Jump to content

Get Rid Of Elo-Based Matchmaking


69 replies to this topic

#61 MischiefSC

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • The Benefactor
  • The Benefactor
  • 16,697 posts

Posted 16 August 2014 - 04:10 PM

View PostKjudoon, on 16 August 2014 - 03:35 PM, said:


Complaining? No just stating historical record. You're the one inferring that I'm winning all the time so it makes it easy to say. I'm losing because of of the Banzai charges and precisely what you claim is your flaw. Too many people practicing the 'take you with me' style instead of being smart and trying to survive the match. I mean, come on... listen to an expert on aggressive warfare, George Patton: "The object is not die for your country, but to make the other poor dumb B* die for his!"

That's what I do every time. Those who don't cost their teams games just about every single time unless inordinately lucky.


Define "Skill" first." If skill is pavlovian twitch instinct, it's not skill. Bad mechs hurt all players. Good players can overcome the flaws of bad mechs. Good mechs make bad players only okay, unless they have REAL skill, not just good reflexes. Good mechs make good players great because they can use all the benefits.


Again, here's the thing.

Aggressive like that in pugs is how top tier players play. I don't say that because I'm a top tier player; I'm not. I am regularly fodder for them though and always, *always* take time to spectate them and see what build they take and what they play.

They play aggressively, almost never defensively. They'll take a defensive position *briefly* while their team settles or if there's an enemy push coming they need to blunt. It's not a 'Banzai' charge. It's going and killing the enemy. It's what wins and you can look over the performance of just about anyone on the top 5 of the contest or in competitive teams and see that. Exception to that was poptart play, however you'll notice that solid poptarts don't just go camp some spot and wait to get rolled. They move into a good firing position, they shoot until they can move up. They're always moving forward, always pushing.

If you're pushing the other team is falling back or dying.

That's what wins. It's how people who win consistently play. That's not an opinion. Go pick anyone you see on the forums who's a top tier player. Click their profile, look at their posts, scroll until you see one about them talking general tactics. Read what they wrote.

Skill? Skill is winning. It's whatever wins the match. That's skill in this sort of game. Good players can rock 6 kills and 6 assists plus 1k damage in a QD - and do it repeatedly while leveling basics. Bad players die in a Timber Wolf SRM/CERML brawler with less than 100 damage. Skilled players roll a Kit Fox and keep their whole team from taking less than 100 total points of damage from the enemy LRMs on Caustic, run off lights and while only doing 120 pts of damage they tagged or NARCed half the enemy team and kept the steel rain off their team the whole match without wandering. They will win a lot more than they lose and that translates into a higher Elo. A skilled player rolls his D-DC middle of his teams pack, communicates with everyone the whole match. He loads modules that boost his teams performance like AMS buffs or sensor range boosts. When the time comes for the push he takes point, lays a quick concise plan and plows the mother ******* road. He tanks when he needs to and stays out of long range fire. When time to brawl comes he sticks and swings, humps face and crushes dreams. Skill is the LRM boat who brings TAG every match and stays in the back of the firing line but present. When his team starts to relaly get hammered he moves up, using the lasers he ALWAYS brings and takes damage to keep teammates alive and shooting. It's the Medium pilot who reads the play and knows when to flank and break pressure, he knows when to distract, when to pull back. He protects his teams assaults and runs the length of the line providing fire support as needed. He provides a safe place for his teams lights to run to that's always nearby if they get chased by several enemy lights/mediums.

That is skill. It's what wins, not what gets damage or kills. That's worthwhile and it doesn't care about tonnage. Sure, good mechs are better than bad mechs but good players are what decides matches.

View PostSaxie, on 16 August 2014 - 03:44 PM, said:



Exactly you can roll some derp build, and roll an abhorrently low BV. BV does not account for skill. Well I guess it could, but then again you're looking at a more complicated match system.


Long range ideal - Elo modified by weapon and chassis. You do good with AC20? Your Elo goes up a bit every time you've got one in. You rock the 3D but not the 4X? Your 3D has a higher Elo than the 4X. You are a fiend with NARC? When you drop with NARC *and* your team includes 2+ LRM boats your Elo is measured higher.

Lot of work though and relatively minimal payoff in the current population count.

#62 Mergatroid Skittle

    Member

  • PipPipPip
  • Survivor
  • 65 posts

Posted 16 August 2014 - 09:11 PM

View PostKiiyor, on 15 August 2014 - 11:32 PM, said:

I just want closer games. Things have gone downhill since 4x3.

Posted Image


This is nice data... but it doesn't seem to have anything to do with ELO-based matchmaking. Do you have anything like this pre vs post ELO-based matchmaking?

I just want closer games too and I keep hearing people complain about the matchmaker. Seems to be a common theme in the forums.

#63 Kiiyor

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Big Daddy
  • Big Daddy
  • 5,565 posts
  • LocationSCIENCE.

Posted 16 August 2014 - 09:39 PM

View PostMergatroid Skittle, on 16 August 2014 - 09:11 PM, said:


This is nice data... but it doesn't seem to have anything to do with ELO-based matchmaking. Do you have anything like this pre vs post ELO-based matchmaking?

I just want closer games too and I keep hearing people complain about the matchmaker. Seems to be a common theme in the forums.


The only way to get ELO data is from the devs, but we can make guesses. I've got match results from around 14 people. They seem to have a fairly decent spread over the different ELO brackets, just by looking at individual win/loss rates, KDR etc. and making some assumptions.

The thing about ELO though, is that being in a higher/lower ELO bracket doesn't necessarily mean you'll be in better teams. The MM doesn't fill teams with people around your ELO level - it just ensures that each player has someone at their ELO on the opposite team. A match may have a handful of high ELO players, quite a few in the middle, and a (C) mech or two.

This means that each and every game has a good chance of representing the whole spectrum of ELO.

#64 MischiefSC

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • The Benefactor
  • The Benefactor
  • 16,697 posts

Posted 16 August 2014 - 10:20 PM

View PostKiiyor, on 16 August 2014 - 09:39 PM, said:


The only way to get ELO data is from the devs, but we can make guesses. I've got match results from around 14 people. They seem to have a fairly decent spread over the different ELO brackets, just by looking at individual win/loss rates, KDR etc. and making some assumptions.

The thing about ELO though, is that being in a higher/lower ELO bracket doesn't necessarily mean you'll be in better teams. The MM doesn't fill teams with people around your ELO level - it just ensures that each player has someone at their ELO on the opposite team. A match may have a handful of high ELO players, quite a few in the middle, and a (C) mech or two.

This means that each and every game has a good chance of representing the whole spectrum of ELO.


Well, the MM tries to fit you in a range.

The problem is that the higher your Elo the harder it is to find people in your range with the correct tonnage.

However, remember - regardless of your Elo you're going to have some matches you're likely to win and some you're likely to lose. Even someone with a 2800 Elo is going to drop in matches with 2 other people as good as him in it and several that are as close as possible.

The idea that everyone drops with derps every match is just ego-fapping. We want to believe we're SOOOO GOOOD that the mm has to sandbag us every match.

Just ain't so.

#65 Funkadelic Mayhem

    Member

  • PipPipPipPipPipPipPipPip
  • Knight Errant
  • 1,811 posts
  • LocationOrokin Void

Posted 17 August 2014 - 12:07 AM

View PostMischiefSC, on 16 August 2014 - 10:20 PM, said:


Well, the MM tries to fit you in a range.

The problem is that the higher your Elo the harder it is to find people in your range with the correct tonnage.

However, remember - regardless of your Elo you're going to have some matches you're likely to win and some you're likely to lose. Even someone with a 2800 Elo is going to drop in matches with 2 other people as good as him in it and several that are as close as possible.

The idea that everyone drops with derps every match is just ego-fapping. We want to believe we're SOOOO GOOOD that the mm has to sandbag us every match.

Just ain't so.

All your white knighting is B$!!! The MM/Elo system is BROKE!

Below is straight from Bryan Eckman's Twitter

Posted Image
Posted Image
Posted Image

Edited by Funkadelic Mayhem, 17 August 2014 - 12:12 AM.


#66 MischiefSC

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • The Benefactor
  • The Benefactor
  • 16,697 posts

Posted 17 August 2014 - 12:23 AM

View PostFunkadelic Mayhem, on 17 August 2014 - 12:07 AM, said:

All your white knighting is B$!!! The MM/Elo system is BROKE!

Below is straight from Bryan Eckman's Twitter

Posted Image
Posted Image
Posted Image


So, variance is 37 to 127 as a max, averages will fall 40-90.

That's all hugely within variance. You'll shift your Elo by ~25 points every match you win/lose that affects it. Can be as much as a 50 pt swing.

So all of that's within 2 or 3 matches. Your Elo probably varies by more than 127 on any given night.

That's why he showed us the picture - anyone who understood it would realize what it meant.

That's clearly not you.

It also proves you've never even looked over the post on Elo or how it works, as this was covered.

It's normal, natural and correct. That's what it's supposed to do.

So teams vary by 37 to 127 and individuals within a team are only off by 100-300. That's not a ton - that's not noobs and elites.

Noobs are about 1,000. Elite is 1800 and up.

If you'd read anything posted prior you'd know that.

You haven't, so you don't, so you're wrong.

Again.

Edited by MischiefSC, 17 August 2014 - 12:36 AM.


#67 ArmandTulsen

    Member

  • PipPipPipPipPipPipPipPip
  • Ace Of Spades
  • 1,184 posts

Posted 17 August 2014 - 02:16 AM

View PostMischiefSC, on 15 August 2014 - 11:25 PM, said:



I miss having a 90%+ win/loss and a KDR of 3 or 4 with pretty much any mech I wanted to take, because dropping in a 4man even without VOIP was like driving a steamroller through a herd of ducklings.

It's alright that you miss that. It was just bad for new and old player alike otherwise though.


He said drop Elo matchmaking, not re-introduce 4-man's in the solo queue.

Dafuq....

Edited by ArmandTulsen, 17 August 2014 - 02:16 AM.


#68 Kjudoon

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Rage
  • Rage
  • 7,636 posts
  • LocationWisconsin

Posted 17 August 2014 - 02:26 AM

While I'm glad to see those pictures, the group queue is the one that scares me. Look at the min/avg/max elo snapshot. Average team elo difference is 99 with the max being 499???

There's your 12-0 roflstomps that happen in under 4 minutes.

If 25 points is a big deal, a variation of 499 is intolerably bad.

I state again, large groups hide bad elo matches. Return to us a third queue for 1-4 man groups like we used to have. Solo players can opt in while the large group queues, allow 4 man lances to opt in. Problem reduced.

View PostArmandTulsen, on 17 August 2014 - 02:16 AM, said:

He said drop Elo matchmaking, not re-introduce 4-man's in the solo queue.


Absolutely!!! Do NOT change the solo queue by allowing anything other than solo players in. It's an expansion of the reason why Skirmish is so good for the community because the brawltards have their sandbox to play in and kill each other in giant mech smash frenzies and the rest of us... can stay the bleep away! Here's the mosh pit... now here's the rest of the concert for the sane people. Over there is the festival seating for large groups who don't care who they get stuck next to, now give us mid range "traditional" seating for those of us who want to have a chair, and not get slam danced into a piece of the stage.

Edited by Kjudoon, 17 August 2014 - 02:33 AM.


#69 Tarys

    Member

  • PipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 166 posts
  • LocationGermany

Posted 17 August 2014 - 03:19 AM

I think there is no ELO-Rating after all. The current tournament is a great thingk to take a look with/against what players you are playing. I had some folks from the top 10 of different weight-classes while i am just hitting some place around 100 in the light ranking (and have no idea how to get better without trying to spread my damage, steal kills and just stay behind and offer no scouting to my team).

So if there really is a rating either my ELO is high enough to play with these players (i sure the heck dont think so), the difference which ELO fights against/with each other is way too big or i am simply that unlucky guy that gets thrown into a battle where he normally aint good enough to stand a chance. In that case people should get a reward in form of extra xp/cbills because they shouldnt be there.

Edited by Tarys, 17 August 2014 - 03:39 AM.


#70 LordKnightFandragon

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • 7,239 posts

Posted 17 August 2014 - 11:33 AM

View PostMischiefSC, on 15 August 2014 - 11:16 PM, said:

No they don't agree with that. If they do they are wrong.

We had Elo-less matchmaking before. It was horrible.

We're never going back there.

Peace out.



And how was it?

The MM we need is tonnage or class based, not some magical WL/KD based bullshit. That is more flawed then anything else...

Put a hard cap on the mechs....12 mechs, 3/3/3/3....so it picks 3 of each weight class, regardless of the pilot's stats....and bam...a match.





1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users