General Discussion

General Discussion[Suggestion] Calculating ELO Rattings

[Suggestion] Calculating ELO Rattings in General Discussion
Hurque

    I know that probably a lot hate is going to be generated by this idea, but I still think that this is a good one.

    ELO Rattings (http://bit.ly/XEWJw). When I was part of the lain (ghost bot - http://www.codelain.com/forum/ ) community, some of the bots were modded to support ELO rattings, and it worked nicely.

    Since we have all the data needed why not to use it ? I know that to calculate ELO with this HUGE dataset is going to take a lot of time, and processing time. But it really dont need to be real time, so maybe 1 or 2 updates/day its more than enough.

    What do you guys think ?

    //As always just ignore the engRish mistakes ;D

    Flaccid Jack

      ^

      CyanKnight

        They hinted at having at least something like a rating system underway, I'm looking forward to the next site update.

        I really really miss this in Dota and it would be absolutely awesome :)

        BtBp: wrAth~

          +1 to this idea, knowing your ELO rating would be nice :)

          dCM.Derpy.Mu

            what is ELO rating?

            Scadi

              TrueSkill is a much better rating system, since ELO is a single person rating (good for chess and such) but shit for multyplayer games (dota, LoL, CS).

              TrueSkill
              The TrueSkill rating system was designed in the 21st century at Microsoft Research, with multi-player and multi-team games in mind. The system uses very complex formulas and has greater precision than any previous rating system. The system was not built from scratch, it has heavy influences from ELO and Glicko. TrueSkill tries to lower the number of games needed to get an accurate rating for a single player. The system is extremly dynamic, allowing both quick movements and being able to slow movements down. TrueSkill also changes the ratings of different players differently if they are in same team. At some point or another a players rating will pretty much freeze. This is the goal of the system. The system believes that it has found the true skill level of that player. Unless that player will show unpredictable performance the rating will not change. In simple words - unless the player improves his real skill the TrueSkill rating will not change.

              http://research.microsoft.com/en-us/projects/trueskill/

              Belga

                +1 for this implementation

                HmC

                  Thumbs up!

                  DeltaFoxtrotBravo

                    Yeah I remember Dotacash (one of the main servers for Dota 1) implementing a similar, if not the same, rating system into their servers. unfortunately no one ever looked at it cause everyone was used to the elo system and you couldnt check it in game and no one understood it really.

                    Trueskill would be nice though, however I hope it wouldn't "freeze" as you say, improvement or regression in skill would be nice to monitor to know trends and mistakes, helping understand strengths and weaknesses

                    Kraju

                      TrueSkill sounds better - go for TS.

                      Koma

                        There must already exist some ELO value in DotA2. the match making is working pretty good, so the system must have some value to compare players.

                        CyanKnight

                          Yeah but they want to keep it hidden (questionable decision), so if we want to have a rating we shall have to work with what we do get. Dotabuff already has all the stats, they just need to find a way to make something like TS work..

                          Koma

                            Your TS calculation just doesn't work if you forget that the set of players it's related to is not the set of all players but those who were given by match making. The statistics wouldn't show how skilled you are... it's just some value like the win percentage... just play a game in very high pool and one with 4 newbies. even a player with 45% win rate playing in very high plays way better than someone with the same amount of games but 55% win rate in a medium game.

                            CyanKnight

                              TS is like ELO, only ELO does not handle multiplayer well. It has not much to do with win percentage.

                              VeRychard
                                Yorum silindi
                                Dungeon Master

                                  I like this