Jump to content

Elo Worthless


298 replies to this topic

#201 Odin

    Member

  • PipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 498 posts

Posted 17 November 2013 - 06:33 AM

If ELO in MWO ever should be a plausible, true measure of your abilities, it will be based only on your personal performance during the match and NOT on its outcome.

LOL its THAT simple.

ELO based on win/loss is in fact a measure of the random groups (PUG) or premade drops, your a part of.
Since this game only offers team-death matches, lets ignore capping, there are only 2 factors to be considered:
  • Your kills
  • Your damage done

Currently there are no other measurable factors in game, which could be taken into account.

#202 Urdasein

    Member

  • PipPipPipPipPipPip
  • 237 posts

Posted 17 November 2013 - 07:11 AM

The problem with elo is the following: there is no reward being "good" because you will always be confronted to players to your level, so you will always be an average player, no matter if you play like {Scrap} or like god.

Without ELO, if you are good, you will have good K/D and good W/R.

The should be two kind of pickup: with ELO and without. Let the players choose..

#203 Roadbeer

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Elite Founder
  • 8,160 posts
  • LocationWazan, Zion Cluster

Posted 17 November 2013 - 08:20 AM

Allow me to ask, because while I have my data, I have dug through the NUMEROUS live threads on the topic and I keep seeing conflicting posts on how the system works. I *KNOW* how it is supposed to work in regards to chess and the BCS, but I can't seem to find the CC post on Elo.

1. Elo is increased based off of the Elo level of who you kill.: I remember that in an Elo post from Paul. At the time, this didn't make sense to me as if two players of equal Elo were slugging it out, and some scrub got the killshot, then the scrubs Elo would receive an undeserved increase.

2. Elo is based off of Win/Loss: [Citation Needed]

The first is what I remember Elo being from the original post on it from the Devs, the second is what I keep seeing thrown around.

If someone could point me to relevant threads on the topic, from the Developers, so I can solidify an opinion about what is "broken". It's odd because I see posts on the topic, from people I group with daily, stating a FAR different experience than my own.

#204 Sug

    Member

  • PipPipPipPipPipPipPipPipPip
  • The People's Hero
  • The People
  • 4,629 posts
  • LocationChicago

Posted 17 November 2013 - 08:49 AM

View PostOdin, on 17 November 2013 - 06:33 AM, said:

If ELO in MWO ever should be a plausible, true measure of your abilities, it will be based only on your personal performance during the match and NOT on its outcome.


Yes


View PostRoadbeer, on 17 November 2013 - 08:20 AM, said:

If someone could point me to relevant threads on the topic, from the Developers, so I can solidify an opinion about what is "broken". It's odd because I see posts on the topic, from people I group with daily, stating a FAR different experience than my own.


It's based only on wins. Team Elo is averaged out.


http://mwomercs.com/...79-matchmaking/

Scroll down. The post from Dec2012 explains how Elo works ...but only in the case of 1 player versus 1 player. Which MWO doesn't use. So it's a bit pointless.

But after that post is the one that explains the MWO system of Elo matchmaking and scoring.

Then they talk about how they had to widen the range of Elo scores in a match because solo players queues were too long. So worse people against better people : /

Edited by Sug, 17 November 2013 - 08:52 AM.


#205 Roadbeer

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Elite Founder
  • 8,160 posts
  • LocationWazan, Zion Cluster

Posted 17 November 2013 - 09:12 AM

Ok, had a fundamental misunderstanding on how it was working, for some reason I had "kill" in my head instead of "win".

I can see how, in theory, it would work, but it seems like they've stopped short of implementing a more definable strata where X Elo group should NEVER meet with Y Elo group. This can be argued that they just don't have the playerbase, but since nobody outside of PGI has those figures, just anecdotal evidence, we'll leave that out of the argument.

I've seen it said that new players had their starting Elo adjusted further down than, what would basically be the middle Elo rage, does anyone have that link?

#206 Odin

    Member

  • PipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 498 posts

Posted 17 November 2013 - 09:38 AM

Roadbeer, new players start at 1100, down from 1300 initially.
Much to high if you ask me. Why not starting at 0? Cos they don't have too many newbs? Perhaps. Waiting time for the match maker seem to be a big thing for PGI.

#207 Roadbeer

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Elite Founder
  • 8,160 posts
  • LocationWazan, Zion Cluster

Posted 17 November 2013 - 09:39 AM

View PostOdin, on 17 November 2013 - 09:38 AM, said:

Roadbeer, new players start at 1100, down from 1300 initially.
Much to high if you ask me. Why not starting at 0? Cos they don't have too many newbs? Perhaps. Waiting time for the match maker seem to be a big thing for PGI.


