate systems for the cause they were intended, and there is no need to game them just because the systems themselves are so horribly designed.]
On the other hand (and I see this as unlikely), if the Women are changed to NET next year, then we should gladly renew our contracts to play Arkansas Pine Bluff, Coppin State, New Hampshire and Incarnate Word. Why? Because NET favors those teams who choose to play opponents who are so bad that they can rack up winning margins of 30 points (and there is little-or-no SoS punishment in NET that I know of, although I'm still learning about it). So if the rules change, then we should adapt to the new rules, right? [Or in the Womens case, if they change the rules from RPI poker to NET poker next year, then do the same thing that made us lose badly in the RPI game, because it will make us win big in the NET game. But again, having the Women play a bunch of stupidly bad teams next year, just because the game switched from RPI to NET, while maybe helping us to get to the tournament, there is no other sensical reason for doing so, and it doesn't help our team improve any, and I for one don't want to be the person that has to tell Lindsay that we need to keep on scheduling seamingly stupidly now, because they changed the rules of the game from one stupid RPI metric to another stupid NET metric. I really wouldn't feel comfortable telling either Pitino or Whalen that they ought to stoop so low.]
Meanwhile, I for one want to work hard to change the system so that both RPI and NET are replaced with something sensible. Something statistically sound and causally sound (that is, so that the cause of getting a good score is being a good team, not just winning some completely arbitrary poker game).
Back in Lindsay's interview, she finishes off by saying the right thing. She says, we're just going to continue to try to win games. Something that lots of us posters have said too. And she's right, it's all we can do, and it's what we want to do at this point. If we win games, we still might be able to enter the NCAA tournament via the back door of a Big Ten championship. If not, we'll be well prepped to do well in the WNIT.
Now jump from what the team must do to what I must do. I've learned a lot of technical details about the SoS/RPI game over the last year (and to a lesser extent, NET). And the whole thing just irks me. It's not fair to the teams, let alone it's statistically nonsensical. It's not fair to the teams who have all worked so hard (not just the Gophers, but every other team who got left out of the NCAA tournament unjustly). I claim we need a better system to replace both RPI and NET, both of which are just bankrupt. I mean, soon we will see Mens teams rushing to sign up to play Coppin State, for crying out loud.
Two of my approaches are as follows. One is to try to enlist important people for my cause. For example, Charlie Creme. Maybe I got off to a bad start on this one by calling him a joke. But I guess I'm hoping to convince him, or shame him, or whatever it takes to enlist his help in changing the system. If Charlie Creme would come out and say, "I know I've made a good living off this RPI thing, but I have to show my support for the growing movement to junk both RPI and NET, because it's the right thing to do. I can still make my living off the new replacement system, and feel better about it too." Charlie, if you're reading me, please help (and I'm sorry I called you a joke - I think what I really meant was that the current RPI/NET system is a giant joke, and poor Charlie if forced to live within this joke of a system - good gracious, I wonder how the poor guy sleeps at night!).
My other approach might be to write a technical paper on the whole RPI/NET mess. Not sure what statistics journal might be willing to publish it, tho, if I do write it. I mean, how many technical journals accept a paper for publication that start out 90% technical and end up 10% rant? If anyone knows such a journal, let me know.