Announcement

Collapse
No announcement yet.

2019 NCAA Tournament Thread - Regionals are the best weekend of hockey all year

Collapse
X
Collapse
 
  • Page of 44
  • Filter
  • Time
  • Show
Clear All
new posts
Previous 1 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 template Next
  • #571
    Re: 2019 NCAA Tournament Thread - Regionals are the best weekend of hockey all year

    I agree that strength of schedule should be considered, but in our sport I think it gets too much consideration.
    MTU: Three time NCAA champions.

    It never get's easier, you just go faster. -Greg Lemond

    Comment

    • #572
      Re: 2019 NCAA Tournament Thread - Regionals are the best weekend of hockey all year

      Originally posted by manurespreader View Post
      I agree that strength of schedule should be considered, but in our sport I think it gets too much consideration.
      Not picking a fight here, but how else can teams be rated comparatively, without some objective SOS metric involved?

      Win % and SOS are really the only two salient stats involved, IMHO.
      Last edited by Fishman'81; 04-04-2019, 08:05 PM.

      Comment

      • #573
        Re: 2019 NCAA Tournament Thread - Regionals are the best weekend of hockey all year

        Originally posted by manurespreader View Post
        I agree that strength of schedule should be considered, but in our sport I think it gets too much consideration.
        MSUM earned a #1 seed with the 40th rated SOS, UMASS the same with the 23rd. Cornell and Harvard earned at-large berths with the 19th and 22nd rated SOS's respectively, Providence the 26th and Bowling Green with the 46th.

        Comment

        • #574
          Re: 2019 NCAA Tournament Thread - Regionals are the best weekend of hockey all year

          I am copying this piece data here, as well as to a connected discussion on the Minnesota thread:

          I'm not sure how to do the analysis of this data, but in regard to regionals, starting in 2006 and not including 2019,
          #1s are 31-21 in round one
          #2s are 27-25, with slightly better records in the last 7 years than before then.

          This set of numbers seems to make sense in a sort of a way.....There should be little difference between 2s and 3s, given that it's often true in the final PWR that one game result different (which isn't much - just one fluky goal) can make the difference in seeding a 2 or a 3.
          More parity, since 1s are not dominant, for sure. But the #1s still have a considerable edge.

          So, in a vacuum, these seem like some what representative results.

          However, what is interesting, but difficult to do scientifically, is something like this:
          In games where one school had a significant travel advantage (mostly the 2 eastern regionals, for obvious reasons, but sometimes in the west as well)....and significant is difficult to define, the 'home teams', regardless of seeding, are running about 60% winners.

          Take that for what you will in the idea of 'what's a good system'....

          As for SOS....The best metric if all you want to count is game results, is KRACH adjusted for home ice.
          The current PWR is close to that. Only replace ASU with PSU to get the usual top 15, and I am looking at post-regional results in some case, so that might affect ASU.

          Comment

          • #575
            Re: 2019 NCAA Tournament Thread - Regionals are the best weekend of hockey all year

            Originally posted by Fishman'81 View Post
            Not picking a fight here, but how else can teams be rated comparatively, without some objective SOS metric involved?

            Win % and SOS are really the only two salient stats involved, IMHO.
            Margin of victory could be appropriate as well. It’s used in several advanced rating systems for football and basketball.
            Code:
            As of 9/21/10:         As of 9/13/10:
College Hockey 6       College Football 0
BTHC 4                 WCHA FC:  1
            Originally posted by SanTropez
            May your paint thinner run dry and the fleas of a thousand camels infest your dead deer.
            Originally posted by bigblue_dl
            I don't even know how to classify magic vagina smoke babies..
            Originally posted by Kepler
            When the giraffes start building radio telescopes they can join too.
            He's probably going to be a superstar but that man has more baggage than North West

            Comment

            • #576
              Re: 2019 NCAA Tournament Thread - Regionals are the best weekend of hockey all year

