+ Reply to Thread
Page 5 of 11 FirstFirst 1 2 3 4 5 6 7 8 9 ... LastLast
Results 61 to 75 of 153
Like Tree55Likes

  Click here to go to the first Rift Team post in this thread.   Thread: How ELO Works in RIFT

  1. #61
    Ascendant ecru's Avatar
    Join Date
    Jul 2013
    Posts
    3,741

    Default

    Quote Originally Posted by Ocho View Post
    You can get in a bad streak, and that's never much fun. That's true of any team competition however. One thing that writeup above doesn't mention is that after the above calculations, the matchmaker will try to balance people on losing streaks between the two teams. That means that half the folks on losing streaks will get their streak snapped in any given match. Short streaks, even 10, are still possible. But we do try to make it as unlikely as we can.
    losing streaks of 10+ is why i quit. cancelled my patron sub, quit raiding, quit pvping, haven't logged on in weeks. the last 3-4 weeks i played, my win/loss rate was roughly 1:4-1:5 solo queueing, but my ELO on pts is still 3600.

    maybe you played rift before you instituted this ridiculous matchmaking system or maybe you didn't so you may or may not know how much better it was before you messed everything up by overcomplicating it, but the fact of the matter is that you are punishing players for playing well and they are quitting your game because of it.

    the reward for being good at rift is eating **** in 9 out of 10 warfronts. why?

  2. #62
    Ascendant Violacea's Avatar
    Join Date
    May 2011
    Posts
    12,128

    Default

    So there is a way to code a game to take into account your current streak as well. Now I know SFIV did this. Always wondered if it was placebo effect or consistent coincidence. Anytime I'd lose like 4 - 5 games, it's like it would pit it against a free win nearly.

    But still, ratings are just numbers.

    In said other game, I am @ 2200pp (overall wins) / 6150 bp (wins with particular character).

    I've beaten people with 3kpp/14000bp. I've also lost to players with 340pp/280bp (I hate these so much).

    And this is ratings based on 1 v 1 performance where they include things like set marks to where to increase your rating, you face a harder curve. Ie. 5000 is when you become a B ranked player with whatever player hits that rating. At that point, instead of getting 22 - 30 points for people with 999bp or less, you get +1. The game assumes you can handle those fights by this time and don't deserve points for them. So the better you get, you still face a harder curve to keep the number climbing. I must say, for a rating system, it is done fairly well.

    But this is for a balanced 1 v 1 game where the objective is crystal clear. Ratings are only going to mesh out into more meaningless figures in an imbalanced team heavy game. The ratings are individual, and the result of a WF is rarely the result of just one individuals efforts. See where this is going?

    Food for thought. I mean, it's probably better to attempt it than not to. But I think some people are putting waaaay to much worth into these ratings systems, or any game with a rating system for that matter.

    I only find this reflectively interesting because the 2 current games I hit hard (qlive and sfiv) both have rating systems now. In the 90's they didn't. And you know what? I can't say the rating system has even drastically changed the experience and the challenges I do or don't face. So again, food for thought - don't put so much end all trust in these rating systems and take it out on devs when you lose a game that you perceived should have been won. Ratings don't mean anything *should* happen.
    Last edited by Violacea; 04-07-2015 at 05:54 PM.
    New round up of some high rank matches
    Chun-Li*E.Honda*Evil Ryu
    Abel*Ibuki*Dhalsim
    Gouken*Zangief*Ryu

  3. #63
    Plane Touched BigTerj's Avatar
    Join Date
    Aug 2012
    Posts
    286

    Default

    Quote Originally Posted by Violacea View Post
    I only find this reflectively interesting because the 2 current games I hit hard (qlive and sfiv) both have rating systems now. In the 90's they didn't. And you know what? I can't say the rating system has even drastically changed the experience and the challenges I do or don't face. So again, food for thought - don't put so much end all trust in these rating systems and take it out on devs when you lose a game that you perceived should have been won. Ratings don't mean anything *should* happen.
    I agree, the ratings system and matchmaking are not all that helpful. I think that a lot of companies do it in order to buffer the "new" players so as to not turn them away from the game, giving them a nice warm fuzzy about PvP and PvP progression.

    Wins/loss is an interesting beast because it doesn't take into account a person's gear. A newly minted 65 PvP badass (while he was in the low lvl brackets) vs a mediocre dude geared with full Dreambreaker, who do you think will win the 1v1 lol? That new 65 may have a way higher ELO rating because of all their low level wins, but is expected to be able to go head to head against some dude in raid gear, it's not exactly a level playing field. Yes, I know there is some bolstering, but its not going to buff a person all that much (when it works properly). Perhaps the carry over ove ELO from low level PvP to 65 PvP is cause for some issues since gear bolstering is currently broken.
    Last edited by BigTerj; 04-07-2015 at 06:43 PM.
    Bigterj - <Grievance> - Wolfsbane

  4. #64
    Ascendant Violacea's Avatar
    Join Date
    May 2011
    Posts
    12,128

    Default

    Well, my ELO is 2901. From 2 days before 3.0. 90%+ solo queuing as a Bladedancer most likely.

    Fear it.
    Last edited by Violacea; 04-07-2015 at 06:50 PM.
    New round up of some high rank matches
    Chun-Li*E.Honda*Evil Ryu
    Abel*Ibuki*Dhalsim
    Gouken*Zangief*Ryu

  5. #65
    Ascendant Violacea's Avatar
    Join Date
    May 2011
    Posts
    12,128

    Default

    Quote Originally Posted by BigTerj View Post
    I agree, the ratings system and matchmaking are not all that helpful. I think that a lot of companies do it in order to buffer the "new" players so as to not turn them away from the game, giving them a nice warm fuzzy about PvP and PvP progression.

    Wins/loss is an interesting beast because it doesn't take into account a person's gear. A newly minted 65 PvP badass (while he was in the low lvl brackets) vs a mediocre dude geared with full Dreambreaker, who do you think will win the 1v1 lol? That new 65 may have a way higher ELO rating because of all their low level wins, but is expected to be able to go head to head against some dude in raid gear, it's not exactly a level playing field. Yes, I know there is some bolstering, but its not going to buff a person all that much (when it works properly). Perhaps the carry over ove ELO from low level PvP to 65 PvP is cause for some issues since gear bolstering is currently broken.
    What you described is the flaw I've noticed with my SF game. Arcade Edition and Ultra are two separate games entirely even though the only difference is slight character changes and 5 new players/maps to use. You can tell a true new player vs a Arcade Edition vet who finally picked up Ultra. It's just never a tell all.

    I mean, some people even have youtube videos where they make troll accounts so they look like a new player, when in reality they are a beast. I never trust numbers, I wait till the fight starts .
    Last edited by Violacea; 04-07-2015 at 06:53 PM.
    New round up of some high rank matches
    Chun-Li*E.Honda*Evil Ryu
    Abel*Ibuki*Dhalsim
    Gouken*Zangief*Ryu

  6. #66
    Ascendant
    Join Date
    Dec 2011
    Posts
    2,357

    Default

    Quote Originally Posted by Xirxe View Post
    as bad as the current ELO rating is, there is no easy fix to it.
    despite all the ELO ratings a big factor in wfs is the mix of roles. a team with no or next to no healers is quite handicaped. should it be forbidden to change roles in wfs, the number of healers in each team at least could be even.
    It seems to me that I've won as many matches as my team's sole healer as I have as one of three. The factor that always stands out to me is the quality of the DPS. If I have good DPS on my team, they focus the enemy DPS down quickly and I have an easier time healing. If my team's DPS is bad, they kill nothing and even three healers are not enough to keep the team up.

    Locking roles would have no effect on the quality of the DPS - nor the quality of the healers - and would deny the players the agency to adjust their team's composition. It is a a terrible idea that needs to die in a fire.

    Quote Originally Posted by Ianto Jones View Post
    They don't wait for the wipe or first engagement to be a loss. I get teamed with people who are in the match, at their keyboards, they're on the battlefield, but they are doing nothing. It's like they don't even know how to play the game. They just stand behind the rest of the team, often in a cluster, hiding from taking damage because they're afraid to die. They stick with the team but don't participate; they don't split off and do objectives; they don't switch specs to suit the team's needs.

    You can see them on the scoreboard at the end of the match, they'll be at or near 0 in all categories, but I know that they were there because I'm healing and I see them in my LOS on the raid frames constantly. It's infuriating. Like, how the **** do you fix that?
    Suggestion: Participation Score. A rather simple score that does nothing but measure how active a player was in a warfront. Every player's score starts at 150.

    At the end of the battle, each player will be judged for 'damage done', 'healing done', or damage taken' based on which of those stats are highest for the player.

    DPS would have their total 'damage done' compared to the average 'damage done' (determined by adding the damage done of all players - including healers - and dividing that by the number of players). Those who have a higher 'damage done' than the average get +1 to their participation score. Those who do not get -1 to their participation score.

    Those who scored highest in healing would have to beat the 'average healing done'. And those who scored highest in damage taken would have to beat the average damage taken. Those who soul recall or quit the match before the end get -10 to their participation score.

