Page 2 of 8 FirstFirst 12345678 LastLast
Results 21 to 40 of 157

Thread: Fun With Numbers 2019: PWR, RPI, KRACH, GRaNT, WCHODR, & other assortments of letters

  1. #21
    Registered User
    Join Date
    Feb 2018
    Posts
    564

    Re: Fun With Numbers 2019: PWR, RPI, KRACH, GRaNT, WCHODR, & other assortments of let

    Quote Originally Posted by ChickHicks86 View Post
    As a quick check, I created an average team with a KRACH rating of 100 and plugged it into my spreadsheet. The RRWP numbers inch closer to the USCHO numbers, but they are still closer to my original predictions.
    My hypothesis is not that there is an additional dummy team used in the RRWP calculation. It's that there is a dummy team used by USCHO to calculate the ratings that you see on their KRACH page, and that you and I both used to calculate what the RRWP should be based upon those ratings, but that USCHO's page is calculating RRWP based upon a different set of ratings that does not include the dummy team. In order to test that, you would need to have a spreadsheet that takes every team's schedule and uses it to generate the ratings, and see if it produces the same or different ratings for each team.

    That spreadsheet is a major piece of work. I believe that Grant has one; I do not, and don't intend to build one.

  2. #22
    #SOAR TonyTheTiger20's Avatar
    Join Date
    Jan 2007
    Location
    Madison, CT
    Posts
    10,692
    Quote Originally Posted by Still Eeyore View Post
    My hypothesis is not that there is an additional dummy team used in the RRWP calculation. It's that there is a dummy team used by USCHO to calculate the ratings that you see on their KRACH page, and that you and I both used to calculate what the RRWP should be based upon those ratings, but that USCHO's page is calculating RRWP based upon a different set of ratings that does not include the dummy team. In order to test that, you would need to have a spreadsheet that takes every team's schedule and uses it to generate the ratings, and see if it produces the same or different ratings for each team.

    That spreadsheet is a major piece of work. I believe that Grant has one; I do not, and don't intend to build one.
    I do have it -- I'm working from home tomorrow so I'm planning on diving into both of your guesses/suggestions!

  3. #23
    Registered User
    Join Date
    Feb 2018
    Posts
    564

    Re: Fun With Numbers 2019: PWR, RPI, KRACH, GRaNT, WCHODR, & other assortments of let

    Quote Originally Posted by TonyTheTiger20 View Post
    I do have it -- I'm working from home tomorrow so I'm planning on diving into both of your guesses/suggestions!
    Because I've thought about building one, I have some questions. Foremost, do you have to hand enter game results, or do you have some place that you pull the data from?

  4. #24
    #SOAR TonyTheTiger20's Avatar
    Join Date
    Jan 2007
    Location
    Madison, CT
    Posts
    10,692
    Quote Originally Posted by Still Eeyore View Post
    Because I've thought about building one, I have some questions. Foremost, do you have to hand enter game results, or do you have some place that you pull the data from?
    I pull from USCHO's composite schedule using an external connection

  5. #25
    Fan of chipmunk-like mascots. ARM's Avatar
    Join Date
    Jan 2001
    Location
    The Gopher State
    Posts
    21,668

    Re: Fun With Numbers 2019: PWR, RPI, KRACH, GRaNT, WCHODR, & other assortments of let

    Quote Originally Posted by TonyTheTiger20 View Post
    I pull from USCHO's composite schedule using an external connection
    I'd use collegehockeystats instead of USCHO if you are starting from scratch. The former has been much more reliable over the years.
    "... And lose, and start again at your beginnings
    And never breathe a word about your loss;" -- Rudyard Kipling

  6. #26
    #SOAR TonyTheTiger20's Avatar
    Join Date
    Jan 2007
    Location
    Madison, CT
    Posts
    10,692

    Re: Fun With Numbers 2019: PWR, RPI, KRACH, GRaNT, WCHODR, & other assortments of let

    Quote Originally Posted by ARM View Post
    I'd use collegehockeystats instead of USCHO if you are starting from scratch. The former has been much more reliable over the years.
    Ha, yeah, I knew someone was going to mention how USCHO is always wrong with this stuff... I've thought about changing over to something else, but decided against it because even though it's just a matter of changing the import, I'd also have to re-match up the columns. Also USCHO's problem seems to be more in the line of having dates/times wrong. Once the official boxes get posted everything seems to match up right.

  7. #27
    Fan of chipmunk-like mascots. ARM's Avatar
    Join Date
    Jan 2001
    Location
    The Gopher State
    Posts
    21,668

    Re: Fun With Numbers 2019: PWR, RPI, KRACH, GRaNT, WCHODR, & other assortments of let

    Quote Originally Posted by TonyTheTiger20 View Post
    Ha, yeah, I knew someone was going to mention how USCHO is always wrong with this stuff... I've thought about changing over to something else, but decided against it because even though it's just a matter of changing the import, I'd also have to re-match up the columns. Also USCHO's problem seems to be more in the line of having dates/times wrong. Once the official boxes get posted everything seems to match up right.
    It has been better in recent years. I remember times 6 or so years ago when there were wholesale problems. I don't recommend going to the effort of changing if you have a working application, but I wouldn't build something new on top of USCHO.
    "... And lose, and start again at your beginnings
    And never breathe a word about your loss;" -- Rudyard Kipling

  8. #28
    #SOAR TonyTheTiger20's Avatar
    Join Date
    Jan 2007
    Location
    Madison, CT
    Posts
    10,692

    Re: Fun With Numbers 2019: PWR, RPI, KRACH, GRaNT, WCHODR, & other assortments of let

    Quote Originally Posted by Still Eeyore View Post
    My hypothesis is not that there is an additional dummy team used in the RRWP calculation. It's that there is a dummy team used by USCHO to calculate the ratings that you see on their KRACH page, and that you and I both used to calculate what the RRWP should be based upon those ratings, but that USCHO's page is calculating RRWP based upon a different set of ratings that does not include the dummy team.
    So, I took out the phantom tie against an average team. After doing that, the rankings didn't converge. So that didn't do it.

    I also thought maybe we're overthinking this and RRWP is just your odds of beating a team with a rating of 100. That didn't do it either.

    This is obnoxious.

  9. #29

    Join Date
    Oct 2005
    Location
    20904/13677/07677/07621
    Posts
    37,689
    Quote Originally Posted by TonyTheTiger20 View Post
    So, I took out the phantom tie against an average team. After doing that, the rankings didn't converge. So that didn't do it.

    I also thought maybe we're overthinking this and RRWP is just your odds of beating a team with a rating of 100. That didn't do it either.

    This is obnoxious.
    They're using Red Kryptonite.

  10. #30
    #SOAR TonyTheTiger20's Avatar
    Join Date
    Jan 2007
    Location
    Madison, CT
    Posts
    10,692

    Re: Fun With Numbers 2019: PWR, RPI, KRACH, GRaNT, WCHODR, & other assortments of let

    I still think Eeyore's guess is right... I need to see if I'm screwing things up by how I'm deleting the tie.

  11. #31
    #SOAR TonyTheTiger20's Avatar
    Join Date
    Jan 2007
    Location
    Madison, CT
    Posts
    10,692

    Re: Fun With Numbers 2019: PWR, RPI, KRACH, GRaNT, WCHODR, & other assortments of let

    Quote Originally Posted by TonyTheTiger20 View Post
    I still think Eeyore's guess is right... I need to see if I'm screwing things up by how I'm deleting the tie.
    Oh. It's because if you take out the tie, the teams aren't well connected because none of the NEWHA have beaten anyone outside the conference. If I make all games involving NEWHA teams exhibitions, the new KRACH values do converge.

    Here's the thing though -- you can still calculate RRWP when that happens by separating the teams into groups (i.e. the NEWHA and everyone else). Take out all the games not involving the NEWHA, calculate KRACH for everyone, and assign a 1.000 winning percentage for everyone vs. the NEWHA teams.

    ...But even when you do that, RRWP still doesn't match up to USCHO's.

    I'm so annoyed.