Agreed, too high, but I think 0 would be too much of a climb out of a REAL Elo hell... maybe 800.

#208 Roland

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • 8,260 posts

Posted 17 November 2013 - 09:50 AM

View PostGhogiel, on 17 November 2013 - 03:41 AM, said:

There is actually an Elo rating for each weight class.

Yeah, but it doesn't really work like that. Or at least, it doesn't actually result in tonnage balanced matches.

Ultimately, I suspect that it's mainly due to the fact that the game has continued to hemorage players, and doesn't have enough players left to support good matchmaking. And that tends to make the problem worse, by making PGI refuse to do necessary improvements to the matchmaker, under the argument that it would "split the player base".

#209 Ghogiel

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • CS 2021 Gold Champ
  • CS 2021 Gold Champ
  • 6,852 posts

Posted 17 November 2013 - 09:53 AM

View PostRoadbeer, on 17 November 2013 - 09:39 AM, said:


Agreed, too high, but I think 0 would be too much of a climb out of a REAL Elo hell... maybe 800.

Elo hell is in the middle. That's where the biggest divergence of skill and experience exists and assuming the threshold is wide enough can potentially pick players on either end of the bell curve.

At either end of the spectrum are the guys who play at a consistent measurable level.

View PostRoland, on 17 November 2013 - 09:50 AM, said:

Yeah, but it doesn't really work like that. Or at least, it doesn't actually result in tonnage balanced matches.

Ultimately, I suspect that it's mainly due to the fact that the game has continued to hemorage players, and doesn't have enough players left to support good matchmaking. And that tends to make the problem worse, by making PGI refuse to do necessary improvements to the matchmaker, under the argument that it would "split the player base".

It doesn't result in tonnage balances. But some good light pilot can be in a similar Elo band as some assault pilot. And once they get enough games, both those players have over time consistently affected the out come of their matches in their respective weight class.

Edited by Ghogiel, 17 November 2013 - 09:57 AM.


#210 Sug

    Member

  • PipPipPipPipPipPipPipPipPip
  • The People's Hero
  • The People
  • 4,629 posts
  • LocationChicago

Posted 17 November 2013 - 09:54 AM

View PostRoadbeer, on 17 November 2013 - 09:39 AM, said:

Agreed, too high, but I think 0 would be too much of a climb out of a REAL Elo hell... maybe 800.


It's more of an Elo Limbo.

The system really is pointless no matter how anyone explains or rationalizes it. Elo only works if it's used with individuals or individual teams that never change.

#211 Odin

    Member

  • PipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 498 posts

Posted 17 November 2013 - 09:55 AM

View PostRoadbeer, on 17 November 2013 - 09:39 AM, said:


Agreed, too high, but I think 0 would be too much of a climb out of a REAL Elo hell... maybe 800.


Well, I am unsure, but ELO Hell is supposed to be a too high ELO, no?
And too high on ELO means here, lots of premade drops in your record. Some, drop only with their group, 4 , 8, 12 ppl.
Your W/L is significantly better then. Voice com and training with partners. Once your starting to solo much, your ELO drops considerably.

I personally don't mind getting matched against better players, I know what I am doing, but the ELO-SHOTGUN newbs are a mess, nothing personal.

#212 Motroid

    Member

  • PipPipPipPipPipPip
  • 261 posts
  • Locationmost likely gone

Posted 17 November 2013 - 09:59 AM

http://mwomercs.com/...79-matchmaking/

This might be what you are looking for, Roadbeer.

#213 Odin

    Member

  • PipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 498 posts

Posted 17 November 2013 - 09:59 AM

View PostSug, on 17 November 2013 - 09:54 AM, said:


It's more of an Elo Limbo.

The system really is pointless no matter how anyone explains or rationalizes it. Elo only works if it's used with individuals or individual teams that never change.



If only players with exactly the same ELO would be matchedW/L or whatever, it would be(still) an improvement .

#214 Sug

    Member

  • PipPipPipPipPipPipPipPipPip
  • The People's Hero
  • The People
  • 4,629 posts
  • LocationChicago

Posted 17 November 2013 - 10:00 AM

View PostGhogiel, on 17 November 2013 - 09:53 AM, said:

Elo hell is in the middle. That's where the biggest divergence of skill and experience exists and assuming the threshold is wide enough can potentially pick players on either end of the bell curve.



Then again, the system is pointless since everything is random and averaged. We're purposely kept close to the middle.

Only insanely bad players or people working together on coms can significantly affect their Elo

Edited by Sug, 17 November 2013 - 10:01 AM.


#215 Roadbeer

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Elite Founder
  • 8,160 posts
  • LocationWazan, Zion Cluster

Posted 17 November 2013 - 10:02 AM

View PostSug, on 17 November 2013 - 09:54 AM, said:


It's more of an Elo Limbo.

The system really is pointless no matter how anyone explains or rationalizes it. Elo only works if it's used with individuals or individual teams that never change.


Oh, I couldn't agree more. It actually seems contrary to the long term goal of the game which is CW.
It's like the meme I put up on the first page of this thread.
Elo = Players matched by skill
Weight = Players matched by tonnage
Elo + Weight = a theoretical constant where W/L will be 50% +/-
-------------------------------------------------------------------------------------------------------------------------------------------------------
CW: Loyalists units fight on the battlefronts in an attempt to shift the battle lines for Merc units to be able to attack planets.
50%(ish) W/L = Battlelines never move.

#216 Sug

    Member

  • PipPipPipPipPipPipPipPipPip
  • The People's Hero
  • The People
  • 4,629 posts
  • LocationChicago

Posted 17 November 2013 - 10:06 AM

View PostOdin, on 17 November 2013 - 09:59 AM, said:

If only players with exactly the same ELO would be matchedW/L or whatever, it would be(still) an improvement .



Or at least closely matched.

There is a HUGE difference between a team of :

1800
1800
966
966
966

vs.

1300
1300
1300
1300
1300

But the matchmaker currently sees those teams as equals because the average Elo is 1300..

#217 Ghogiel

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • CS 2021 Gold Champ
  • CS 2021 Gold Champ
  • 6,852 posts

Posted 17 November 2013 - 10:07 AM

View PostSug, on 17 November 2013 - 10:00 AM, said:



Then again, the system is pointless since everything is random and averaged. We're purposely kept close to the middle.

Only insanely bad players or people working together on coms can significantly affect their Elo

Another one lol. Speak for yourself, I affect the games I play in every time I drop, for better or worse.

#218 Motroid

    Member

  • PipPipPipPipPipPip
  • 261 posts
  • Locationmost likely gone

Posted 17 November 2013 - 10:12 AM

View PostSug, on 17 November 2013 - 10:00 AM, said:



Then again, the system is pointless since everything is random and averaged. We're purposely kept close to the middle.

Only insanely bad players or people working together on coms can significantly affect their Elo

I think keeping us in the ELO middle is "working as intended".
Didn't someone say that the goal of MM is 50/50 w/l ratio for everyone?
If that is true, MM works for me because after more than 10000 drops (solo, 2-4 and 12) my w/l ratio is 5490/4538.
It has always been the same cycle. First a massive win streak followed by a massive loss streak. And so on.

#219 Odin

    Member

  • PipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 498 posts

Posted 17 November 2013 - 10:13 AM

Agreed Sug.

So, the problem is, we are so few? In closed beta we had around 2500 to maximal 5000 players in the count, as I remember.
I don't wont to violate NDA, hope this isn't one.

Well isn't the player base consistent?

#220 Sug

    Member

  • PipPipPipPipPipPipPipPipPip
  • The People's Hero
  • The People
  • 4,629 posts
  • LocationChicago

Posted 17 November 2013 - 10:14 AM

View PostSug, on 17 November 2013 - 10:00 AM, said:

Only insanely bad players or people working together on coms can significantly affect their Elo


View PostGhogiel, on 17 November 2013 - 10:07 AM, said:

Another one lol. Speak for yourself, I affect the games I play in every time I drop, for better or worse.


You can affect the outcome of a game, but eventually you will be averaged out with worse players, and are unable to significantly affect your Elo because you only change your Elo signifigantly by winning against a team you have low odds of beating, or losing to a team you had high odds of beating.


Paul Inouye said:


Some people also asked to have the description simplified. Here's the summary:
  • The Match Maker uses a scoring system to determine if your team is more likely to win or lose based on your team's average Elo rating.
  • If the Match Maker determines that you're going to lose, but you actually win, then your Elo score is going to go up and the enemy's score is going to go down.
  • If the Match Maker determines that you're going to win and you actually win, then your Elo score isn't going to change very much (if at all). The same applies to a prediction of loss and you actually lose, your score may drop but it will be slight.

  • Thus


    View PostSug, on 17 November 2013 - 10:00 AM, said:

    Only insanely bad players or people working together on coms can significantly affect their Elo

    Edited by Sug, 17 November 2013 - 10:15 AM.






    13 user(s) are reading this topic

    0 members, 13 guests, 0 anonymous users