Announcement

Collapse
No announcement yet.

2012 NCAA Tournament: Bracketology

Collapse
This topic is closed.
X
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Re: 2012 NCAA Tournament: Bracketology

    Originally posted by mnstate0fhockey View Post
    They're going to do whatever makes the most money. Being "fair" isn't their top concern.
    The "fair" was taken care of by requiring off-campus venues as far as NCAA is concerned.

    Comment


    • Re: 2012 NCAA Tournament: Bracketology

      Originally posted by Numbers View Post
      Did y'all see that Moy's latest masterpiece left Ferris in St Paul with Minnesota? I read it about an hour ago, and I can't remember the details. I think that after he took care of the interconference swaps (NoDak for OSU to avoid Ferris, and Miami for Denver to avoid UMD), he then swapped Den/BU for Miami/UMD for attendance at Green Bay and Bridgeport, and he swapped Maine for Michigan State for the sake of attendance at Worcester and St Paul (like St Paul needs any help.....).

      Leaving:
      St Paul : 4-FSU v 14-NoDak; 8-Minnesota v 10-Michigan State (total of 36 combined seedings)
      Green Bay : 2-Michigan v 15-Cornell; 5-UMD v 11-Miami (33)
      Bridgeport : 3-UML v 13-OSU; 6-BU v 12-Denver (34 the hard way)
      Worcester : 1-BC v 16-AHA Champ; 7-Union v 9-Maine (33)

      Comments??
      I assume he put Michigan in Ferris with accordance to the NCAA guideline
      No. 1 seeds are placed as close to home as possible in order of their ranking 1-4.
      The problem is that Michigan is getting on a plane no matter what so that rule no longer applies. That's why, for example, #2 overall Denver was placed in Albany rather than Fort Wayne or St. Paul in 2010.

      Comment


      • Re: 2012 NCAA Tournament: Bracketology

        Priceless, has someone ever come out and said "a flight is a flight"? I mean, I know what you're saying, but not every flight is the same distance (obviously). As someone else pointed out, it shouldn't be just about the team's travel. Taking into consideration overall fan travel should be a concern as well.

        Anyway, the way I view the current scenario is #1 BC goes to Worcester--obvious to everyone. Yes, #2 Michigan is closer to GB than St. Paul, but is it really necessary to sacrifice bracket integrity so that the lowest #1 seed (Ferris) is now paired with the lowest #2 seed (Minnesota)? I just don't think that it is, especially when you consider a pretty good regional can still be put together in GB without Michigan.

        Comment


        • Re: 2012 NCAA Tournament: Bracketology

          Originally posted by hattrick16 View Post
          My error, the word "hockey" is misplaced. "big institutions" in terms of money is what I was trying to say. Minnesota, Yale and Holy Cross have the financial resources to do something like this that many schools do not. I can't speak to Tech's finances.
          I think you're overestimating the financial exposure of the host schools. It's really more of a matter of legwork than cost (see http://www.ncaa.org/wps/wcm/connect/...9b9f5b09d3bfa5). Of course you can argue that schools with larger athletic departments can spare the manpower.

          I think that the reason that some of us are not overly excited by the current situation with regard to "hosting" advantages is that it used to be worse. Sometimes a higher seeded team had to play a lower seeded team on the lower seeded team's home rink, not just a venue that was more convenient for the lower seeded team and its fans. This happened more in the West, because there were more venues in the East that were the right size and were reasonably located.

          I personally don't agree that it would be better to let the higher seeded team host (though reasonable people differ on this one a lot; the reason that's usually given is attendance, not "fairness"). IMO, tolerating the relatively few times that it's very unfair to a higher seeded team is better than always giving a big advantage to a team based on a system as flawed as the PWR. Other than resorting to the "smoke filled room", we need some method for selecting who's in the tournament, but I think it would be a big mistake to award an advantage as big as home rink based on the PWR.

          Comment


          • Re: 2012 NCAA Tournament: Bracketology

            So, now I am thinking of the end of the year PWR. I have a few thoughts.

            1) I believe we will get to see what the committee does with this scenario about Ferris and Michigan. The reason is that the most likely ending of the CCHA playoffs would have Michigan and Ferris playing in the finals. If both of them end up as #1s, then the conference champ is going to be set into Green Bay initially -as per guidelines. Then, the 2nd part of the equation is that even though i would wish for better, I feel it likely that Minnesota will end up a low #2 or a high #3 - between 7 and 10 in the PWR. So, that sets up perfectly this situation #4 (Ferris or Michigan) being paired with the 7/10 or 8/9 game. Then what, committee members??? We can guess all we want, but the reality will come out of the discussions in that room.

            2) And, I posted this in the Minnesota season thread as well, I have developed a new thought about losses. It is often said that losing to a low RPI team kills you. My thought: not so much. I mean, losing is always bad for your own win%, so winning is preferable. However, take Minnesota as example. Early this year, they swept UMD, and then split with Vermont. What if they split with UMD, and swept Vermont instead? Well, their RPI is identical, no change. Their TUC record suffers, because they now have a loss to UMD, rather than a win. The Vermont loss hurts them in CommOpp, but only against HE teams, and it does hurt bad there. But, having a loss to UMD hurts against Denver, UND, ColoColl, etc. So, the CommOpp factor is likely a wash. So, my conclusion is: if you have to lose, lose to team that is not a TUC. Beat all the high ranking teams you can.

            Comments?

            Comment


            • Re: 2012 NCAA Tournament: Bracketology

              Originally posted by Numbers View Post
              So, now I am thinking of the end of the year PWR. I have a few thoughts.

              1) I believe we will get to see what the committee does with this scenario about Ferris and Michigan. The reason is that the most likely ending of the CCHA playoffs would have Michigan and Ferris playing in the finals. If both of them end up as #1s, then the conference champ is going to be set into Green Bay initially -as per guidelines. Then, the 2nd part of the equation is that even though i would wish for better, I feel it likely that Minnesota will end up a low #2 or a high #3 - between 7 and 10 in the PWR. So, that sets up perfectly this situation #4 (Ferris or Michigan) being paired with the 7/10 or 8/9 game. Then what, committee members??? We can guess all we want, but the reality will come out of the discussions in that room.

              2) And, I posted this in the Minnesota season thread as well, I have developed a new thought about losses. It is often said that losing to a low RPI team kills you. My thought: not so much. I mean, losing is always bad for your own win%, so winning is preferable. However, take Minnesota as example. Early this year, they swept UMD, and then split with Vermont. What if they split with UMD, and swept Vermont instead? Well, their RPI is identical, no change. Their TUC record suffers, because they now have a loss to UMD, rather than a win. The Vermont loss hurts them in CommOpp, but only against HE teams, and it does hurt bad there. But, having a loss to UMD hurts against Denver, UND, ColoColl, etc. So, the CommOpp factor is likely a wash. So, my conclusion is: if you have to lose, lose to team that is not a TUC. Beat all the high ranking teams you can.

              Comments?
              When I have Minnesota sweep Vermont and split with Duluth, their RPI goes from .5455 to .5464. That sounds like the loss to Vermont was worse than a loss to Duluth would have been.
              North Dakota
              National Champions: 1959, 1963, 1980, 1982, 1987, 1997, 2000, 2016

              Comment


              • Re: 2012 NCAA Tournament: Bracketology

                Maybe you can clarify, then FS23. I was under the impression that the 21% opponents win% and the 54 % opps' opps' win% were calculated while excluding the games played against the team whose RPI you are calculating. Is that true, or not?

                In other words, I was thinking that the 'strength of schedule' part of the RPI was not dependent on games that include "Minnesota" in this case. And, if that were the case, then the SOS part of the equation would not change in the scenario I described, correct?
                Last edited by Numbers; 02-22-2012, 06:39 PM.

                Comment


                • Re: 2012 NCAA Tournament: Bracketology

                  Originally posted by Numbers View Post
                  Maybe you can clarify, then FS23. I was under the impression that the 54% opponents win% and the 21 % opps' opps' win% were calculated while excluding the games played against the team whose RPI you are calculating. Is that true, or not?

                  In other words, I was thinking that the 'strength of schedule' part of the RPI was not dependent on games that include "Minnesota" in this case.
                  I don't know the RPI all that well, but if you beat Team B which has a .800 win % and a .400 opp win %, it would be better than beating Team C which has a .400 win % and a .800 opp win % because (.54 * .800) + (.21 * .400) > (.54 * .400) + (.21 *.800). (the actual numbers you get there are .516 > .384) Again, I'm not an RPI guru (just plugged in the different results using Whalen's calculator), but it makes sense that you are better off beating better teams...with the one caveat being that it is occasionally better to lose to a team on the cusp of being a TUC if you have beaten them many times in order to save that TUC record against that team (i.e. you are 5-0-0 against North Dakota, and a sixth win would knock them off from being a TUC. Unless you have an extremely high record aginast TUCs, you are better off losing in 6th game to keep North Dakota above .500 in the RPI, thus salvaging at least a 5-1-0 TUC record...assuming that the drop in your RPI would not lose you more comparisons than the 5-1-0 extra TUC record).
                  North Dakota
                  National Champions: 1959, 1963, 1980, 1982, 1987, 1997, 2000, 2016

                  Comment


                  • Re: 2012 NCAA Tournament: Bracketology

                    Originally posted by Numbers View Post
                    2) And, I posted this in the Minnesota season thread as well, I have developed a new thought about losses. It is often said that losing to a low RPI team kills you. My thought: not so much. I mean, losing is always bad for your own win%, so winning is preferable. However, take Minnesota as example. Early this year, they swept UMD, and then split with Vermont. What if they split with UMD, and swept Vermont instead? Well, their RPI is identical, no change. Their TUC record suffers, because they now have a loss to UMD, rather than a win. The Vermont loss hurts them in CommOpp, but only against HE teams, and it does hurt bad there. But, having a loss to UMD hurts against Denver, UND, ColoColl, etc. So, the CommOpp factor is likely a wash. So, my conclusion is: if you have to lose, lose to team that is not a TUC. Beat all the high ranking teams you can.
                    I thought I saw that had MN beat VT they'd presently be #1 or 2 in the PWR?

                    Comment


                    • Re: 2012 NCAA Tournament: Bracketology

                      Originally posted by Slap Shot View Post
                      I thought I saw that had MN beat VT they'd presently be #1 or 2 in the PWR?
                      I believe their conclusion was it was better to lose to Vermont (bad team) than Duluth (good team), but I don't believe that to be true.
                      North Dakota
                      National Champions: 1959, 1963, 1980, 1982, 1987, 1997, 2000, 2016

                      Comment


                      • Re: 2012 NCAA Tournament: Bracketology

                        Originally posted by Fighting Sioux 23 View Post
                        I don't know the RPI all that well, but if you beat Team B which has a .800 win % and a .400 opp win %, it would be better than beating Team C which has a .400 win % and a .800 opp win % because (.54 * .800) + (.21 * .400) > (.54 * .400) + (.21 *.800). (the actual numbers you get there are .516 > .384) Again, I'm not an RPI guru (just plugged in the different results using Whalen's calculator), but it makes sense that you are better off beating better teams...with the one caveat being that it is occasionally better to lose to a team on the cusp of being a TUC if you have beaten them many times in order to save that TUC record against that team (i.e. you are 5-0-0 against North Dakota, and a sixth win would knock them off from being a TUC. Unless you have an extremely high record aginast TUCs, you are better off losing in 6th game to keep North Dakota above .500 in the RPI, thus salvaging at least a 5-1-0 TUC record...assuming that the drop in your RPI would not lose you more comparisons than the 5-1-0 extra TUC record).
                        So, fs23, I tried Whelan's calculator, and got the same result you did - swap the Vermont loss and a UMD win, and the RPI rose by .0009. Looking closely at the RPI breakdown there, the Opponent's win% part of the calculation did not change. That means that UMD's win% and Vermont's win%, when calculating Minnesota's RPI, do not include the Minnesota games. What does change, however, is the opp's opp's win % part of the calculation. In this case, UMD's win% goes up (because we are giving them a win over Minnesota rather than a loss), and since Minnesota has played so many games against teams who have played UMD, the Opp's opp's win% for Minnesota goes up.

                        Wow. That gets really complicated. So, now I think I need to revise my idea. If I were Minnesota, and could choose one of these 2 (sweep versus AA, split versus BC) or (split versus AA, sweep versus BC), it would be better to choose the sweep with BC. Since AA has a win versus Minnesota, that comes back to the Gophers' RPI every time they play a conference game. And, the sweep against BC helps the TUC record, and the CommOpp record against eastern teams.

                        Comment


                        • Re: 2012 NCAA Tournament: Bracketology

                          Originally posted by Slap Shot View Post
                          I thought I saw that had MN beat VT they'd presently be #1 or 2 in the PWR?
                          That was an end of the year calculation from a Gopher fan, assuming Minnesota won out. Here, I am trading a win versus UMD for a win versus Vermont.

                          I hope that's clear.

                          Comment


                          • Re: 2012 NCAA Tournament: Bracketology

                            Originally posted by Numbers View Post
                            That was an end of the year calculation from a Gopher fan, assuming Minnesota won out. Here, I am trading a win versus UMD for a win versus Vermont.

                            I hope that's clear.
                            Yes.

                            Comment


                            • Re: 2012 NCAA Tournament: Bracketology

                              I was using the do it yourself calculator this morning, and found something very interesting to me - and this shows how delcate the PWR really is.

                              I just guessed at reasonable results for this weekend. What I found was that if UMD gets 3 pts this weekend, at least one scenario would move them all the way to #2 in the PWR, and it would be on the strength of TUC record. But, give them only a split, and they would be #6 in the same scenario.

                              Take that for what you want - flawed system (I agree), tightly bunched pack (I also agree). But, that is the way it is...

                              Comment


                              • Re: 2012 NCAA Tournament: Bracketology

                                Originally posted by Numbers View Post
                                I was using the do it yourself calculator this morning, and found something very interesting to me - and this shows how delcate the PWR really is.

                                I just guessed at reasonable results for this weekend. What I found was that if UMD gets 3 pts this weekend, at least one scenario would move them all the way to #2 in the PWR, and it would be on the strength of TUC record. But, give them only a split, and they would be #6 in the same scenario.

                                Take that for what you want - flawed system (I agree), tightly bunched pack (I also agree). But, that is the way it is...
                                The real flaw is in looking at PWR before all games are complete. If they end up #2 in PWR, its not because they "moved up". It was the only position they ever had, as PWR is meant to be calculated at the end.

                                Everything before that is just for speculation and entertainment purposes.
                                PSNetwork / XBOX GamerTag: xJeris
                                Steam Profile

                                Sports Allegiance
                                NFL: CHI; MLB: MN, NYM; NHL: MN, MTL; NCAAB: MN, UNLV; NCAAF: MN, MIA; NCAAH: MN; Soccer: USA, Blackburn

                                Comment

                                Working...
                                X