  12. #32
    #SOAR TonyTheTiger20's Avatar
    Join Date
    Jan 2007
    Location
    Madison, CT
    Posts
    10,692

    Re: Fun With Numbers 2019: PWR, RPI, KRACH, GRaNT, WCHODR, & other assortments of let

    Quote Originally Posted by Still Eeyore View Post
    My hypothesis is not that there is an additional dummy team used in the RRWP calculation.
    For what it's worth -- I actually tried the same thing CH86 did, because if the ratings were calculated by adding a tie versus a fictional team, then as far as the system knows, there's 41 teams playing each other. So, if you plug them in as another opponent while calculating RRWP, I thought maybe that would be where they were getting it. But as CH86 pointed, out, that ain't it...

    UPDATE: I pulled men's KRACH for every place I could find on Google (which is to say, Jim Dahl's and CHN, plus USCHO) and they were all different... lol. CHN's RRWP calculates the way I expected it to. Jim Dahl's doesn't have RRWP. A few historical KRACH posts that had RRWP (eLynah's) also matched what we would expect for RRWP.

    USCHO's RRWP didn't match any of them, so now I'm thinking they're notusing the removed-tie KRACH to calculate RRWP.

    I really have no idea what calculation USCHO is doing for RRWP in men's or women's, but really it just appears to be straight up wrong. I'll probably build RRWP (the way we expect it to be calculated) into my spreadsheet in the next few weeks.

  13. #33
    Registered User
    Join Date
    Nov 2018
    Posts
    49

    Re: Fun With Numbers 2019: PWR, RPI, KRACH, GRaNT, WCHODR, & other assortments of let

    If the season ended today, the NCAA tourney committee would have an interesting seeding vs. travel decision to make with OSU climbing up to #4 in the Pairwise. Would they send the #8 seed CHA champ (presumably Robert Morris or Mercyhurst) to Columbus to save a flight? Would Wisconsin then be stuck with a Princeton team that gave them a couple close games in Madison back in October? I realize there's a good couple months of hockey to play (especially conference tourneys will change these seed), but I'm curious what the NCAA would do if they were dealt this hand in March?

    Using the current Pairwise rankings, this would be the bracket:
    8 CHA Champ @ 1 Wisconsin (Flight)
    5 Princeton @ 4 Ohio State (Flight)
    6 Clarkson @ 3 Northeastern (Bus)
    7 Cornell @ 2 Minnesota (Flight)
    2005 Piston Cup Champion

  14. #34
    #SOAR TonyTheTiger20's Avatar
    Join Date
    Jan 2007
    Location
    Madison, CT
    Posts
    10,692

    Re: Fun With Numbers 2019: PWR, RPI, KRACH, GRaNT, WCHODR, & other assortments of let

    I believe Robert Morris is the only team that can bus to Ohio State, off the top of my head. So if they make the tournament, there's a 100% chance they send them to Columbus.

    Looking like they may be changing that for next year though.......

    That would send 8 RMU to 4 OSU, 7 Cornell to 1 Wisconsin, 6 Clarkson to 2 Minnesota, and 5 Princeton to 3 Northeastern. 2 flights, 2 busses.

    Such a stupid rule, my God.

    EDIT: No, Mercyhurst is a bus to OSU too. So if MU or RMU wins the CHA and OSU is top 4, that's where they'll be going.

  15. #35
    Registered User
    Join Date
    Nov 2018
    Posts
    49

    Re: Fun With Numbers 2019: PWR, RPI, KRACH, GRaNT, WCHODR, & other assortments of let

    PairWise question:

    How are ties in the overall PairWise rankings interpreted come tournament selection time? I know that RPI is the tiebreaker when awarding PairWise points between two teams, but how does it work when a number of teams are tied when all is said and done? If a number of teams were tied at say #5, things could get really interesting in a hurry if there's not a clear cut tiebreaker.
    2005 Piston Cup Champion

  16. #36
    Fan of chipmunk-like mascots. ARM's Avatar
    Join Date
    Jan 2001
    Location
    The Gopher State
    Posts
    21,668

    Re: Fun With Numbers 2019: PWR, RPI, KRACH, GRaNT, WCHODR, & other assortments of let

    Quote Originally Posted by ChickHicks86 View Post
    PairWise question:

