Jump to content

To Lower Elo?


37 replies to this topic

#1 HiProfile

    Member

  • Pip
  • Survivor
  • 18 posts

Posted 30 January 2015 - 04:43 PM

The million dollar question: what can be done to lower your ELO, presuming it's high and matchmaker expects me to "carry harder"? I've been thinking about "playing" matches where my score ends at near-zero due to a loadout of a single small pulse laser. I'd use a fresh "patsy" mech to help track these stats, probably a Commando or Locust. Will this method lower the necessary stats to lower my ELO?

I'm having a serious problem with I can only describe as "ELO Hell". I desperately want normal "ELO Neutral" matches again. I now get put on teams where 1/3 of the players can't get over 100 damage and I get 5+ losses in a row. The Victor event gave the strongest visual cue (plus a 10-loss streak) & the Cataphract event reinforced the point - at no point were there similar winning streaks compared to losing streaks.

Before you ask if it's fair to the others in those games, consider this. I've had days where 3/4 of my first dozen matches could have had 10 disconnects on the enemy team w/o changing the outcome. It felt so "off" I started tracking this 3 months ago. Based on those stats & my profile stats (current vs archived), the formulas I pulled from my old college stats book are telling me something isn't right. It certainly feels nothing like the other FPS games I played where it was truly random - the server just filled an empty slot with the next available player.

#2 terrycloth

    Member

  • PipPipPipPipPipPipPip
  • Little Helper
  • Little Helper
  • 769 posts

Posted 30 January 2015 - 04:46 PM

Every loss should lower your Elo. It's the only thing that affects it.

#3 Wintersdark

    Member

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

Posted 30 January 2015 - 05:18 PM

You have different Elo scores in different weight classes. Your Elo in a locust is entirely unrelated to your Elo rating in a Catapult, for example.

Next, the Matchmaker doesn't put you with low Elo players and "expects you to carry harder", it puts you with as many people at your level that it can. This may or may not work out, depending on your Elo and what's available in the queue when you click launch.

The important part to realise here is that you are not special. The matchmaker doesn't build matches around you, it doesn't care about you at all. You're just a player with your particular Elo rating in whatever weight class you're in, no more or less important than anyone else in the match. It certainly doesn't construct matches just to screw with you.

#4 Agent 0 Fortune

    Member

  • PipPipPipPipPipPipPipPipPip
  • Bad Company
  • Bad Company
  • 3,403 posts

Posted 30 January 2015 - 06:14 PM

I realized this post contained a few mistakes, I refer you to CocoaJin's post which is more accurate.
http://mwomercs.com/...st__p__4151934]

Edited by Agent 0 Fortune, 31 January 2015 - 12:44 PM.


#5 HiProfile

    Member

  • Pip
  • Survivor
  • 18 posts

Posted 30 January 2015 - 07:43 PM

Your comment about weight classes may be true but still only filler words. With 3500 matches the ELO isn't going to be much different between classes for someone who regularly plays all 4. I've switched between classes on some evenings and met the same particular people - who also appear in different classes each time too. Speaking of multiple weight classes, the last 5 chassis I tried to level up (light, medium, heavy, and 2 assaults) had W:L ratios of HALF my current/archived W:L ratio after getting basic efficiencies. You may say sample size is not large enough, but that's 15 mech variants across >225 games. I finished one mech's basic in 6 decent matches, while the other two took 20 matches EACH. I can see a bad build getting crap EXP, but not cutting my W:L in half. HALF.

Regarding matching the most "like" players together, it's hard to believe that happens often given the s*** matches I've seen MULTIPLE TIMES IN A ROW. When less than 1/4 of the team (myself and 1-2 others) do 500+ dmg and at least HALF the team literally doing less than 150 each, I find it hard to believe that the MEAN ELO of all 24 players is even half what it is for the 2-3 "try-harders" on the losing team. In case you doubt, I have a screenie of 2 guys in the 600's and 5 guys doing jack dmg: 28, 32, 72, 73, 111, 141! I can see 1-2 assault mech getting caught alone, but nearly 2 full lances??

I'm not saying I'm special or that PGI agents are out to get me, however I would say matchmaker must be making an inordinate amount of players ALSO feel this way. I've felt this way for at least 3 months, if not longer.

#6 Felicitatem Parco

    Professor of Memetics

  • PipPipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 13,522 posts
  • LocationIs Being Obscured By ECM

Posted 30 January 2015 - 07:48 PM

To deliberately lower your Elo ranking, you have to lose on purpose. Please don't do this.

#7 That Dawg

    Member

  • PipPipPipPipPipPipPipPip
  • FP Veteran - Beta 1
  • 1,876 posts

Posted 30 January 2015 - 07:53 PM

Losing on purpose isn't good. Happens too much as it is.

Edited by That Dawg, 30 January 2015 - 07:54 PM.


#8 FerretGR

    Member

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

Posted 30 January 2015 - 08:03 PM

View PostHiProfile, on 30 January 2015 - 04:43 PM, said:

I now get put on teams where 1/3 of the players can't get over 100 damage and I get 5+ losses in a row.


So you're playing with subnormals and you think you have a HIGH ELO? mmkay.

There is a common factor in all of your losses.

#9 Tarl Cabot

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Tai-sho
  • Tai-sho
  • 7,659 posts
  • LocationImperial City, Luthien - Draconis Combine

Posted 30 January 2015 - 08:10 PM

View PostProsperity Park, on 30 January 2015 - 07:48 PM, said:

To deliberately lower your Elo ranking, you have to lose on purpose. Please don't do this.


The way your Elo drops is in two ways. Once both teams are put together, if your team's Elo is higher than the other side and is expected to win but loses then your Elo will drop lots. If your team is expected to lose and does lose, your Elo may drop a little, if at all.

#10 Mycrus

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Overlord
  • Overlord
  • 5,160 posts
  • Google+: Link
  • Twitter: Link
  • LocationFilipino @ Singapore

Posted 30 January 2015 - 08:11 PM

Use a new userid... problem solved

#11 Brody319

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • The Ominous
  • The Ominous
  • 6,273 posts

Posted 30 January 2015 - 08:12 PM

you can make people think your elo is garbage by using an unpainted champion.

that's how you lure the enemy into a false sense of security. and when they push trying to kill you, that is when you rip off your noob disguise and turn to them like
Posted Image

then you murder their entire team.

#12 HiProfile

    Member

  • Pip
  • Survivor
  • 18 posts

Posted 30 January 2015 - 08:18 PM

View PostAgent 0 Fortune, on 30 January 2015 - 06:14 PM, said:

if you were the "ringer" on your team your Elo doesn't move at all.

Furthermore, the matchmaker fills up 1 team at a time

Changing your personal Elo is never going to work long term, because when there are too few players it isn't used, and where there are lots of players it will quickly return to equilibrium.

The only real solution is population increase.


First it's asinine if ELO can go up so easily but takes a ripe scenario to go down. If I was advanced in a bowling league each time I got a strike but held at that level a few games after becoming a parapalegic, someone may find that fishy. I wish school had been like that - have an F on a test not count because the class average was under my current GPA. I could have busted my ass the first week then slacked the rest and kept a 4.0 GPA.

The second part of your reply is the reason I think the current system is completely worthless. Think about when you were in grade school (kids of all ages, growth stages, etc) and you stood in line getting picked for dodgeball. It usually came out relatively even. Think about what the teams would be like if you picked 2 random kids from different schools, matched them up with their doppelgangers, then faced those teams against each other. That's what MWO is like - people of ALL sorts of abilities, not kids who grew up playing with each other. Until they build a set of 24 and then split it down the middle (random or via ELO), this matchmaker is worthless.

BTW I find it interesting that matchmaker is supposed to work better with a larger sample size, yet all sorts of bats*** crazy things happen when the server gets flooded during these events.

#13 CocoaJin

    Member

  • PipPipPipPipPipPipPipPipPip
  • 3,607 posts
  • LocationLos Angeles, CA

Posted 31 January 2015 - 12:01 AM

View PostAgent 0 Fortune, on 30 January 2015 - 06:14 PM, said:

It is more complicated that just "lowering your Elo".
First, your Elo only goes down when your team loses, and your personal Elo is lower than the enemy teams average. This means that if you were the "ringer" on your team your Elo doesn't move at all.

Furthermore, the matchmaker fills up 1 team at a time, so the first team gets all the Elo outliers (people at the top or bottom of the Elo ranking). This is because "wait in queue time" determines the first player picked in a match for which all other Elo is built around. Guess who waits in queue longest, it is the upper and lower Elo values (especially those in the most popular weight bracket). Now, that you have someone in the fringe as your "primary elo target". This means all high (or low) Elo players (at least those with long queue wait times) are placed on the same team, once that team reaches 12 players the next team forms, and guess what?

Those players are probably all drawn from the the center of the Elo curve because team1 took all the outliers. This means that team1 is going to have either much higher or much lower Elo than team 2. The only times this is not true is when there is enough population to support players in the extreme Elo brackets.

Changing your personal Elo is never going to work long term, because when there are too few players it isn't used, and where there are lots of players it will quickly return to equilibrium.

The only real solution is population increase.


Reference source plz

#14 Ghogiel

    Member

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

Posted 31 January 2015 - 01:52 AM

View PostAgent 0 Fortune, on 30 January 2015 - 06:14 PM, said:

and your personal Elo is lower than the enemy teams average

It never used to. Sauce?

#15 Joseph Mallan

    ForumWarrior

  • PipPipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • FP Veteran - Beta 1
  • FP Veteran - Beta 1
  • 35,216 posts
  • Google+: Link
  • Facebook: Link
  • LocationMallanhold, Furillo

Posted 31 January 2015 - 02:40 AM

View PostGhogiel, on 31 January 2015 - 01:52 AM, said:

It never used to. Sauce?
I agree. This is the first I read that Agent.

#16 Ghogiel

    Member

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

Posted 31 January 2015 - 02:50 AM

View PostJoseph Mallan, on 31 January 2015 - 02:40 AM, said:

I agree. This is the first I read that Agent.
Yeah it's been explained to work compeltely differently than that. As in it's afaik an averaged team Elo that goes into the equation that is used to find the number of points players are adjusted, not a single player against the avergaed Elo of the other team.

#17 F4T 4L

    Member

  • PipPipPipPipPipPipPip
  • 767 posts

Posted 31 January 2015 - 02:53 AM

View PostAgent 0 Fortune, on 30 January 2015 - 06:14 PM, said:

Furthermore, the matchmaker fills up 1 team at a time


Why do you believe this? I've seen others make the same claim, but AFAIK the devs have never said the mm works this way.

#18 That Dawg

    Member

  • PipPipPipPipPipPipPipPip
  • FP Veteran - Beta 1
  • 1,876 posts

Posted 31 January 2015 - 05:56 AM

so......all I'm getting out of this thread is.........I get to keep posting screens in the Carry Harder thread? ....offs...

View PostF4T 4L, on 31 January 2015 - 02:53 AM, said:

the devs have never said the mm works this way.



pro tip: I'd never use "the devs never said" in any argument...

Because. PGI

#19 Sug

    Member

  • PipPipPipPipPipPipPipPipPip
  • The People's Hero
  • The People
  • 4,627 posts
  • LocationChicago

Posted 31 January 2015 - 06:45 AM

Hey speaking of intentionally losing does any know what I can do to start intentionally winning?

Cause right now I only win half my matches no matter what i do. I can get 8 kills and lose 8 - 12 or tab out and forget I was playing and come back ten minutes later to a 12 - 0 win.

#20 CocoaJin

    Member

  • PipPipPipPipPipPipPipPipPip
  • 3,607 posts
  • LocationLos Angeles, CA

Posted 31 January 2015 - 07:56 AM

Devs said that ELO changes are based on the outcome of the match, win/lose. After the teams are made, the MM makes an assumption of the outcome based on the two team's ELOs. If the outcome is different from the presumed outcome, as in the lower ELO team wins, then the ELOs of players change, with the underdogs getting an increase in ELO and the previously assumed winning team players getting a decrease.

If the presumed winning team wins, there is no ELO change for anybody...or at best there is a very small change, but honestly, I'm 99% sure I read that there would be no change.

The MM does the best it can to always create a balanced matchup, but it's already accepted there will be some acceptable deviation in the two team's ELOs...that is, unless there is a short fall of acceptable players in the queue to match, in this case, the MM does the best it can with what it has.





1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users