Jump to content

Can We Get The Old Weight-Based Matchmaking Back?


37 replies to this topic

#1 Ninja Snarl

    Member

  • PipPipPipPipPipPip
  • 310 posts

Posted 18 March 2013 - 06:27 PM

In all honesty, the current matchmaker isn't worth squat. It's supposed to be balanced based on player skill, yet you get teams with all competent players against teams where half the players can't have been in MWO for more than a week. Not to mention the enormous tonnage and equipment imbalances; how in the world is it even worthwhile to implement a matchmaking system that pits multiple Atlases against a team with zero Atlases and only one or two assaults, and then gives that same side multiple ECM systems and the other none?

In other words, the ELO is worse than useless; it's an active detriment to the game. We'd be better off having the old system, where at least we don't have games become impossible to win thanks to the matchmaker giving one side a bunch of lights on Alpine or having the teams imbalanced by 150-200 tons.

#2 Zervziel

    Member

  • PipPipPipPipPipPipPip
  • The Marauder
  • The Marauder
  • 909 posts
  • LocationVan Zandt

Posted 18 March 2013 - 06:39 PM

I remember seeing someone post a pic where the enemy team literally had an entire assault lance whereas his team had two assaults and one had dced. Gotta love getting screwed.

Edited by Zervziel, 18 March 2013 - 06:39 PM.


#3 Ranek Blackstone

    Member

  • PipPipPipPipPipPipPip
  • 860 posts
  • LocationUSA

Posted 18 March 2013 - 06:41 PM

View PostZervziel, on 18 March 2013 - 06:39 PM, said:

I remember seeing someone post a pic where the enemy team literally had an entire assault lance whereas his team had two assaults and one had dced. Gotta love getting screwed.


I think I remember this pic. I know one dude posted getting 5 Atlas mechs, 2 heavies and a med against a team of 2 heavies 3 meds and 3 lights. The lighter team won.

#4 Ninja Snarl

    Member

  • PipPipPipPipPipPip
  • 310 posts

Posted 18 March 2013 - 06:47 PM

View PostRanek Blackstone, on 18 March 2013 - 06:41 PM, said:

I think I remember this pic. I know one dude posted getting 5 Atlas mechs, 2 heavies and a med against a team of 2 heavies 3 meds and 3 lights. The lighter team won.

That doesn't actually mean anything, though. If your team gets Alpine and you've got a mix of mediums, heavies, and assaults but no fast mediums or lights, the enemy team is almost guaranteed a victory if they have 2-3 lights. Conquest in general is easily ruined by the current useless matchmaking system, because it's entirely possible to get a highly mobile team facing off against a slow team; it doesn't matter if one side has 5 slow mechs left and the other side only one fast mech if the enemy resources are at 500 and they have 4 points capped. The match was a foregone conclusion before any actual fighting, simply because one side was randomly given the correct tools and the other side wasn't matched against it.

#5 Ryan Kryzanowski

    Member

  • Pip
  • 13 posts

Posted 18 March 2013 - 06:50 PM

I had 2 games with 5 Atlas DDC's tonight. That's real fun.

#6 Levi Porphyrogenitus

    Member

  • PipPipPipPipPipPipPipPipPip
  • Mercenary
  • Mercenary
  • 4,763 posts
  • LocationAurora, Indiana, USA, North America, Earth, Sol, Milky Way

Posted 18 March 2013 - 06:51 PM

My experience has generally been that when I'm in a group with much lower Elo players I tend to have wildly disparate team skill comps (usually they're terrible). When I solo-queue though it seems to be much better at finding roughly equivalent teams. I've had proportionally far more close matches in solo-queue since Elo than I did before.

Still, I would not object to them reworking the rate at which it widens the weight category restrictions. I'd also be a fan of them reworking how Elo is used to match people. I'd rather it do individual matches and then mash them together into teams, rather than do team matches based on average Elo for the group. I'd also support Elo tiers to speed up the Elo matching process.

#7 Wintersdark

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • 13,375 posts
  • Google+: Link
  • Twitter: Link
  • LocationCalgary, AB

Posted 18 March 2013 - 06:53 PM

https://plus.google....sts/hjvGypeRBkz

3 D-DC's, a 3L, a Splatcat, and a CTF, TBT and DRG

vs.

AWS-9M, AS7-D, Ilya, K2 (stock), C4. They also had a D-DC, but it DC'd before the battle started.

8v5, 4 ECM vs. none (and the poor C4 was an LRM boat), and 5 of the 8 man team were arguably the most powerful mechs in the game. Even without the CTF, TBT and DRG, that battle would have been grossly one-sided.

#8 Davers

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 9,886 posts
  • Facebook: Link
  • LocationCanada

Posted 18 March 2013 - 07:04 PM

I know everyone is venting, but you all do know that the Devs said Elo was going to be tuned before they even put it in, right?

#9 Belorion

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 5,469 posts
  • LocationEast Coast

Posted 18 March 2013 - 07:11 PM

I think the variable weight adds to the fun, plus I think it does do weight balancing to some degree. Just not the 1 - 1 it used to.

#10 MuonNeutrino

    Member

  • PipPipPipPipPipPip
  • 478 posts
  • LocationPlanet Earth, Sol System, Orion Arm, Milky Way Galaxy, Local Group, Virgo Supercluster

Posted 18 March 2013 - 08:23 PM

I just wish it would balance light mech numbers in conquest. That gametype can be SO destroyed by a bad matchup. Had two games today - one on caustic where the enemy team had 4 lights and we had nothing that moved faster than 70 kph, and another on alpine with a 4 ecm raven premade and a jenner on their side and a raven, spider and a bunch of slow mechs on ours (and yeah, that one would have been **** regardless, but having only 2 lights made it *even worse*).

#11 Zero Neutral

    Member

  • PipPipPipPipPipPipPipPip
  • Elite Founder
  • Elite Founder
  • 1,107 posts
  • LocationEast Coast USA

Posted 18 March 2013 - 08:30 PM

Progress is about moving forward not bringing back the old... the matchmaker system is still a work in progress not regress.

Edited by Zero Neutral, 18 March 2013 - 08:32 PM.


#12 SpartanFiredog317

    Member

  • PipPipPipPipPip
  • The Decimator
  • The Decimator
  • 176 posts
  • LocationMighty MO

Posted 18 March 2013 - 08:51 PM

I think they should implement matchmaker in a Role Specific way similar to the traditional MMO 5 man groupfinder and the way I remember Americas Army (the PC game) functioning. Standardize matchmaker loadouts by class and function. 2 lights, 2 mediums, 2 heavy, 2 Assault with 2 EWAR spots and 1 Command module per team. ...This would serve to keep matches far more balanced AND enhance the concept of team play and further define the different roles of the weightclasses. Would be a lot closer to how a fireteam functions on the battlefield.

Edited by SpartanFiredog317, 18 March 2013 - 08:52 PM.


#13 DaisuSaikoro Nagasawa

    Member

  • PipPipPipPipPipPipPip
  • Mercenary
  • Mercenary
  • 974 posts
  • Google+: Link
  • LocationTaipei, Taiwan

Posted 18 March 2013 - 08:56 PM

View PostMuonNeutrino, on 18 March 2013 - 08:23 PM, said:

I just wish it would balance light mech numbers in conquest. That gametype can be SO destroyed by a bad matchup. Had two games today - one on caustic where the enemy team had 4 lights and we had nothing that moved faster than 70 kph, and another on alpine with a 4 ecm raven premade and a jenner on their side and a raven, spider and a bunch of slow mechs on ours (and yeah, that one would have been **** regardless, but having only 2 lights made it *even worse*).


View PostMuonNeutrino, on 18 March 2013 - 08:23 PM, said:

I just wish it would balance light mech numbers in conquest. That gametype can be SO destroyed by a bad matchup. Had two games today - one on caustic where the enemy team had 4 lights and we had nothing that moved faster than 70 kph, and another on alpine with a 4 ecm raven premade and a jenner on their side and a raven, spider and a bunch of slow mechs on ours (and yeah, that one would have been **** regardless, but having only 2 lights made it *even worse*).


Splitting up into 3 groups of 2 or 3 and moving to and defending the separate capture points (you need 3 to win ultimately) wouldn't work?

#14 SpartanFiredog317

    Member

  • PipPipPipPipPip
  • The Decimator
  • The Decimator
  • 176 posts
  • LocationMighty MO

Posted 18 March 2013 - 08:57 PM

in an interesting twist, you could make it so the only person that could call Arty/Air would be the teammember with the command module, You could have it equiped on your mech and designate the target, but you'd have to be able to communicate with your lance commander and he'd have to press a hotkey to ok the strike (both you and your commander out of enemy ECM) ... would give another tactical element to the EWAR game.

#15 Kassatsu

    Member

  • PipPipPipPipPipPipPipPip
  • The Devoted
  • The Devoted
  • 1,078 posts
  • LocationColorado

Posted 18 March 2013 - 09:00 PM

View PostWintersdark, on 18 March 2013 - 06:53 PM, said:

https://plus.google....sts/hjvGypeRBkz

3 D-DC's, a 3L, a Splatcat, and a CTF, TBT and DRG

vs.

AWS-9M, AS7-D, Ilya, K2 (stock), C4. They also had a D-DC, but it DC'd before the battle started.

8v5, 4 ECM vs. none (and the poor C4 was an LRM boat), and 5 of the 8 man team were arguably the most powerful mechs in the game. Even without the CTF, TBT and DRG, that battle would have been grossly one-sided.


That's not even my worst match. My worst match had my team consist of everything 40-50 tons with one single Dragon and one lone 3L against a team of four D-DCs and a 3L, I don't even care to remember what the other three were. The 3L ran off and got himself killed in seconds and the rest of my team didn't stand a chance against the invisible wall of Atlas slowly creeping around the map.

#16 TehSBGX

    Member

  • PipPipPipPipPipPipPip
  • FP Veteran - Beta 1
  • FP Veteran - Beta 1
  • 911 posts

Posted 18 March 2013 - 09:09 PM

I've been playing like a week and a half, and I'm starting to admit Elo is wonky. This is what happens to me 4-5 games in a row where I get steamrolled then one game where everyone seems to be at the same level of skill as me.

On the Rare Occasions where Elo does it's job I tend to score in the middle of my team, I either get a kill or two or get a nice amount of assists I may not last the whole round But I actually accomplish something.

#17 CheezPanther

    Member

  • PipPipPipPipPip
  • 125 posts

Posted 18 March 2013 - 09:18 PM

TBH wieght matching prolly has better odds of giving u a balanced match than ELO....

ELO is broken and prolly more based on luck than anything else.. IE are u lucky to be the team with 3 or 4 3L's or D-DC's....

CAuse right now the way mech balance is and the number of OP builds around winning a match comes down to who has more of the OP builds than the other..

And of course lances of 3-4 3L's trumps the other teams lineup hands down..

#18 slayerkdm

    Member

  • PipPipPipPipPipPip
  • Elite Founder
  • Elite Founder
  • 395 posts

Posted 18 March 2013 - 09:22 PM

I think it will get better, I dont think going backwards is the best solution.

It will always have flaws, heck, some of it depends on what mech you use. For instance, I mostly still drive my HBK 4G, well, its win loss ratio is not too good, and its solo stats are well, unspectacular.

Then I roll out my Illya. Well, now I have a better mech, with what I presume is a low ELO and I just find myself having a field day. I played my HBK all night after coming back from vacation. Did well in it for what I think is a poor variant, getting a couple kills some games, no kills others, worst damage 98, best 493.

Then for grins I take out the Illya. 4 kills, 753 damage, lopsided match. My Illya's win/loss ratio is double my HBK.'s.

So same player, playing as well as I can each game, but with the superior mech, I am just wrecking things. I think its the combination of the mech and my low ELO from playing the HBK. I figure that my HBK is probably mostly weaker than most other mediums, so I am likely putting my team at a disadvantage from the start. I play my best, but the HBK,'s potential is likely lower than who they matched me up against. Now I run the Illya, Its likely my variant is stronger than what they match against me, and my team is at an advantage. So I have very good success with it, both personnaly, and for the team.

If I mostly ran my Illya, and then occasionally ran my HBK, I imagine I would find the HBK games, even more difficult to win and contribute effectively, as the game is looking at me, like I was in my Illya.

On whole though, I dont have a big problem with the MM. Sometimes its good, sometimes bad, but I find it interesting.

#19 valkyrie

    Member

  • PipPipPipPipPipPipPip
  • 508 posts

Posted 18 March 2013 - 09:37 PM

Supposedly ELO scores are by weight class, so for example I have a good Medium and Heavy ELO, while my Assault ELO is pretty much non-existent (I hate being slow, and have maybe 30 games in my D-DC, all of which were before ELO seeding began).

Unfortunately, this means two things:
1. When I run in my Hunchback, I'm probably going to be squared off against an equally skilled player, who is likely running a -3L, Splatcat, D-DC, or something similar to it
2. IF I step in my Atlas, the enemy team may very well be screwed since I know very well what I'm doing in it and have it set up as a proper D-DC, but I haven't played enough games in that class for ELO to match me properly.

These two things are issues on their own, but combined with the mass weight discrepancies and the issues that causes on Conquest? Yikes.

#20 Khobai

    Member

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

Posted 18 March 2013 - 09:53 PM

The problem isnt the ELO matchmatching. The problem is role warfare imbalance.

If a light mech and assault mech were both equal... it wouldnt matter how many of each got placed on your team at the start.

The reason it matters so much now, is because theyre not both equal.





1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users