General Discussion

General DiscussionI know this isn't important to most people but...

I know this isn't important to most people but... in General Discussion
123

    I am curious, how are these numbers determined?

    http://dotabuff.com/heroes/phoenix/players

    I am looking at thier WR and KDA, etc, and most of them have a smaller WR and KDA. Some have 30 games too, same as me. I also place a shit ton of wards.

    http://dotabuff.com/players/100477876/heroes

    Could someone explain it to me?

    123

      No one really knows?

      Сука Spirit

        They play in the 5.5k-6k mmr bracket, you don't

        123

          So then shouldn't it also say "Only people with 30 games for the hero AND are ranked 5.5k+"?

          Сука Spirit

            Quote: Hero Score is a composite statistic that weighs multiple factors including DBR (difficulty), win rate (success), kda (impact) and matches played (risk).

            swoleytrinity

              If a player with 5K and a player with 4K have the same stats the player with 5K will place higher due to a higher "difficulty".

              It doesn't mean they are all and only 5K but it will pick the best players at the highest bracket available. When the rankings for Phoenix came out one of the first on there had like a 3.4K rating but was bumped off the list as soon as higher tier players came on.

              I do feel it is a little shit as if you have a massively good DBR rating (pre DBR takedown) then you are almost guaranteed a place on the list even if you are subpar on the hero.

              MILNOR

                If thats true then this list is pointless. Say the most skilled Timbersaw in the world who knows the hero inside out, completely dominates with him but is trash with other heroes and has a low rating; that guy shouldnt be on the timber list? Why?

                Also on the 'Heroes success' feature of plus I have a Sven at 64% win rate but the success % s something like 33%.. These seem to be nothing but random numbers to me.

                Gustaphos

                  @ RM LORDY

                  Agreed. I don't get the success %'s it would be nice what the calculation is with it and where they are pulling the #'s from. Doesn't make sense. I don't know how you could calc what ||| is saying difficulty with win rate, and KDA. There are missing factors of where it calc's the difficulty. I mean how do they add all 4 of those together?

                  Also Agreed on the timber equation. Everyone that is a pubber like me has their go to guys they are great at then are garbage as the rest. He should be on the list! Doesn't make sense otherwise.

                  123

                    > RM & -hg-

                    I agree. I mean, I think the thirty game requirement is fine, the kda part is fine, etc, but the mmr requirement seems ridiculous to me. Someone with a lower MMR would have to play far more games with that hero to get on that list than someone who is already sitting in the 5k+ bracket when the hero came out. Not everyone wants to carry themselves to 5.5k pub stomping with heros like slark, many like playing strictly support heros where it is far harder and takes much longer to get to those mmr's.

                    But anyway, im sure the super 1337 players who just pub stomped to 5.5k with slark or some other pup stomper will be here to slap us down and tell us why we're totally wrong shortly.

                    Ce commentaire a été édité
                    Gustaphos

                      ^Agreed. I like support more bc it is rare in my MMR range to have someone that will buy support items.

                      I mean now hearing me say this... I could try focusing on a game changing hero thats OP like slark or something, practice then start just strictly picking him. Then after I get my MMR out of the trash can go back to playing support roles since I enjoy that more.

                      Сука Spirit

                        Are you all saying that 3k people with 100% winrate on a certain hero play that hero better than the 5k with only 50%?
                        No, they don't.

                        123

                          > |||

                          Are you saying that it is theoretically impossible?

                          Сука Spirit

                            Yeah, if you are better than dendi you shouldn't be 3-4k. There is a lot to playing well a hero besides understanding the hero, ie: game sense, map awareneses, decision making

                            123

                              Lol.

                              1. It's unlikely, not impossible.
                              2. It's unhealthy to worship others as gods.
                              3. Dendi messes up plenty.
                              4. So does everyone else.

                              Ce commentaire a été édité
                              Сука Spirit

                                Worship? lol? I was merely referring to him as an example. The logic behind the difficulty is perfectly resonable, stop whining about how your name is not listed because honestly you are just a trench tier player.
                                I thought you were just looking for an explanation but it seems like you are just one of those players who don't even know how shit they are and keep telling people that they are better than what their stats say.
                                http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect