test content
What is the Arc Client?
Install Arc

Elo system is malfunctioning

zaodonnzaodonn Member Posts: 109
edited March 2014 in PvE Discussion
As everyone knows, the Elo system doesn't work.

5 brand new level 60s in blue gear who have never played any PvP matches (no elo rank) are being matched against a 5-man premade Elite guild group with all Rank 10s/Perfects.

And not just occasionally, but every single match. Yesterday I played 30 matches, and all 30 had this problem. 5 low gear/new players vs 5 elite players (sometimes premade, sometimes pugs).

Please fix the Elo system, as it is horribly broken atm.
Post edited by zaodonn on

Comments

  • vedran541vedran541 Member Posts: 199 Arc User
    edited March 2014
    Usually when i solo que i get those pugs on my team. 9k gs tops, running straight to their home point on their rental ponies.
  • killzoneexkillzoneex Member Posts: 94 Arc User
    edited March 2014
    I think it's because those elite guilds are removing all gear before queueing up. Happened to me on a new 60 I tried doing PVP with.
  • zaodonnzaodonn Member Posts: 109
    edited March 2014
    Its not based on gear, its based on your win/loss ratio. Elite guilds who are playing lots of games and winning should have a high rating, and never be matched against new lvl 60s.
  • cerebralpyrecerebralpyre Member Posts: 24 Arc User
    edited March 2014
    Or youre expecting an additional 5 lvl 60s in all blue gear to be on at the same time as you?
    Cerebralpyre 13.2k SW
    Pheryllt - 13.8k HR
    Ginko - 10.3k DC


    Puzzl Bochs - lvl38 CW
  • edited March 2014
    This content has been removed.
  • akemnosakemnos Member Posts: 597 Arc User
    edited March 2014
    the should have done a standard number for the ELO. for example MTG starts with a base 1600 rating and then you go from there. the top players are in the 2000+ range.
  • ambisinisterrambisinisterr Member, Neverwinter Moderator Posts: 10,462 Community Moderator
    edited March 2014
    Elo is not directly based on Gear Score. That is a complete misconception.

    It is more on a win loss equation so the issue you are describing is more along the lines that the "elite players" haven't done enough matches to get properly ranked.
    akemnos wrote: »
    the should have done a standard number for the ELO. for example MTG starts with a base 1600 rating and then you go from there. the top players are in the 2000+ range.

    This is likely how it is. That is how a normal Elo system works.
    Thing is everybody stats at the same place so it will take time to get players where they need to be.
  • zaodonnzaodonn Member Posts: 109
    edited March 2014
    Elo is 100% definitely broken, absolutely verified in-game.

    Players with 100s of matches (since Elo came out) are ABSOLUTELY being put up against new 60s in blue gear with no elo ranking.
  • ambisinisterrambisinisterr Member, Neverwinter Moderator Posts: 10,462 Community Moderator
    edited March 2014
    Also known as people who happened to win against even lower ranked opponents.

    Gear Score is not a skill rating.
    Plus in order to get the players spread out more the restrictions are a bit more relaxed to reduce the initial queue times. Those will be tightened again as the player rankings mature.

    All of this was expected. -.-
  • vasillesvasilles Member Posts: 33 Arc User
    edited March 2014
    I noticed every match teams being like that ;
    a-1 or 2 full skilled 25-0 type player :mad:
    b- 2 x Avarage :rolleyes:
    c-1x or 2x Total newbie :confused:
    You can count yourself full skilled :cool: but main tactic should be avoid that full skilled ones if gwf or GF ambush them if other classes, you can easly manage win match if you detect who comes from which range skill.
    Join SCUM Feel Hate Of Others.
  • velynnavelynna Member Posts: 336 Arc User
    edited March 2014
    killzoneex wrote: »
    I think it's because those elite guilds are removing all gear before queueing up. Happened to me on a new 60 I tried doing PVP with.

    Disregarding the fact that ELO doesn't work according to GS, I sincerely doubt that would ever be the case. In the premades I run with my guild, we don't want to run into PUGs any more than you want to run into us. There is no fun in doing nothing, waiting for the game to finish while you hang out in spawn.
  • godlysoul2godlysoul2 Member Posts: 661 Bounty Hunter
    edited March 2014
    Matches are much more balanced overall and the system is working well.

    I believe the problem is the amount of players of similar high ranking. If I had to guess what happened I would think one of the ways they implemented to help decrease matchmaking time was by widening the gap of who can be paired with who based on rankings. Generally most of my matches are balanced, but sometimes you do get blowout matches. I believe in those cases there just aren't enough people q-ing that are around your rank, so the system settles for pairing a "not so even" match to help avoid the crazy long que time.

    That is just a random speculation though.

    I also believe it has to do with people still just joining the system or not having played enough matches for it to recognize how skilled they are. Maybe people are logging on alts to pvp on the system that haven't been used on it yet or people were on vacation/took a break and are just coming back.

    Who knows. Ultimately the matchmaking system does a good job of relatively pairing up teams of similar skill most of the time. It is definitely better than pre-elo even if its not perfect.
  • mehguy138mehguy138 Member Posts: 1,803 Arc User
    edited March 2014
    Matches are still as unbalanced as before, but now it takes 10 times more time to get into the match.
    M6 almost drains your soul given how boring it is. (c) joocycuzzzzzz
  • pers3phonepers3phone Banned Users, Neverwinter Beta Users Posts: 0 Arc User
    edited March 2014
    After today's patch, we ran out into 4-5 men premades each time we were 4-5. At one point we faced full premade while 4, and the matchmaking allocated us a horrible HR that was nowhere to be found :\ So there are still some issues with filling the missing spots with competent people before throwing them against full PM.