    How are ties in the overall PairWise rankings interpreted come tournament selection time? I know that RPI is the tiebreaker when awarding PairWise points between two teams, but how does it work when a number of teams are tied when all is said and done? If a number of teams were tied at say #5, things could get really interesting in a hurry if there's not a clear cut tiebreaker.
    RPI breaks that tie as well. So if you want to know how the teams rank, it almost always follows the RPI order.
    "... And lose, and start again at your beginnings
    And never breathe a word about your loss;" -- Rudyard Kipling

  17. #37
    #SOAR TonyTheTiger20's Avatar
    Join Date
    Jan 2007
    Location
    Madison, CT
    Posts
    10,692
    Quote Originally Posted by ARM View Post
    RPI breaks that tie as well. So if you want to know how the teams rank, it almost always follows the RPI order.
    It's a little wild this year. BC is locked into losing the comparison to UMD and there are a couple other flipped comparisons too. More so than usual anyway.

  18. #38
    #SOAR TonyTheTiger20's Avatar
    Join Date
    Jan 2007
    Location
    Madison, CT
    Posts
    10,692

    Re: Fun With Numbers 2019: PWR, RPI, KRACH, GRaNT, WCHODR, & other assortments of let

    First bracketology of the year: https://www.bcinterruption.com/bosto...a-northeastern

    I really, really, really don't know what the committee would do if it ended this way. You can keep perfect bracket integrity even with minimizing flights (which, as I scream into the wind every year, is THEIR ONLY DIRECTIVE), but there is a ton of money and driving time to be saved by sending Cornell to Clarkson, Princeton to Minnesota, and BC to Northeastern. It'll be interesting to see where this all shakes out.

    Also, with two teams pretty much being required to fly out west (unless OSU gets in, which I don't see happening given their general free fall plus having 2 more with Wisconsin left, or if someone random wins the WCHA, which, again, lol nope), the winner of the CHA won't likely change anything, since you're going to have 2 flights and 2 bus trips regardless. The CHA champ would be going to the 1 (unless someone really crazy wins another conference that's ranked lower than the CHA champ) and that's that.

  19. #39
    Registered User
    Join Date
    Feb 2017
    Posts
    13

    Re: Fun With Numbers 2019: PWR, RPI, KRACH, GRaNT, WCHODR, & other assortments of let

    Forgive my ignorance as I'm new to how the ncaa playoff format works. I'm wondering as to how these teams are picked and the playoffs haven't even started yet. If a Princeton or Clarkson or Cornell in the ECAC lose in the playoffs, am I understanding right that it doesn't mean squat. They would go to the final 8 and the team that upset them wouldn't. If this is the case, why have playoffs? Im confused!! I would appreciate any information on this please and thank you.

  20. #40
    Registered User
    Join Date
    Mar 2012
    Posts
    1,159

    Re: Fun With Numbers 2019: PWR, RPI, KRACH, GRaNT, WCHODR, & other assortments of let

    Quote Originally Posted by 99hockey View Post
    Forgive my ignorance as I'm new to how the ncaa playoff format works. I'm wondering as to how these teams are picked and the playoffs haven't even started yet. If a Princeton or Clarkson or Cornell in the ECAC lose in the playoffs, am I understanding right that it doesn't mean squat. They would go to the final 8 and the team that upset them wouldn't. If this is the case, why have playoffs? Im confused!! I would appreciate any information on this please and thank you.
    The top eight teams in the Pairwise make the NCAA tournament, unless a team from outside the top 8 wins their conference tournament and thus wins that conference's "automatic bid". Right now, we already "know" that the CHA winner will be someone not in the top eight. So Grant's description above (below) is assuming the NCAA will consist of the top seven in the Pairwise plus the CHA winner. If there was to be a 'upset winner' of one of the other conferences, then the NCAA field would be the 'top six' plus CHA winner plus upset winner.

    Right now we 'know' that Wisconsin. Minnesota and probably Northeastern can't fall far enough to find themselves out of the tournament. So we 'know' it will be Wisconsin, Minnesota, Northeastern, and the CHA winner. The other four spots are very much up for grabs.

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •