Announcement

Collapse
No announcement yet.

No more RPI?

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #16
    "It's about time the NCAA tournament modernized its selection process" - ESPN.com

    Joe Lunardi, Senior Writer
    9Posted: 08/22/18, :50 AM CT

    ESPN's Bracketologist Joe Lunardi says the NCAA's new metric to evaluate tourney teams is a step in the right direction, but questions remain about its effectiveness.


    "The RPI is dead, apparently, so long live NET (the new NCAA Evaluation Tool).

    If nothing else, credit the NCAA for a clever acronym. Forgive me for withholding any additional credit for the time being. Had the RPI long outlived its usefulness? Had schools and conferences learned how to "game" the system? Was the tiresome approach -- "it's only a sorting tool" -- failing to recognize the RPI's outsized influence within the NCAA men's basketball selection committee?

    Yes, yes and yes.

    Cutting to the chase, then, we are left with two unanswered questions as the RPI is laid to rest:

    • Is NET a better metric for team selection and seeding?

    • Will the selection committee utilize it correctly?

    To answer the first question, even without a complete breakdown of the new formula, NET almost can't help but improve upon the RPI. The only thing worse than RPI's emphasis on "who you played" over "how you played" was the committee's seeming inability to evaluate outliers. This is how we'd get inexplicable outcomes such as Wichita State (30-4) being a No. 10 seed in 2017...."

    Comment


    • #17
      Cap for home should be lower than cap for away.
      "I not sure that I've ever been around a more competitive player or young man than Fred VanVleet. I like to win more than 99.9% of the people in this world, but he may top me." -- Gregg Marshall 12/23/13 :peaceful:
      ---------------------------------------
      Remember when Nancy Pelosi said about Obamacare:
      "We have to pass it, to find out what's in it".

      A physician called into a radio show and said:
      "That's the definition of a stool sample."

      Comment


      • 1979Shocker
        1979Shocker commented
        Editing a comment
        It's probably already taken into account. A win on the road by 10 is better than a win at home by 10.

        And vice versa. I loss at home by 10 is worse than a loss on the road by 10.

    • #18
      Originally posted by RoyalShock View Post
      The NCAA is adopting a new metric - NET.

      The NCAA has developed the NCAA Evaluation Tool (NET) a ranking system that will replace the RPI as the primary sorting tool used to evaluate teams.


      I'm not sure 10 points is a big enough cap, but at least they are factoring margin of victory into the metric. I also think late-season games ought to carry a little bit more weight to account for players who returned from injury or became eligible after the 1st semester.
      Agreed on both counts, Royal. Your idea from another post of adding a gradually declining benefit for winning by more than 10 strikes me as very logical. And late-season games simply do matter more, mostly because as you note they better reflect the rosters and rotations teams will be taking into tournament play, but also partly because they can be more pressurized in terms of conference races, and (at least in my view) how a team performs when the chips are really down should count for something.

      Comment


      • #19
        How about just stripping the team names and the conference names from the profile sheets and let the numbers do the talking

        Comment


        • Kung Wu
          Kung Wu commented
          Editing a comment
          Makes too much sense.

      • #20
        Originally posted by Boss1786 View Post
        How about just stripping the team names and the conference names from the profile sheets and let the numbers do the talking
        If there are no team names, how would the committee know which numbers to give more weight to that year, though?

        Comment


        • #21
          Is RPI buried and in an official grave yet? Gotta pee real bad, looking to water the new grass seed.

          Comment


          • C0|dB|00ded
            C0|dB|00ded commented
            Editing a comment
            Burrah will be standing shoulder to shoulder with you. Careful though, he tends to stomp instead of shake.


            T


            ...:cool:

        • #22
          I think the KU model for scheduling will still trick the system. For their cupcake home games, they schedule the projected winners of some really bad conferences. Those opponents go on to produce great W/L records as they cruise through their conference games. That distorts strength of schedule. It "looks" like KU played a 25-win team, but the source of those wins doesn't seem to be discounted enough to offset the "value" of beating a severely overmatched team with a good-looking W/L record.

          KU got a Quadrant 2 win last year by trouncing South Dakota State by 30+ in Lawrence. It seems that type of a game will still be rewarded by strength of schedule considerations.

          The future's so bright - I gotta wear shades.
          We like to cut down nets and get sized for championship rings.

          Comment


          • FadedCrown
            FadedCrown commented
            Editing a comment
            We also got Q2 win by beating SDSU in Wichita.

          • Steeleshocker
            Steeleshocker commented
            Editing a comment
            We've taken the same approach in our scheduling. We just had the valley dragging down the RPI for many years.

            It's not rocket surgery.

          • another shocker
            another shocker commented
            Editing a comment
            and yet the mvc sent a team to the final four again last year.

            rpi be damned.

        • #23
          Originally posted by jdshock View Post

          If there are no team names, how would the committee know which numbers to give more weight to that year, though?
          Right? All those sub .500 P5 team's would need to watch out....you would actually have to finish in the top quarter of you conference to guarantee a spot....

          Comment


          • #24
            Originally posted by jdshock View Post

            If there are no team names, how would the committee know which numbers to give more weight to that year, though?
            Knowing the NCAA...they'd probably just look at the TV ratings per team. Lol
            FINAL FOURS:
            1965, 2013

            NCAA Tournament:
            1964, 1965, 1976, 1981, 1985, 1987, 1988, 2006, 2012, 2013, 2014, 2015, 2016, 2017, 2018, 2021

            NIT Champs - 1 (2011)

            AP Poll History of Wichita St:
            Number of Times Ranked: 157
            Number of Times Ranked #1: 1
            Number of Times Top 5: 32 (Most Recent - 2017)
            Number of Times Top 10: 73 (Most Recent - 2013, 2014, 2015, 2017)

            Highest Recent AP Ranking:
            #3 - Dec. 2017
            #2 ~ March 2014

            Highest Recent Coaches Poll Ranking:
            #2 ~ March 2014
            Finished 2013 Season #4

            Comment


            • #25
              So, you have a new system in place to rank teams. Then you make all of the calculations proprietary and dont release them to anyone. You wont back test it, publicly at least, to allay fears that the system is once again gamed to produce intended resluts that benefit the P5. <sarcasm font begin> Nice job NCAA. We all trust ya. <sarcasm font end>
              Go Shocks!

              Comment


              • #26
                "Why NET could be a good first step toward improving the NCAA tournament" - ESPN

                John Gasaway, ESPN Insider
                Posted: 08/29/18, 11:45 AM CT

                Some coaches are optimistic that NET, the tournament's new metric for evaluating teams, will be an improvement over using RPI. But it might depend on what happens next.


                "Last week the NCAA announced it will no longer use the Ratings Percentage Index in selecting and seeding the NCAA tournament field. It was a big step, and the NCAA is to be commended for doing something it very plainly did not want to do for years. (Fine, make that years and years and years.) Well done.

                In place of the RPI, we now have the NCAA Evaluation Tool, or the NET. In the NCAA's words, the new metric "relies on game results, strength of schedule, game location, scoring margin, net offensive and defensive efficiency and the quality of wins and losses."

                That sounds straightforward enough, and, by and large, the coaches I've spoken with this week seem cautiously optimistic about this strange new post-RPI world we now inhabit.

                Under the old set of rules still in play last spring, for instance, the NCAA passed over Nebraska and its 13-5 Big Ten record due to a perceived lack of quality wins. So perhaps we shouldn't be surprised to find that head coach Tim Miles is open to a new way of evaluating teams...."

                Comment


                • SB Shock
                  SB Shock commented
                  Editing a comment
                  I do not know why anybody would be “optimistic” about this? The NCAA track record on fixing anything is in the negative numbers.

                  The NCAA won’t release past year using the formula, won’t release any details of their formulation except some general gibberish.

                  I will bet if u take KenPom and RPI, play around with the weighting average, then u will come close to replicating it.

              • #27
                Wonder if the rpi will still be available to track. Be interesting to see it compared to the new NET at the end if the year.
                FINAL FOURS:
                1965, 2013

                NCAA Tournament:
                1964, 1965, 1976, 1981, 1985, 1987, 1988, 2006, 2012, 2013, 2014, 2015, 2016, 2017, 2018, 2021

                NIT Champs - 1 (2011)

                AP Poll History of Wichita St:
                Number of Times Ranked: 157
                Number of Times Ranked #1: 1
                Number of Times Top 5: 32 (Most Recent - 2017)
                Number of Times Top 10: 73 (Most Recent - 2013, 2014, 2015, 2017)

                Highest Recent AP Ranking:
                #3 - Dec. 2017
                #2 ~ March 2014

                Highest Recent Coaches Poll Ranking:
                #2 ~ March 2014
                Finished 2013 Season #4

                Comment


                • Stickboy46
                  Stickboy46 commented
                  Editing a comment
                  The RPI is a well known formula. I’m sure several sites will have it running this year still.

              • #28
                Originally posted by AZ Shocker View Post

                Knowing the NCAA...they'd probably just look at the TV ratings per team. Lol
                You've cracked the code.

                Comment


                • #29
                  “Losers Average Losers.” ― Paul Tudor Jones

                  Comment

                  Working...
                  X