              Also, when using ratings systems, you need to understand if they are supposed to be predictive or not. Some systems are designed to pick winners of future hypothetical games. Other (most?) are designed to pick the teams that performed the best over the season.
              Code:
              As of 9/21/10:         As of 9/13/10:
College Hockey 6       College Football 0
BTHC 4                 WCHA FC:  1
              Originally posted by SanTropez
              May your paint thinner run dry and the fleas of a thousand camels infest your dead deer.
              Originally posted by bigblue_dl
              I don't even know how to classify magic vagina smoke babies..
              Originally posted by Kepler
              When the giraffes start building radio telescopes they can join too.
              He's probably going to be a superstar but that man has more baggage than North West

              Comment

              • #577
                Re: 2019 NCAA Tournament Thread - Regionals are the best weekend of hockey all year

                Originally posted by dxmnkd316 View Post
                Also, when using ratings systems, you need to understand if they are supposed to be predictive or not. Some systems are designed to pick winners of future hypothetical games. Other (most?) are designed to pick the teams that performed the best over the season.
                Predictive systems, which are interesting to the user of course, seem inappropriate for an NCAA tournament. It seems much better to reward the teams for their performance already done. Of course, this is a point given to argument.

                What is more interesting in this idea:
                In the early days, the NCAA tournament was a sort of bonus. Most teams only played schedules of teams close to them. So, the tournament was a way to match regions of the country against each other as well.
                It has changed, of course.

                Comment

                • #578
                  Re: 2019 NCAA Tournament Thread - Regionals are the best weekend of hockey all year

                  Originally posted by Numbers View Post
                  Predictive systems, which are interesting to the user of course, seem inappropriate for an NCAA tournament. It seems much better to reward the teams for their performance already done. Of course, this is a point given to argument.
                  100% agree.
                  Code:
                  As of 9/21/10:         As of 9/13/10:
College Hockey 6       College Football 0
BTHC 4                 WCHA FC:  1
                  Originally posted by SanTropez
                  May your paint thinner run dry and the fleas of a thousand camels infest your dead deer.
                  Originally posted by bigblue_dl
                  I don't even know how to classify magic vagina smoke babies..
                  Originally posted by Kepler
                  When the giraffes start building radio telescopes they can join too.
                  He's probably going to be a superstar but that man has more baggage than North West

                  Comment

                  • #579
                    Re: 2019 NCAA Tournament Thread - Regionals are the best weekend of hockey all year

                    Originally posted by dxmnkd316 View Post
                    Margin of victory could be appropriate as well. It’s used in several advanced rating systems for football and basketball.
                    Margin of victory is worthless in hockey. It may have some value in basketball, because of the larger sample size of points scored. I don't see how you can definitively determine anything by analyzing MOV in hockey. What? Does a 5-2 win "mean" more than a 6-1 win? Everyone knows that you can win 7-0 one night and lose the next. It's not a sport where enough "points" are scored to rely on a margin of victory stat.

                    And the only thing that irritates me more are those stupid "total goals" series that we used to have. The object of the game is to WIN. I don't care if it's 1-0 or 25-2. There is NO DIFFERENCE. And I don't even like total goals being used as a tiebreaker for some of these tournaments (like World Juniors, for example).
                    Last edited by chickod; 04-05-2019, 10:32 AM.

                    Comment

                    • #580
                      Re: 2019 NCAA Tournament Thread - Regionals are the best weekend of hockey all year

                      Originally posted by chickod View Post
                      Margin of victory is worthless in hockey. It may have some value in basketball, because of the larger sample size of points scored. I don't see how you can definitively determine anything by analyzing MOV in hockey. What? Does a 5-2 win "mean" more than a 6-1 win? Everyone knows that you can win 7-0 one night and lose the next. It's not a sport where enough "points" are scored to rely on a margin of victory stat.

                      And the only thing that irritates me more are those stupid "total goals" series that we used to have. The object of the game is to WIN. I don't care if it's 1-0 or 25-2. There is NO DIFFERENCE. And I don't even like total goals being used as a tiebreaker for some of these tournaments (like World Juniors, for example).
                      The only reason they ever used total goals was to justify a two-game series and not be embarrassed by playing a mini-game after a team wins 8-0 and loses 3-4. The fact that both these rationale were stupid helps explain why that loathsome format is no longer used.
                      Cornell University
                      National Champion 1967, 1970
                      ECAC Champion 1967, 1968, 1969, 1970, 1973, 1980, 1986, 1996, 1997, 2003, 2005, 2010
                      Ivy League Champion 1966, 1967, 1968, 1969, 1970, 1971, 1972, 1973, 1977, 1978, 1983, 1984, 1985, 1996, 1997, 2002, 2003, 2004, 2005, 2012, 2014, 2018, 2019, 2020

                      Comment

                      • #581
                        Re: 2019 NCAA Tournament Thread - Regionals are the best weekend of hockey all year

                        In addition, the empty net goals are pretty much unique to hockey. There's a huge difference between a 5-2 game in which a team leads throughout and a game that 3-2 after 58 minutes and ends 5-2.

                        Comment

                        • #582
                          Re: 2019 NCAA Tournament Thread - Regionals are the best weekend of hockey all year

                          Originally posted by Numbers View Post
                          Predictive systems, which are interesting to the user of course, seem inappropriate for an NCAA tournament. It seems much better to reward the teams for their performance already done. Of course, this is a point given to argument.
                          Two sides of the same coin..? The predictions would have to based on (the season's) performance.

                          Comment

                          • #583
                            Re: 2019 NCAA Tournament Thread - Regionals are the best weekend of hockey all year

                            Originally posted by chickod View Post
                            Margin of victory is worthless in hockey. It may have some value in basketball, because of the larger sample size of points scored. I don't see how you can definitively determine anything by analyzing MOV in hockey. What? Does a 5-2 win "mean" more than a 6-1 win? Everyone knows that you can win 7-0 one night and lose the next. It's not a sport where enough "points" are scored to rely on a margin of victory stat.

                            And the only thing that irritates me more are those stupid "total goals" series that we used to have. The object of the game is to WIN. I don't care if it's 1-0 or 25-2. There is NO DIFFERENCE. And I don't even like total goals being used as a tiebreaker for some of these tournaments (like World Juniors, for example).
                            I think your hatred for total goal series, which were stupid though I was never around for them, is clouding your judgment on the use of MOV. In basketball, teams that are up get fouled and shoot free throws. How does that not disqualify it based on your criteria. Those points are essentially ENGs, hell, it’s in the name, ‘free’.

                            And to answer your question, yes, if A beats B 5-0, it’s objectively better than C beating B 1-0.
                            Code:
                            As of 9/21/10:         As of 9/13/10:
College Hockey 6       College Football 0
BTHC 4                 WCHA FC:  1
                            Originally posted by SanTropez
                            May your paint thinner run dry and the fleas of a thousand camels infest your dead deer.
                            Originally posted by bigblue_dl
                            I don't even know how to classify magic vagina smoke babies..
                            Originally posted by Kepler
                            When the giraffes start building radio telescopes they can join too.
                            He's probably going to be a superstar but that man has more baggage than North West

                            Comment

                            • #584
                              Re: 2019 NCAA Tournament Thread - Regionals are the best weekend of hockey all year

                              Originally posted by CLS View Post
                              In addition, the empty net goals are pretty much unique to hockey. There's a huge difference between a 5-2 game in which a team leads throughout and a game that 3-2 after 58 minutes and ends 5-2.
                              no.... pretty much each is a W
                              a legend and an out of work bum look a lot alike, daddy.

                              Comment

                              • #585
                                Re: 2019 NCAA Tournament Thread - Regionals are the best weekend of hockey all year

                                I understand that there was A dude out in vegas who had AIC over St. Cloud and a couple of other Minnesota teams winning for that same night all of whom won. His 10 dollar bet turned into 26,000 some odd.

                                Nice.
                                MTU: Three time NCAA champions.

                                It never get's easier, you just go faster. -Greg Lemond

                                Comment

                                Previous 1 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 template Next
                                Working...
                                X