    Every time a new match is being made, the match-making system would pull those with the highest Participation Scores first, and match them by ELO. The players remaining in the pool would form another match, again chosen by Participation Score first and balanced by ELO. The Participation Score would not be used to balance matches, but to create better match-ups by placing active players with active players and pooling less active players with the less active players.

    This would not really be a measure of skill, since any decent active player would be getting +1 to his participation score at the end of every battle, to the point that it would hit the thousands and keep on going. It would, however, serve the worthy purpose of keeping AFKers and serial quitters out of the matches of the more competitive players.

    Afkers and quitter who find themselves at sub-150 scores would be placed with other Afkers and quitters, and could climb out of their low ranking by simply participating. Of course, there are some who will game the system to give themselves a low score and play well when it matters in order to get their weeklies and daily warfront charges done, but so long as they are not making a mess of my warfront matches, I couldn't care less.

  7. #67
    Ascendant Ianto Jones's Avatar
    Join Date
    Feb 2011
    Posts
    6,018

    Default

    I'd like to see a simple APM column on the WF scoreboard. It's not perfect (none of the metrics are) but it'd give you an idea of the people who are not actively participating. They could probably find an easy way to incorporate it into matchmaking, too.

    The system is just designed to get you to win 50% of the time regardless of your personal performance. I'm pretty sure this was meant to stop new players to the game from having overly-negative experiences, and being turned off from PvP altogether, but the flip side of that is that it actively punishes knowledgeable veterans who tend to be better at the game -- and they're at least as necessary as the fresh blood.

    This has to change. That simple.

  8. #68
    Plane Walker
    Join Date
    Nov 2014
    Posts
    404

    Default

    So I just made a sarcastic post in HK chat saying "Good PvPer looking for 2 players with low ELO so I can get my guns for hire done." With in 30 seconds I had 4 people respond to me saying they will group with me.

    How much more of this will we see when ELO becomes public?

  9. #69
    Plane Walker regall's Avatar
    Join Date
    Aug 2012
    Posts
    459

    Default

    Quote Originally Posted by Hailolyergory View Post
    So I just made a sarcastic post in HK chat saying "Good PvPer looking for 2 players with low ELO so I can get my guns for hire done." With in 30 seconds I had 4 people respond to me saying they will group with me.

    How much more of this will we see when ELO becomes public?
    Did you get your quest done with them?

  10. #70
    Shadowlander
    Join Date
    Mar 2015
    Posts
    33

    Default

    Quote Originally Posted by Hailolyergory View Post
    So I just made a sarcastic post in HK chat saying "Good PvPer looking for 2 players with low ELO so I can get my guns for hire done." With in 30 seconds I had 4 people respond to me saying they will group with me.

    How much more of this will we see when ELO becomes public?
    I would imagine we'll see quite a bit of this, but really it's not a huge deal. It would depend on how quickly they're ELO rating adjusts, but essentially by doing this they are trading short term gain for long term pain. Their ELOs are going to go up if you carry them, thus they will eventually not be able to game the system at some point (since gaming it relies on them having low ELOs) and they will have ELOs that aren't reflective of their skill, which will most likely result in lengthy losing streaks. The same would apply to the person doing the carrying. Of course, this all relies on the ELO system actually doing what it's supposed to do.

    If it became widespread we would just see matches where the skill level between playeres in a given warfront would vary quit a bit since it would be likely that both teams would have people trying to game the system. I.E. both teams having 3 players with very high ELOs and 12 players with very low ELOs which in theory could actually be an even match.

  11. #71
    Ascendant Violacea's Avatar
    Join Date
    May 2011
    Posts
    12,128

    Default

    Quote Originally Posted by Hailolyergory View Post
    So I just made a sarcastic post in HK chat saying "Good PvPer looking for 2 players with low ELO so I can get my guns for hire done." With in 30 seconds I had 4 people respond to me saying they will group with me.

    How much more of this will we see when ELO becomes public?
    Why would people tie weights to their feet and jump in the water?
    New round up of some high rank matches
    Chun-Li*E.Honda*Evil Ryu
    Abel*Ibuki*Dhalsim
    Gouken*Zangief*Ryu

  12. #72
    Ascendant butkus's Avatar
    Join Date
    Oct 2011
    Posts
    2,008

    Default

    Quote Originally Posted by ecru View Post
    losing streaks of 10+ is why i quit. cancelled my patron sub, quit raiding, quit pvping, haven't logged on in weeks. the last 3-4 weeks i played, my win/loss rate was roughly 1:4-1:5 solo queueing, but my ELO on pts is still 3600.

    maybe you played rift before you instituted this ridiculous matchmaking system or maybe you didn't so you may or may not know how much better it was before you messed everything up by overcomplicating it, but the fact of the matter is that you are punishing players for playing well and they are quitting your game because of it.

