Jump to content

Elo Issue? (Super High Elo Players)


205 replies to this topic

#101 Zeh

    Member

  • PipPipPipPipPipPip
  • 343 posts

Posted 14 March 2013 - 12:37 PM

View PostWarZ, on 14 March 2013 - 12:32 PM, said:


Did I not state clearly that I was not missile boating ? Read first.



With that in mind, please rethink your post.

Sorry, missed that. Then yes, that's quite amazing. Were you playing at a severe off-time, like NA morning? It could be at that time of day there were only 5 good players on, you and a 4-man premade on the other team. Were those matches all against different players? That looks like a crazy aberration to me. How's your match history? Mostly wins?

#102 RussianWolf

    Member

  • PipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 2,097 posts
  • LocationWV

Posted 14 March 2013 - 12:37 PM

were any of the 5 cap wins?

I had one of those last night. killed 4 ran 800+ damage. last man standing on my team and lost to a spider (a stalker doesn't win races). I won't put the loss on my team as they did what they were supposed to do (wore enemies down and killed them), I did what I was supposed to (wore enemies down and killed them), but we were out played. No dishonor in the loss and I think we all had fun.

#103 Corvus Antaka

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Knight Errant
  • Knight Errant
  • 8,310 posts
  • Twitch: Link
  • LocationInner Sphere

Posted 14 March 2013 - 12:41 PM

good. keeps thing interesting and drop times down. more challenge for the good players.

#104 Sprouticus

    Member

  • PipPipPipPipPipPipPipPipPip
  • Bridesmaid
  • 2,781 posts
  • LocationChicago, Il, USA

Posted 14 March 2013 - 12:42 PM

my win % on mediums was roughly 65%

my win rate on Heavies is 70.8% (LL cataphract mainly). As it happens, my hit % and % max dmage in LL's is best of all weapons. I find them the easiest weapons to use (even more than medium lasers)

I am 7 games over 500 (17-7). Assuming my heavy mech ELO started at 1300 (it probably didnt due to pre-seeding, but still) and assume an average of 25 points per game either way, (yes, I know it is impossible to really know Im just spitballing here), that is not nearly up to the top.


From what it sounds like here, (amongst a very select crowd, those who post on the boards) the win % for most of us runs around 60-70%. No one who has posted is less than 50%.


60% is not horrible I suppose, but frankly anything below that for a large group means that there is another large group that is in the <40% win range. Which has to suck.


I don't know if Elo uses low Elo players to balance high Elo players. I hope not. It goes against the whole idea of being match against equal talent.

I will say that if you ARE in a 4 man and find yourself getting matches with poor players, there ARE tactical options to minimize the impact of this. The biggest (and only one Im going to post) is that you should NOT let them be cannon fodder. Take the burden of being in front upon yourselves. Sure, sometimes they will scatter or hang back and you end up getting torn apart. But a few typed messages and all 4 players in your group acting in unison tends to corral the pugs into a group think mentality and you will often win in spite of your pugs. All you need is 1-2 kills to get ahead enough to make up for the bad pugs.

#105 Serapth

    Member

  • PipPipPipPipPipPipPipPip
  • 1,674 posts

Posted 14 March 2013 - 12:45 PM

View PostStaggerCheck, on 14 March 2013 - 11:34 AM, said:


When you say 'bad ole days', do you mean for yourself or your competion during the brutal PuGstomp timeframe? We were all seeing results exactly as you describe months ago before ELO. If you were on the winning team the majority of that time, which your elite ELO status would seem to dictate, many people reading the OP are not feeling too badly for your current situation.

Does it suck being grouped with players that do not have the same game understanding as you do? Yuppers. Did it suck for the people just getting started with the game to be matched up against a group of your friends for months and months? Yuppers. I imagine the matchmaker is working things out. Adapt to your current situation... all the Pubbies did to get to this point.



No, I pug almost 100%.

The bad ole days I'm referring to is back when the match maker would put a 4man premade on one side, and 4 trial mechs on yours.

For a while, once elo went in, live was grand... good skill, no trial mechs...

Now though, since last patch, its back to being awful about half the time.

#106 shabowie

    Member

  • PipPipPipPipPipPipPip
  • 877 posts

Posted 14 March 2013 - 12:46 PM

View PostFerretGR, on 14 March 2013 - 11:58 AM, said:

I can't speak to that. I assume it's something like a schoolyard pick based on closeness to whatever target value the system sets. One good way to do it would be to make the target value the average Elo rating of the group in your situation. My only point is that it's NOT done by averaging the Elo value of the team, which is the mechanism by which this whole &quot;matching 4 scrubs with 4 goods&quot; is supposed to be happening.


Not the average rating, the aggregate value of all players on the team. It has to try to balance that, even thought it still fails miserably at times.

Things I am still seeing with Elo:

1. Me and three other pretty good players drop with four complete tools vs 8 decent to good players on other team.
2. My group drops with another 4 man against unknown opposition.
3. My group drops with 4 pubs against 2 4 mans on other side.

Edited by shabowie, 14 March 2013 - 12:49 PM.


#107 Serapth

    Member

  • PipPipPipPipPipPipPipPip
  • 1,674 posts

Posted 14 March 2013 - 12:49 PM

View PostRussianWolf, on 14 March 2013 - 12:19 PM, said:

History of ELO

No ELO - "The competition is too varied. It's either 8-0 or 8-1 with teams rolling pugs every game."

New EO - "Wait times are too long. We want to play the game."

Revised ELO - "I'm being teamed with crap players because I'm a god."

Make up your minds.

This game does not have the population to pit 16 high ELO players against each other except on rare occasions. That's the fact. So you will be playing with players lower skilled. Given that noobs are placed in ELO in the middle to start with.

Be thankful there is a Mechwarrior game at all, and pray that it lasts long enough to see your "problems" gone.

Now, continue your crying.



See, thats the problem, you missed a stage. New ELO only sucked for two days... and that was a bug. They hotfixed it.

Then ELO was ******* wonderful. The game was as good as it has ever been. Great matches, many of them coming down to the last player or two.

Then... last patch, and now its just ****.

#108 FerretGR

    Member

  • PipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 1,445 posts

Posted 14 March 2013 - 12:52 PM

View Postshabowie, on 14 March 2013 - 12:46 PM, said:

Not the average rating, the aggregate value of all players on the team. It has to try to balance that, even thought it still fails miserably at times.

You're putting the chicken before the egg, or the other way around, or whatever. The matchmaker, according to the Command Chair post, doesn't look at the grouped players at all. It looks at a target value (how this is set or chosen I don't know, maybe the first player to drop into a game, it's a mystery), and finds players looking for games who are close to that value. I doubt that in its current, rudimentary form, it's good at distributing players evenly between teams once it's selected players for the game, which might explain what some folks are seeing here. Be patient, I guess ;)

Edited by FerretGR, 14 March 2013 - 12:53 PM.


#109 mekabuser

    Member

  • PipPipPipPipPipPipPipPip
  • Philanthropist
  • 1,846 posts

Posted 14 March 2013 - 12:52 PM

just wtf are you complaining about? OP./ Loose 70 % of what based on the stats u posted?

#110 shabowie

    Member

  • PipPipPipPipPipPipPip
  • 877 posts

Posted 14 March 2013 - 12:54 PM

View PostFerretGR, on 14 March 2013 - 12:52 PM, said:

The matchmaker, according to the Command Chair post, doesn't look at the grouped players at all.


It definitely does not say that.

#111 RiceyFighter

    Member

  • PipPipPipPipPipPipPip
  • Elite Founder
  • Elite Founder
  • 608 posts

Posted 14 March 2013 - 12:58 PM

View Postmekabuser, on 14 March 2013 - 12:52 PM, said:

just wtf are you complaining about? OP./ Loose 70 % of what based on the stats u posted?


I don't play groups all the time. Is just that isolated day where I played in 4 man.

#112 Kdogg788

    Member

  • PipPipPipPipPipPipPipPipPip
  • Elite Founder
  • Elite Founder
  • 2,314 posts

Posted 14 March 2013 - 01:11 PM

Reading the original post, it seems to me that ELO is working as it should. If they gave you better teammates, you'd probably win 90% of your matches if you're ratings are as high as you say, and that would be worse for the game. So enjoy the challenge if you guys are the elite of MWO.

-k

#113 FerretGR

    Member

  • PipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 1,445 posts

Posted 14 March 2013 - 01:13 PM

View Postshabowie, on 14 March 2013 - 12:54 PM, said:

It definitely does not say that.

"the match maker starts trying to make a match for an Elo of say 1300 and will pull in players to those teams closest to those values;"

It's not as explicit as the bit I responded about earlier, I'll give you that ;) but I read this as saying that the matchmaker is using only the target value and is pulling players in without consideration for any factor other than that value.

#114 shabowie

    Member

  • PipPipPipPipPipPipPip
  • 877 posts

Posted 14 March 2013 - 01:16 PM

View PostFerretGR, on 14 March 2013 - 01:13 PM, said:

It's not as explicit as the bit I responded about earlier, I'll give you that ;) but I read this as saying that the matchmaker is using only the target value and is pulling players in without consideration for any factor other than that value.


There is nothing there that says the target value is individual rather than an aggregate of a group dropping, and a team aggregate for general balance purposes. The fact it's always one team filling up more slowly supports my thinking, not yours.

It has to take into consideration the aggregate value of the team, otherwise every match will still be drastically lopsided (as it makes 8 individual potentially compounding "close enough" matchups that are off a bit, without balancing it all out in the aggregate) and it wouldn't make sense for good players to be seeing people who can barely control their mechs. Use a little common sense. The matchmaker can not individually match the scores of a drop that has 2 4 mans on one side, and a 4 man group on the other, which is still happening.

Edited by shabowie, 14 March 2013 - 01:25 PM.


#115 von Pilsner

    Member

  • PipPipPipPipPipPipPipPip
  • 1,043 posts
  • LocationColorado

Posted 14 March 2013 - 01:24 PM

From: http://mwomercs.com/...ost__p__1639451

Paul Inouye said:

CLARIFICATION:

I've been asked to clarify the Elo system above. It makes sense for a single player but not for teams. Does a player on 1 team get rated against a player that he beat on the other team?

Here's how a team works:

At the start of a match, all player's on ONE team have their Elo ratings totaled and divided by 8 (max players). You may realized that this is simply the AVERAGE of a team's total Elo. Team 1's average and Team 2's average are then used to calculate the probability of win (as per the formulas above). If Team 1 beats Team 2, then the appropriate math as above is applied to each player using the probability score calculated by the team averages.

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.
  • The more games you participate in, the more accurate the Match Maker becomes and you will start seeing that you are playing against people of relative equal skill.


It does not appear to take Elo score into account when forming a team, only when balancing that team with another... So it looks like the tryhards and the bads can end up on the same team.

#116 FerretGR

    Member

  • PipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 1,445 posts

Posted 14 March 2013 - 01:35 PM

View Postshabowie, on 14 March 2013 - 01:16 PM, said:

There is nothing there that says the target value is individual rather than an aggregate of a group dropping, and a team aggregate for general balance purposes.


Sure there is. The matchmaker starts with a target number, so the target number is determined before there's a team to aggregate. Granted, I don't know how the target number is determined, because it can't really be random, or it might not meet the needs of the online players. Perhaps the first player to be placed in a room determines the target number, in which case you're right, an aggregate of a foursome (group dropping) might very well be used if the foursome are the first occupants of the room. But I think the wording precludes the target number being determined after any matches have been made, which means players are probably not being picked for the opposite team based on a team aggregate. Again, perhaps I'm wrong, but I'd like to see evidence.


View Postshabowie, on 14 March 2013 - 01:16 PM, said:

It has to take into consideration the aggregate value of the team, otherwise every match will still be drastically lopsided (as it makes 8 individual potentially compounding "close enough" matchups that are off a bit, without balancing it all out in the aggregate)


Well, it should, but that doesn't necessarily mean that it is. Perhaps that's what's responsible for the mismatched games some players are complaining about. It doesn't mean every match will be lopsided, either, it just means the potential of such happening is there. If it's random, it's random, it could randomly assign everyone well, and it could randomly screw over one team.

No need to be all "common sense." I think I'm being reasonable.

ETA: these are good questions to be posed to the devs, IMHO: How is the target number determined? How are groups balanced, exactly, with examples, in the current implementation of Elo? How are players distributed by team once the matchmaker picks them for the game? None of this has been made clear.

Edited by FerretGR, 14 March 2013 - 01:42 PM.


#117 shabowie

    Member

  • PipPipPipPipPipPipPip
  • 877 posts

Posted 14 March 2013 - 02:06 PM

FerretGR, Von pilsner just posted something new that appears to prove you wrong.

#118 Comguard

    Member

  • PipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 652 posts
  • LocationBavaria, Germany

Posted 14 March 2013 - 02:08 PM

Ok, where do you see your ELO?

#119 Mazzyplz

    Member

  • PipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • 3,292 posts

Posted 14 March 2013 - 02:17 PM

i haven't read the whole thing, but i'll respond to OP directly.


no, ELO is not the problem, there is a problem for sure; and it's 4 man groups themselves.

at lower ELO brackets, 4 man have teamwork between the 4 of them and they can destroy low ELO players, just by 4 man maneuvers and ignoring the rest of their own team.


4 man in HIGH ELO brackets, are a complete liability for your team.

they way more often than not lose the match; and there's a simple reason; they don't cooperate with the 4 pugs they have.

that's it.
the root of the problem you're experiencing.

4 man's are UBER lame in high ELO matches. they get obliterated unless they work with their pugs.

pug commanders are frequently curbstomping these so called teams.
i talk from experience of about a hundred drops since ELO was implemented.
whenever i get a team i ask them to cooperate or just roll my eyes hard and expect a beating

#120 Khobai

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPipPip
  • Elite Founder
  • Elite Founder
  • 23,969 posts

Posted 14 March 2013 - 02:21 PM

pugs arnt real people. wonk.





5 user(s) are reading this topic

0 members, 5 guests, 0 anonymous users