    Also when I solo queued I got in match with a few known PvPers and we had a very close nice match, which was pretty cool, since almost all my previous solo queues were either with pugs with all kinds of skill level (but no PvPers), or full enemy premades which never had any challenge whatsoever.

    So overall, some nice games today. Not yet sure if luck or ELO functioning better, but I was pleasantly surprised to be put in a match with actual PvPers instead of pugs when I solo queued, and that the queue fed our premade only other premades and not pugs.
  • xthebluespiritxxthebluespiritx Member Posts: 0 Arc User
    edited March 2014
    just imagine if you were able to use the ray gun on call of duty multiplayer, while everyone else has pistols. thats how i feel when i play nw pvp, im the dude with the pistol
  • zolronzolron Member Posts: 30 Arc User
    edited March 2014
    pers3phone wrote: »
    After today's patch, we ran out into 4-5 men premades each time we were 4-5. At one point we faced full premade while 4, and the matchmaking allocated us a horrible HR that was nowhere to be found :\ So there are still some issues with filling the missing spots with competent people before throwing them against full PM.

    Also when I solo queued I got in match with a few known PvPers and we had a very close nice match, which was pretty cool, since almost all my previous solo queues were either with pugs with all kinds of skill level (but no PvPers), or full enemy premades which never had any challenge whatsoever.

    So overall, some nice games today. Not yet sure if luck or ELO functioning better, but I was pleasantly surprised to be put in a match with actual PvPers instead of pugs when I solo queued, and that the queue fed our premade only other premades and not pugs.

    Pretty sure my PUG group came up against 4 of you, and that horrible HR :) Even so, the group I was in had 12-16k GS players, but as usual we lost pretty badly as we lacked the communication and overall didn't seem especially skilled/experienced. I'm not sure where you got the only pre-made vs premade from ?

    My experiences have been all over the map really. Still going against full/near-full pre-mades occasionally when I solo que and players range from 8k to 17k GS. I still go with the cryptic tuned down the matchmaking to where ELO barely accounts for anything theory :P
  • izatarizatar Member, NW M9 Playtest Posts: 1,161 Arc User
    edited March 2014
    OP seems to have queued as a 5 man premade. I am sure the algorithm gives an elevated priority for premades. Since not many premades are around, and most of them are going to be organized, it is very likely you would be put up against them. How would you have it? Your 5 man premade gets put up against 5 random players? Thats not equitable.
  • nameexpirednameexpired Member Posts: 1,282 Arc User
    edited March 2014
    Elo is not directly based on Gear Score. That is a complete misconception.

    It is more on a win loss equation so the issue you are describing is more along the lines that the "elite players" haven't done enough matches to get properly ranked.
    So you are saying that GS factors into the matchmaking, though "not directly"?
    Imaginary Friends are the best friends you can have!
  • pers3phonepers3phone Banned Users, Neverwinter Beta Users Posts: 0 Arc User
    edited March 2014
    zolron wrote: »
    Pretty sure my PUG group came up against 4 of you, and that horrible HR :) Even so, the group I was in had 12-16k GS players, but as usual we lost pretty badly as we lacked the communication and overall didn't seem especially skilled/experienced. I'm not sure where you got the only pre-made vs premade from ?

    My experiences have been all over the map really. Still going against full/near-full pre-mades occasionally when I solo que and players range from 8k to 17k GS. I still go with the cryptic tuned down the matchmaking to where ELO barely accounts for anything theory :P

    Oh yeah I think we ran into each other :) Hello there (you are a GF, yes?). The games we speak about are different. The one we both were in was a sad exception to the PM vs PM and it was as lame as PM vs Pug can get, sorry about that. Good thing you posted, since I completely forgot about that game since it just ended fast.

    The match I was talking about was a full premade we went against while we were 4 and got the baddie HR. To make things worse one of out GWFs DCed for like 5 mins but was able to return and we still won.
  • ambisinisterrambisinisterr Member, Neverwinter Moderator Posts: 10,462 Community Moderator
    edited March 2014
    I don't know the details but I do know seeing gear score difference is not evidence the system is not working.

    The Elo System is designed to judge skill, not gear score. That is how an Elo system works. If you look up the Elo System you will find no reference to gear scores.

    As such even in PvE I have described players either being able to outplay their gear score or use gear to compensate for a lack of playskill. You could have a person with 18K gear score be able to play only as well as a person is expected to play with 13K gear score and on the opposite side of the equation a person with 8K gear score could play as well as a person with 13K gear score.

    Gear Score alone has never been and never will be a good way to judge skill.
    Any claim that the system isn't working because of the gear scores of other players are simply not even in the realm of validated. In fact it is more likely that gear score plays no role than a small role which is why gear scores have absolutely no correlation to the functionality of the Elo System.

    It could take quite a while for things to level out and that is all that is being shown by saying gear scores are weirdly mixed.
  • yogokouyogokou Member, Neverwinter Beta Users Posts: 0 Arc User
    edited March 2014
    I'm thinking that matchmaking forms teams based on Elo averages now to save time i.e. 2 guys with good ranking, 2 averages and a newbie VS 3 man premade and two newbies, or something like that. I've been getting hit with top premades after a 4-5 win streak everytime, so from my experience it definitely works to a degree. Once people gravitate to their rankings, it might shift more to matching people from same rankings together.
Sign In or Register to comment.