Announcement

Collapse
No announcement yet.

RPI 2013/14: Sunny, with a slight chance of drivel

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

  • Re: RPI 2013/14: Sunny, with a slight chance of drivel

    Originally posted by JCrawford View Post
    Where are you coming up with these numbers? Depending on where I look (collegehockeynews.com / rpihockey.net) RPI is currently either 12th or 13th in PWR after yesterday's games. Even the RPI table on this site (http://www.uscho.com/rankings/rpi/d-i-men/) has Rensselaer at 12th in RPI.
    I am using the RPI and PWR calculator at wa2aea.com . I have found through USCHO's calculations that they are not correctly doing the weighting for win percentage, and neither site is calculating the Quality Wins Bonus that is new for this year, as described by several notes seen in the summer. It is true that RPI is high in RatingsPI prior to the Quality Wins Bonus, but are brought down immensely because they have not beat any quality opponents yet this year, while the several teams that have passed RPI in that calculation have either won or tied games against "quality opponents". For specific examples, I'm willing to discuss it on the several RatingsPI and PWR threads, or we can discuss via private message.

    Comment


    • Re: RPI 2013/14: Sunny, with a slight chance of drivel

      Originally posted by FlagDUDE08
      Don't believe me? Do the calculations out yourself. I actually e-mailed Chris Lerch about these phenomena, and he did mention to me that they will not adjust their calculations until January.
      I'll continue to remain skeptical until I start seeing others who have been doing this for years come up with the same numbers. Sorry if I'm skeptical of the "it's this because I say it is" line of argument.

      Comment


      • Re: RPI 2013/14: Sunny, with a slight chance of drivel

        Originally posted by FlagDUDE08 View Post
        I am using the RPI and PWR calculator at wa2aea.com . I have found through USCHO's calculations that they are not correctly doing the weighting for win percentage, and neither site is calculating the Quality Wins Bonus that is new for this year, as described by several notes seen in the summer. It is true that RPI is high in RatingsPI prior to the Quality Wins Bonus, but are brought down immensely because they have not beat any quality opponents yet this year, while the several teams that have passed RPI in that calculation have either won or tied games against "quality opponents". For specific examples, I'm willing to discuss it on the several RatingsPI and PWR threads, or we can discuss via private message.
        FlagDUDE, I'm trying to get my head around PairWise. What's the definition of a "quality opponent?"
        Let's give a cheer for old Rensselaer... You bring the whiskey, I'll bring the beer!

        Comment


        • Re: RPI 2013/14: Sunny, with a slight chance of drivel

          I just watched the Colgate game on RPI tv, very good coverage and I thought pretty much non biased. Very strange game, thought two of the majors were very questionable.
          It all starts with the goaltending.

          Comment


          • Re: RPI 2013/14: Sunny, with a slight chance of drivel

            Originally posted by FlagDUDE08
            Don't believe me? Do the calculations out yourself. I actually e-mailed Chris Lerch about these phenomena, and he did mention to me that they will not adjust their calculations until January.
            It'd be nice to view the source to see how exactly the calculations are done in the WA2AEA calculator, but the software publisher seems to be violating the GPL:

            Convey the object code by offering access from a designated
            place (gratis or for a charge), and offer equivalent access to the
            Corresponding Source in the same way through the same place at no
            further charge.
            The terms stating "These applications are produced under the GNU General Public License. Downloads of the applications are free, but any source code requests require a $5 donation per application, per request." seem to violate the "equivalent access" clause of the license terms. If the software is available as a free download, then so must the source code. http://www.gnu.org/licenses/gpl-faq....lowDownloadFee

            Comment


            • Re: RPI 2013/14: Sunny, with a slight chance of drivel

              Originally posted by JCrawford View Post
              I'll continue to remain skeptical until I start seeing others who have been doing this for years come up with the same numbers. Sorry if I'm skeptical of the "it's this because I say it is" line of argument.
              Not surprising given our past quarrels, but understandable given this is the first year that college hockey has used both home/away weighting and the quality wins bonus in its calculations.

              USCHO uses 1.2 if you are away and 0.8 if you are at home, regardless of game outcome. However, many other sources that I have found, most of them being for basketball but still relevant because they, to quote you, "have been doing this for years", show that the higher weight is applied to a road win or home loss, while the lower weight goes to a home win or road loss. This makes sense because given the phenomena of sports, there is a slight bias in possible outcomes that the home team is expected to win. Basketball may use a higher weighting (theirs is 1.4/0.6), but the method of calculation is still the same. The rest of the WP, OWP, and OOWP is correct, and I have used the sources you have previously cited to ensure that I am calculating those correctly.

              The Quality Wins Bonus is the factor that is really providing the difference that you are seeing in the weighting. From USCHO's article:

              Quality Wins Bonus
              The Quality Wins Bonus is implemented to reward performance against the strongest competition, as measured by the top 20 teams in the RPI (calculated per above).

              The maximum bonus of 5.00 points in any single game is earned with a win against the #1 team in the RPI. A bonus of 4.75 points is earned for a win against the #2 team, 4.50 is earned for a win against the #3 team, and so on until the smallest bonus of 0.25 points, which is earned for a win against the #20 team.

              As a tie is one-half of a win, the bonus for a tie against any top 20 team is one-half of the corresponding win bonus against the same team.

              Read more: http://www.uscho.com/2013/09/20/road...#ixzz2kMAnDlMT
              RPI is 14th prior to this Quality Wins Bonus, but has yet to defeat or tie any teams that are in the top 20, so they don't get any bonus points. I list the top 20 teams in the "NCAA Selection Process" thread.

              Comment


              • Re: RPI 2013/14: Sunny, with a slight chance of drivel

                Originally posted by RPI_2003 View Post
                FlagDUDE, I'm trying to get my head around PairWise. What's the definition of a "quality opponent?"
                A "quality opponent", as defined by the selection criteria, is a team that is in the Top 20 of RatingsPI prior to any calculation of quality wins bonus. BU is actually 28th in this calculation, so they are not yet a quality win for us.

                Comment


                • Re: RPI 2013/14: Sunny, with a slight chance of drivel

                  We remain at #10 in the USCHO poll http://www.uscho.com/rankings/.
                  sigpic

                  Let's Go 'Tute!

                  Maxed out at 2,147,483,647 at 10:00 AM EDT 9/17/07.

                  2012 Poser Of The Year

                  Comment


                  • Re: RPI 2013/14: Sunny, with a slight chance of drivel

                    Originally posted by Ralph Baer View Post
                    We remain at #10 in the USCHO poll http://www.uscho.com/rankings/.
                    Zoiks Scoob, look who's #11....yeah Shaggy but THEY STILL SUCK......
                    Brandon Pirri: Turk would give it to Jagr. That's awesome! Jagr's untouchable. And he would give it to Jagr.

                    Comment


                    • Re: RPI 2013/14: Sunny, with a slight chance of drivel

                      Originally posted by turk181 View Post
                      Zoiks Scoob, look who's #11....yeah Shaggy but THEY STILL SUCK......
                      Gotta respect the fact that they were UNANIMOUSLY picked to finish last in the ECAC by everyone whose opinion matters, and yet here they are.
                      Hockey is serious business. Everything else is just a game.

                      RPI Hockey, Syracuse Football, New York Rangers, New York Giants.

                      Comment


                      • Re: RPI 2013/14: Sunny, with a slight chance of drivel

                        Originally posted by AspyDad View Post
                        Well put indictment of Paul Stewart. The bias that I have talked about exists.
                        ......"Stewy, you've made one lousy call after another tonight and I'm sick of you trying to stick it to my players," ....http://www.hockeybuzz.com/blog/Paul-...5#.UoEXtSj_Tao
                        Brandon Pirri: Turk would give it to Jagr. That's awesome! Jagr's untouchable. And he would give it to Jagr.

                        Comment


                        • Re: RPI 2013/14: Sunny, with a slight chance of drivel

                          Originally posted by Ralph Baer View Post
                          We remain at #10 in the USCHO poll http://www.uscho.com/rankings/.
                          Best news of all - 6 ECAC teams in top 20 (and 3 more getting votes), not bad for a lousy conference.
                          Not a real doctor, nor do I play one on TV!
                          Clarkson 1990

                          Comment


                          • Re: RPI 2013/14: Sunny, with a slight chance of drivel

                            Originally posted by turk181 View Post
                            ......"Stewy, you've made one lousy call after another tonight and I'm sick of you trying to stick it to my players," ....http://www.hockeybuzz.com/blog/Paul-...5#.UoEXtSj_Tao
                            Stewy's own words shows that he has rabbit ears.
                            "Rick not want to be know as tough guy, Rick want to be known as team guy" Rick "Mongo" Bennett

                            Comment


                            • Re: RPI 2013/14: Sunny, with a slight chance of drivel

                              Originally posted by Ralph Baer View Post
                              We remain at #10 in the USCHO poll http://www.uscho.com/rankings/.
                              Not much to argue about with those numbers but I still feel that 2 teams with 4 losses each are ranked too highly(regardless of the toughness of their schedule) and why Harvard and Brown got votes is a bit of a mystery. I would have ranked QU higher(perhaps 4th) after watching the Yale game Saturday. But just still too early to make much out of this. Another couple weekends and we should have a much better idea of whether these polls mean anything.
                              Take the shortest distance to the puck and arrive in ill humor

                              Comment


                              • Re: RPI 2013/14: Sunny, with a slight chance of drivel

                                Originally posted by AspyDad View Post
                                Stewy's own words shows that he has rabbit ears.
                                Show me a coach or referee who doesn't and I'll show you an Obamacare web site that doesn't crash!!
                                I've been lost since 1989. I got a breakaway on a frozen river and have been skating ever since.

                                Comment

                                Working...
                                X