    the reward for being good at rift is eating **** in 9 out of 10 warfronts. why?
    why?who else is going to carry the pvers in wfs so they can get their pve marks.wfs are just a tool for pvers to get geared up so they can get the hit they need to get into raids.

    on live i have never seen so many players just sitting around afking in wfs since the day i started playing.in the past myself and players that i played with went into wfs to PLAY,we did not sit around.we enjoyed pvping in rift so we choose to run wfs/dailys ect. not now though,now they have forced pvers into wfs that do not want to be there.their main goal is to get raids marks to they can buy raid gear to raid with,thats it.this whole elo thing is just a way to try and cover up the flaws with rifts pvp.

    at some point someone at trion has to say wtf, this new system we made seems to have created more problems then what its solved.maybe having to have two gear sets was not so bad after all.
    Last edited by butkus; 04-08-2015 at 05:13 AM.
    I did not invent Warladin I perfected it!
    rifts #1 warlord!
    rifts #1 pull tank!
    Your weakness builds me!
    Butkus@seastone

  13. #73
    Ascendant butkus's Avatar
    Join Date
    Oct 2011
    Posts
    2,008

    Default

    Quote Originally Posted by Nefelia View Post
    It seems to me that I've won as many matches as my team's sole healer as I have as one of three. The factor that always stands out to me is the quality of the DPS. If I have good DPS on my team, they focus the enemy DPS down quickly and I have an easier time healing. If my team's DPS is bad, they kill nothing and even three healers are not enough to keep the team up.

    Locking roles would have no effect on the quality of the DPS - nor the quality of the healers - and would deny the players the agency to adjust their team's composition. It is a a terrible idea that needs to die in a fire.



    Suggestion: Participation Score. A rather simple score that does nothing but measure how active a player was in a warfront. Every player's score starts at 150.

    At the end of the battle, each player will be judged for 'damage done', 'healing done', or damage taken' based on which of those stats are highest for the player.

    DPS would have their total 'damage done' compared to the average 'damage done' (determined by adding the damage done of all players - including healers - and dividing that by the number of players). Those who have a higher 'damage done' than the average get +1 to their participation score. Those who do not get -1 to their participation score.

    Those who scored highest in healing would have to beat the 'average healing done'. And those who scored highest in damage taken would have to beat the average damage taken. Those who soul recall or quit the match before the end get -10 to their participation score.

    Every time a new match is being made, the match-making system would pull those with the highest Participation Scores first, and match them by ELO. The players remaining in the pool would form another match, again chosen by Participation Score first and balanced by ELO. The Participation Score would not be used to balance matches, but to create better match-ups by placing active players with active players and pooling less active players with the less active players.

    This would not really be a measure of skill, since any decent active player would be getting +1 to his participation score at the end of every battle, to the point that it would hit the thousands and keep on going. It would, however, serve the worthy purpose of keeping AFKers and serial quitters out of the matches of the more competitive players.

    Afkers and quitter who find themselves at sub-150 scores would be placed with other Afkers and quitters, and could climb out of their low ranking by simply participating. Of course, there are some who will game the system to give themselves a low score and play well when it matters in order to get their weeklies and daily warfront charges done, but so long as they are not making a mess of my warfront matches, I couldn't care less.
    you cant simply award players more points for more damage and healing done.what about the guy thats caps 10 stones in port but did very little damage compared to the clown that sat on the bridge so he could top the meters?
    Last edited by butkus; 04-08-2015 at 05:03 AM.
    I did not invent Warladin I perfected it!
    rifts #1 warlord!
    rifts #1 pull tank!
    Your weakness builds me!
    Butkus@seastone

  14. #74
    RIFT Fan Site Operator Riane's Avatar
    Join Date
    Apr 2011
    Posts
    5,612

    Default

    Quote Originally Posted by Nefelia View Post
    At the end of the battle, each player will be judged for 'damage done', 'healing done', or damage taken' based on which of those stats are highest for the player.
    You do realize that this is what the ELO used to be, right? It's flawed because you have people (me) stuck with getting really awful teams all the time because of high ELO.

  15. #75
    Champion of Telara
    Join Date
    Jul 2011
    Posts
    1,300

    Default

    just checked the pts to see what my ELO numbers were. It was discouraging. the highest score I had was on a level 10 that hadn't pvp'd yet, which was 1500.
    My main was 1/2 that.
    Maybe I need to bunnyhop more.
    Last edited by SmidgePlays; 04-08-2015 at 06:14 AM.
    Smidge
    Wolfsbane

+ Reply to Thread
Page 5 of 11 FirstFirst 1 2 3 4 5 6 7 8 9 ... LastLast

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts