Should the ELO Ranking System Be Used For Boxing?

Collapse
Collapse
  • Time
  • Show
Clear All
new posts
  • DeadLikeMe
    ................
    Unified Champion - 10,00-20,000 posts
    • Dec 2011
    • 11968
    • 748
    • 1,409
    • 26,662

    #11


    Euro bums and Pac and May come to mind when I read that section

    Comment

    • Chups
      Banned
      Unified Champion - 10,00-20,000 posts
      • May 2004
      • 18400
      • 1,835
      • 1,281
      • 52,879

      #12
      The ELO rating which is primarily used in Chess and in most of the gaming industry is FLAWED. The new less FLAWED system is called TSR (True Skill Rating)....that would better represent the true skill of boxers.

      Comment

      • DeadLikeMe
        ................
        Unified Champion - 10,00-20,000 posts
        • Dec 2011
        • 11968
        • 748
        • 1,409
        • 26,662

        #13
        Originally posted by Chups
        The ELO rating which is primarily used in Chess and in most of the gaming industry is FLAWED. The new less FLAWED system is called TSR (True Skill Rating)....that would better represent the true skill of boxers.
        Just because it is flawed doesn't mean there are more viable or popular options out there.

        Comment

        • gingerbreadman
          Undisputed Champion
          Platinum Champion - 1,000-5,000 posts
          • Aug 2010
          • 2245
          • 348
          • 55
          • 11,136

          #14
          I'm a chess player myself, and the first problem I can think of is that modern-day chess is subject to rating inflation (more people playing, leading to today's guys having higher ratings than Bobby Fischer even though he would most probably **** them). In boxing, though, I'm pretty sure you'd suffer the opposite, namely deflation from fewer people boxing now than in, say, the 50s. I think boxrec must use a similar system, and if you look at their historical rankings, it seems that the earlier eras are WAY stronger than nowadays (SRL ranked no. 12 at welter seems the most glaring). Anyways, Arpad Elo himself said that his own rating system was about as accurate as a cork on a line on a fast-moving river. Not feasible in my opinion, though I'd love to be proven wrong.

          Comment

          • Chups
            Banned
            Unified Champion - 10,00-20,000 posts
            • May 2004
            • 18400
            • 1,835
            • 1,281
            • 52,879

            #15
            Originally posted by DeadLikeMe
            Just because it is flawed doesn't mean there are more viable or popular options out there.
            lololololololol............

            Comment

            • BoxingGenius27
              Banned
              Super Champion - 5,000-10,000 posts
              • Dec 2009
              • 8502
              • 370
              • 463
              • 9,603

              #16
              Originally posted by DeadLikeMe
              Something like "margin of victory" would have to be included as well (or a composite average of the 3 cards) to factor in if something is truly a "good win."

              If I didn't have so much to do at work today (as I'm typing this on NSB lol)...I would actually sit down and formulate a rough algorithm of some kind. If other people show interest in this and contribute I'm sure we could come up with something by the end of the weekend.
              Any good news; need help with anything?

              I work with a few engineers who are genius's in excel. Is there anything I can run past them?

              Comment

              • zdiddy
                Amateur
                Interim Champion - 1-100 posts
                • Jan 2012
                • 10
                • 0
                • 1
                • 7,340

                #17
                Did anything ever come of this? I started exploring ranking systems and stumbled across this thread but it seemed to end quite abruptly.

                Comment

                Working...
                TOP