Jump to content

Elo Working As Intended


88 replies to this topic

#21 Barnaby Jones

    Member

  • PipPipPipPipPipPip
  • Survivor
  • Survivor
  • 434 posts
  • LocationTexas

Posted 26 February 2013 - 12:30 AM

View Postharuko, on 25 February 2013 - 11:39 PM, said:

Dude...no offense, but are you drunk?



No hes just a troll. He doesn't know half of what hes talking about, and the other half he disagrees with.

If he can some how make it the fault of premades, he will twist and turn his "logic" to make it so. But dont you dare imply his favorite thing in the world right now (ELO) is broken, bcause it absolutely has to be something else thats at fault. Best to just not feed him.

#22 elbloom

    Member

  • PipPipPipPipPipPip
  • 413 posts

Posted 26 February 2013 - 12:33 AM

tonnage difference should be 10-15% at a MAXIMUM. this is silly like it is.....

reminder: this is no single player game anymore.

Edited by elbloom, 26 February 2013 - 12:34 AM.


#23 Rina Fujimoto

    Member

  • PipPipPipPipPipPipPip
  • Urban Commando
  • Urban Commando
  • 526 posts
  • LocationSF

Posted 26 February 2013 - 12:55 AM

View Postgunghoblazes, on 26 February 2013 - 12:16 AM, said:

Posted Image



cough cough

I don't want to sound condescending but you can't really compare that to the one in the opening post, considering 2 of those are Awesomes (the subpar 8V variant and an 8R which I'm assuming wasn't even an SRM 8R and some ******** LRM 8R), one's a ****** STK variant who did under 100 damage, a deaths knell, and a Dragon of all things, while having a kurita 4 man to help you out.

It's not like you went against 2 4 mans, 7 being composed of the most armored mech in the game, so it really isn't the same in my eyes.

#24 Riffleman

    Member

  • PipPipPipPipPipPipPip
  • 968 posts

Posted 26 February 2013 - 01:06 AM

The fact the matchmaker dident throw both those 4 mans against each other on opposite sides is the biggest load of crap. System sucks.

#25 gunghoblazes

    Member

  • PipPipPipPipPip
  • 137 posts
  • LocationEverywhere at once.

Posted 26 February 2013 - 01:10 AM

View Postharuko, on 26 February 2013 - 12:55 AM, said:

I don't want to sound condescending but you can't really compare that to the one in the opening post, considering 2 of those are Awesomes (the subpar 8V variant and an 8R which I'm assuming wasn't even an SRM 8R and some ******** LRM 8R), one's a ****** STK variant who did under 100 damage, a deaths knell, and a Dragon of all things, while having a kurita 4 man to help you out.

It's not like you went against 2 4 mans, 7 being composed of the most armored mech in the game, so it really isn't the same in my eyes.


lol

#26 gunghoblazes

    Member

  • PipPipPipPipPip
  • 137 posts
  • LocationEverywhere at once.

Posted 26 February 2013 - 01:25 AM

we were out tonned just as much as you were.

#27 Karl Split

    Member

  • PipPipPipPipPipPipPip
  • 727 posts

Posted 26 February 2013 - 01:32 AM

BWAHAHAHAHAHAH WORKING AS INTENDED ELO IS FINE I SEE NO ISSUES CARRY ON

I also notice the usual ELO and PGI apologists are out in force to defend it / redirect blame to the mythological team stack. Tried for a laugh team stacking the other night, as everyone had different ELO's lo and behold none of us could get into the same match, so yeah obviously team stacking and not a ****** matchmaker

Edited by Karl Split, 26 February 2013 - 01:41 AM.


#28 Shazarad

    Member

  • PipPipPipPipPipPipPip
  • Elite Founder
  • Elite Founder
  • 525 posts
  • LocationTexas

Posted 26 February 2013 - 01:39 AM

View PostLindonius, on 25 February 2013 - 11:54 PM, said:


This has been happening to me a lot recently in my Jenner or Hunchback. I get a couple of kills, do a sit check, see that I'm the only one left, manage to get a few more kills before I go down. Final score shows me and perhaps one other valiant soul with decent damage and kills at the top of our leaderboard, followed by 6 guys who couldn't hit a cow's backside with a banjo.


This happens to me all the damn time. ALL THE TIME! Before ELO and after.

#29 TheGreatNoNo

    Member

  • PipPipPipPipPipPip
  • Bridesmaid
  • Bridesmaid
  • 448 posts

Posted 26 February 2013 - 02:28 AM

Posted Image

I decided to try out a Jenner for once. The four other jenners ran ahead of the group, then all died. Match maker needs a fine tuning I`d say.

I was part of a group that had all the non-jenners save myself. The other jenners were pugs, or another group. Not sure.

Edited by Lost One, 26 February 2013 - 02:34 AM.


#30 xenoglyph

    Member

  • PipPipPipPipPipPipPipPip
  • 1,480 posts
  • LocationSan Diego

Posted 26 February 2013 - 02:38 AM

I actually had a pretty bad match right now.

My team was Atlas DC + Atlas D + 6 mediums

versus

3 Atlas DCs, an Awesome, 3 mediums and a commando

We didn't lose by too much, so in defense of ELO I guess you could say the results showed it was almost a fair matchup, but damn we had an uphill battle against that kind of tonnage disparity.

It doesn't seem to happen very often, but it happens.

#31 jakucha

    Member

  • PipPipPipPipPipPipPipPipPip
  • Knight Errant
  • 2,413 posts

Posted 26 February 2013 - 02:41 AM

It works nicely, but really does need tweaking. Still getting a few too many lopsided matches, though not nearly as many as before obviously.

#32 Leimrey

    Member

  • PipPipPipPipPip
  • 169 posts

Posted 26 February 2013 - 04:05 AM

View Postjakucha, on 26 February 2013 - 02:41 AM, said:

It works nicely, but really does need tweaking. Still getting a few too many lopsided matches, though not nearly as many as before obviously.

Hey, guys, Elo works, except it doesn't.

The whole point of Elo is to eliminate the occurrence of these "a few too many lopsided matches", which it fails to do. Honestly, I have no idea how it's supposed to work, but I do know that I should NOT be carrying assault players in my heavy or medium mechs, especially when these players are, presumably, as skilled (or horrible at the game, I won't be the judge of my own performance) as me. Why does the game fail to provide me with a team of my skill level if the Elo system works?

Posted Image

Ironically, I had more balanced solo matches when there was no Elo matchmaking, because I was at least sure that the enemy team will not have 2 extra assaults or lights, which can make or break a match. Feel free to laugh at this screenshot and dismiss it as irrelevant anecdotal evidence if you wish, I could make a whole archive of these. Seriously, I think I will be taking a long break from this game because I do not find it fun playing on a team with players who are significantly less skilled than the enemy team players over and over again.

Edited by Leimrey, 26 February 2013 - 11:06 AM.


#33 Lyteros

    Member

  • PipPipPipPipPipPip
  • 456 posts
  • LocationGermany

Posted 26 February 2013 - 09:44 AM

PGI ELO will not work much better then complete random drops because it calculates ELO purely on team win / loss, which consists of 8 random people with random mechs with random fittings. Using only the outcome of this and disregarding everything else just will not work...

It's like measuring your body hight but rolling some dice first to find out where you start the measurement, may it be the toe or the shoulder.

More toughts and explanation about those errors and why this ELO system will not work nearly as good as it could is here.

#34 Wrenchfarm

    Member

  • PipPipPipPipPipPipPipPip
  • Big Brother
  • Big Brother
  • 1,039 posts

Posted 26 February 2013 - 09:52 AM

View PostSlXSlXSlX, on 25 February 2013 - 11:37 PM, said:


Look, don't take it personally. Let me rephrase, my delivery can be barsh at times but give me a chance to clarfiy. Elaine thought you guys were closely paired, or it was the tail end of her desperately trying to match "someone" vs those two 4 mans. I don't mean you lacked in skill in this game in general, but in ratio to them, they did seem to have you outclassed to a certain extent.

I will give you this however, part of wot's mm does try to pair a heavy for a heavy, or at least a tier 9 for a tier 9, 3 arty for 3 arty, etc... so yes, a closer weight between teams will help, and I do want Elaine to be at her best. Her dancing after all, leaves a littlle to be desired. I am sure you personally are very good, and im not here to attack you, im here to day..... dont balme elaine, what we have here is not an example of her failing.. it is the result of 2 things.

1. A community too shallow to draw the 8 super pros needed to combat these stat padding elites (my guess is they knew their stuff, no?), and
2. the danger of having 4 mans which can overly influence individual pub matches due to their size in ratio to the entire team size.

Yet your title resounds in, this is Elaines fault, and you OP does as well. I think we can call this a spurrious relationship at best.

Grow the community, Elaine needs all the variables she can to throw into her formulas.

The community size and 4 man dominance issues, are closely connected. PGI, you have drug your feet on addressing this for a very long time. When will you finally do something about it?

20%.... 20%. Please, make it happen. 4/20 = 20% and 2/8 is close enough, 25%. Options exist, and I urge the devs t oconsider them.

.......

Cue founder army flaming me, whilst screaming lances are 4 guys, and canon this and that, and the other side of the argument that the devs jsut got 8v8 un-laggy, and only intend to go to 12 v 12, so too bad so sad right?

Well these kind of matches, while rarer now with Elaine, will continue to happen all to often as long as the dev team leaves the mechanics in place (4 mans) which allow it to happen.

I have no words.

#35 John MatriX82

    Member

  • PipPipPipPipPipPipPipPipPip
  • Philanthropist
  • Philanthropist
  • 2,398 posts
  • LocationItaly

Posted 26 February 2013 - 09:55 AM

This happens to me as well. As soon as I pilot my Atlases or stalkers I can still be able to win matches. As soon as I start using CTFs or Catapults or Dragons I see noticeable tonnage differences in the enemy team.

As soon as I move down and employ unskilled trebuchets (it's a lot of time that I don't take for a spin my HBKs or the Cicada) I usually end up paired to teams with 200 and more tons of me. Working as intended, I've been told by the support right today.

Therefore I think I'll have to stick with Assaults for the rest of my MWO playtime and never invest anymore in anything lighter.

#36 Mercules

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • The Nimble
  • The Nimble
  • 5,136 posts
  • LocationPlymouth, MN

Posted 26 February 2013 - 09:57 AM

http://mwomercs.com/...-making-update/

They are aware. The only way to get Elo working is to play matches so there is data to work off of.

#37 Rina Fujimoto

    Member

  • PipPipPipPipPipPipPip
  • Urban Commando
  • Urban Commando
  • 526 posts
  • LocationSF

Posted 26 February 2013 - 10:49 AM

View PostMercules, on 26 February 2013 - 09:57 AM, said:

http://mwomercs.com/...-making-update/

They are aware. The only way to get Elo working is to play matches so there is data to work off of.

I dunno, I've been playing a whole lot up until a day ago and my matches never seemed to get any better, the players I'd get matched up with were worse than randoms from the old system while the enemy team would usually be a swarm of cheese mechs with no sense of weight cohesion or anything.

It's also why I quit MWO for the time being, because it's no fun trying to take my new 2X out for a spin only to be matched up with 4 splat cats despite being the only heavy on our team.

It's like getting a moldy crappy sandwich at a restaurant, complaining about it, and then they tell you
"Keep eating it! We promise it'll taste better eventually! Maybe!"

#38 Mercules

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • The Nimble
  • The Nimble
  • 5,136 posts
  • LocationPlymouth, MN

Posted 26 February 2013 - 10:51 AM

Except Elo isn't going to get better if you don't give it data to place you in your proper spot.

#39 Rina Fujimoto

    Member

  • PipPipPipPipPipPipPip
  • Urban Commando
  • Urban Commando
  • 526 posts
  • LocationSF

Posted 26 February 2013 - 10:53 AM

View PostMercules, on 26 February 2013 - 10:51 AM, said:

Except Elo isn't going to get better if you don't give it data to place you in your proper spot.


View PostMercules, on 26 February 2013 - 10:51 AM, said:

Except Elo isn't going to get better if you don't give it data to place you in your proper spot.

Our scores were already supposed to be pre-seeded at the beginning of this month before they ever even added elo weren't they?

Edited by haruko, 26 February 2013 - 10:53 AM.


#40 Leimrey

    Member

  • PipPipPipPipPip
  • 169 posts

Posted 26 February 2013 - 11:02 AM

View PostMercules, on 26 February 2013 - 10:51 AM, said:

Except Elo isn't going to get better if you don't give it data to place you in your proper spot.

Except no amount of seeding will fix the weightclass disparity we're getting constantly, since the system apparently prioritizes player Elo score above the weight of the mech they are piloting, resulting in situations where 2 players of similar Elo may have to face each other while piloting mechs of absolutely different weightclasses.





1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users