Jump to content

My Thoughts On Elo (Yep, Another One)


156 replies to this topic

#81 Aethon

    Member

  • PipPipPipPipPipPipPipPipPip
  • Mercenary
  • 2,037 posts
  • LocationSt. Louis, Niles, Kerensky Cluster

Posted 26 February 2013 - 06:27 AM

View PostSlXSlXSlX, on 26 February 2013 - 06:18 AM, said:

If i had a match like critical fumbles, id want to make a thread about it.

http://mwomercs.com/...78#entry1953278

I might even want to sing and dance about it.




I just had an 8-6 battle...good nail-biter. Felt good...hope I see more. :lol:

Also, lol @ video

Still, is solo-dropping the key to getting good matches? Or is it common with groups somehow?

Edited by Aethon, 26 February 2013 - 06:28 AM.


#82 Critical Fumble

    Member

  • PipPipPipPipPipPipPip
  • Knight Errant
  • 810 posts

Posted 26 February 2013 - 06:33 AM

View PostAethon, on 26 February 2013 - 06:14 AM, said:

How often are you getting matches that end like that? Are you playing in a group, or solo?

THAT close? Pretty rare. I think I've had one, maybe two others like that from when Elo went live. Which is to say, noticeably more often than before. Or it could be a fluke or simply what I'm expecting to see; the devs would know, though, hopefully we'll be getting some more concrete data in the dev update soon. I play solo, and the game has improved for me.

Friendly advice, if you're having trouble with uncoordinated PUGs, offer some coordination. The one constant in my recent good matches - someone steps up. Have a plan, offer the plan, and if they refuse to work with you - that's their loss.


View PostSlXSlXSlX, on 26 February 2013 - 06:18 AM, said:

If i had a match like critical fumbles, id want to make a thread about it.

http://mwomercs.com/...78#entry1953278

I might even want to sing and dance about it.

Spoiler


You really like that vid, don't you xD

That's actually why I saved it. I've been meaning to start a thread about good matches as a soft counter to the unchecked derision on the forums.

What do you think - "The Thread of Valhalla"? Or will the FRRs beat me up over that one?


View PostAethon, on 26 February 2013 - 06:27 AM, said:


I just had an 8-6 battle...good nail-biter. Felt good...hope I see more. :lol:

Also, lol @ video

Still, is solo-dropping the key to getting good matches? Or is it common with groups somehow?

Elo is possibly still a n00b, PGI may need to teach it how to make it work as well for groups as it does solos.

#83 SlXSlXSlX

    Member

  • PipPipPipPipPipPipPip
  • 666 posts

Posted 26 February 2013 - 06:33 AM

View PostAethon, on 26 February 2013 - 06:27 AM, said:


I just had an 8-6 battle...good nail-biter. Felt good...hope I see more. :lol:

Also, lol @ video

Still, is solo-dropping the key to getting good matches? Or is it common with groups somehow?

The key is, to not team stack in teh first place. If you want ot run a 4 man, take half silly huilds, and accept taht your gonna lsoe some.

The other key is to look at losing half your matches as a good thin as balance, and to not come on the forums with pitch forks chasing down Elaine.

Shes a good girl, and wants us all to have "max fun".

Also not teaming with ppl well beyond your skill level helps, they end up carrying you, and inflating your stats. Then when they sign off and you go solo sam, your going to ahve a bad time.

#84 Aethon

    Member

  • PipPipPipPipPipPipPipPipPip
  • Mercenary
  • 2,037 posts
  • LocationSt. Louis, Niles, Kerensky Cluster

Posted 26 February 2013 - 06:39 AM

View PostSlXSlXSlX, on 26 February 2013 - 06:33 AM, said:

The key is, to not team stack in teh first place. If you want ot run a 4 man, take half silly huilds, and accept taht your gonna lsoe some.

The other key is to look at losing half your matches as a good thin as balance, and to not come on the forums with pitch forks chasing down Elaine.

Shes a good girl, and wants us all to have "max fun".

Also not teaming with ppl well beyond your skill level helps, they end up carrying you, and inflating your stats. Then when they sign off and you go solo sam, your going to ahve a bad time.


That is why I have endeavoured to master the Jenner first. :lol: It really is fun, even though I have a really hard time dueling with Ravens. I really do not want to be 'one of those 3L guys', lol.

Had a really, really good duel with someone in another Jenner a bit earlier...best fun I have had in MWO in a while...until a friendly Centurion came over and nailed him. :)

Edited by Aethon, 26 February 2013 - 06:40 AM.


#85 Havok1978

    Member

  • PipPipPipPipPipPip
  • Fury
  • Fury
  • 371 posts
  • LocationTexaz!!

Posted 26 February 2013 - 06:41 AM

View PostThirdstar, on 26 February 2013 - 04:40 AM, said:


The only person I see whining in the whole thread (and that other one) is you.

Waaaaaaaah I want my win rate back! Those damn PUGs are actually beating me! NO FAIR!


You really dont get it do you? its not about pug stomping its about FAIL pugs on our team. Its about complete lack of even an idea of how to function as a team.. You Sir are obviously the type of pug that benefits from riding skilled players coat tails.. Now get off my lawn and away from my bridge,,,,,

#86 PhoenixFire55

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 5,725 posts
  • LocationSt.Petersburg / Outreach

Posted 26 February 2013 - 06:43 AM

View PostSlXSlXSlX, on 26 February 2013 - 05:14 AM, said:

Thirdstar is right, and im glad hes sharing the burden of arguing against the bads, not bad as in bad @ mechs, but bad as in, wants to farm pubs and enjoy 8/10 matches ending in their favor kinda bad.


Let us imagine for a second that the only way we can play this game is solo.

Then let us take a skilled player, not among the top 100 players in game, but near that in top 10% players. With ELO he will only be matched against players of near same skill level. This will mean that he will win about 50% of his games and lose about 50% of his games.

Now let us take a non-skilled player, one from the bottom 10% of players in game. With ELO he will also be matched against players of near same skill level. This will mean that he will also win about 50% of his games.

In matches that have people of same skill, all memebers of the team on average do about same amount of damage, kills, assists, spotting regardless if players are from top ELO zone, middle ELO zone or botton ELO zone. In turn it means that player that is better then 90% of players in game and a player that is worse then 90% of players in game will be getting same amounts of c-bills and exp, as well as haveing same stats.

Best players in game have all rights to expect that 8 out of 10 games will end up in their favor, worst players in game have no rights to expect that that less then 8 out of 10 games will end up not in their favor. With ELO both types will have 5 out of 10 games ending up in their favor. I fail to see how ELO is fair. Just as I fail to see how people who bring up arguments to show that ELO isn't fair are bad.

What ELO does is encouraging noobiness, no matter how bad you play you will still be winning 50% of your games and you will still be getting same c-bills and exp as everybody else. What people will do is set their mech into afk mode for a day to drop ELO down into worst possible zone, then play for a day, stomping enemy teams practically on their own and getting way more c-bills and exp by doing that. Now tell me, am I bad for not wanting THIS to happen?

#87 PhoenixFire55

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 5,725 posts
  • LocationSt.Petersburg / Outreach

Posted 26 February 2013 - 06:50 AM

View PostSlXSlXSlX, on 26 February 2013 - 05:21 AM, said:


You can play very selfishly, completely boning your team, in fact hamstringing them, and end up with top score. That does not mean your good. That is not how good is measured. But by all means, continue.


You can not do 100 smth damage, steal all 8 kills and still be at the top of your team. Whoever done the most damage will be on the top of the team. Even if you try to stay behind, hide and come out in the end to steal all kills you will not be able to do it on regular basis. Most likely you'll end up always losing because you aren't helping your team.

View PostSlXSlXSlX, on 26 February 2013 - 05:44 AM, said:


OMG, we have a random queue, and an 8 man queue. What you say would help, is currently available!



Not 'random' queue and '8-man' queue. 'Random' queue and 'groups-of-all-sizes' queue. There is a big difference.

View PostSlXSlXSlX, on 26 February 2013 - 05:48 AM, said:

You arguments, have been countered, but you jsut keep going on. I cant blame his frustration with you.


Show me where they were countered.

#88 SlXSlXSlX

    Member

  • PipPipPipPipPipPipPip
  • 666 posts

Posted 26 February 2013 - 06:51 AM

View PostPhoenixFire55, on 26 February 2013 - 06:43 AM, said:

Best players in game have all rights to expect that 8 out of 10 games will end up in their favor, worst players in game have no rights to expect that that less then 8 out of 10 games will end up not in their favor.


No you dont. In wot the top 3% have a win rate of 57-58% or better. Thats the cream of the crop. In wot,a 57% win rate means a heck of alot.

You do not hvae that "right" in any way shape or form. That you would even make an assertion like this, tells us everything we need to know about you.

Id say you disgust me, but I'll jsut call you a bad sport.

Have fun im off to work, Sport.

#89 PhoenixFire55

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 5,725 posts
  • LocationSt.Petersburg / Outreach

Posted 26 February 2013 - 06:58 AM

View PostSlXSlXSlX, on 26 February 2013 - 06:51 AM, said:

No you dont.


What does it have to do with me? I said 'best players'.

Bringing WoT which is a different game does not make sense either. THE best player can take down 3-4 mechs in practically any given game. Chances that 7 other players in his team will be able to take down 4-5 remaining mechs are about 80%. 8 out of 10.

#90 FerretGR

    Member

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

Posted 26 February 2013 - 07:33 AM

View PostAethon, on 26 February 2013 - 06:06 AM, said:

If you ride the bus with some homeless people, does that make you a bum?


Difference being, of course, bus riders aren't sorted by a piece of software that attempts to place folks of similar rank together. Whereas Elo places you with players at or around your skill level as well as it can. I'll accept that there are occasional fails, but if game after game, you're paired with nothing but scrubs, I'd start to look elsewhere for the fail, if I were in your shoes.

#91 Thirdstar

    Member

  • PipPipPipPipPipPipPipPipPip
  • 2,728 posts
  • LocationIndia

Posted 26 February 2013 - 07:40 AM

View PostPhoenixFire55, on 26 February 2013 - 06:43 AM, said:

Best players in game have all rights to expect that 8 out of 10 games will end up in their favor, worst players in game have no rights to expect that that less then 8 out of 10 games will end up not in their favor. With ELO both types will have 5 out of 10 games ending up in their favor. I fail to see how ELO is fair. Just as I fail to see how people who bring up arguments to show that ELO isn't fair are bad.


It boggles the mind that you think this is a valid argument.

#92 Yokaiko

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Bad Company
  • Bad Company
  • 6,775 posts

Posted 26 February 2013 - 07:46 AM

View PostSlXSlXSlX, on 26 February 2013 - 06:18 AM, said:

If i had a match like critical fumbles, id want to make a thread about it.

http://mwomercs.com/...78#entry1953278

I might even want to sing and dance about it.





You have never had a 5 kill +900 point match? I'd call it a good game, but from unique, or dance worthy.

#93 Vasces Diablo

    Member

  • PipPipPipPipPipPipPip
  • 875 posts
  • LocationOmaha,NE

Posted 26 February 2013 - 07:51 AM

2 points about ELO:

1. It will take time. It compiles data on players over the long run. We'll eventually end up with a bell curve of sorts that everyone is on.

2. Yes, it will currently put some low ELO players in matches with high ones to achieve an equal average. However, it will always try and match equal players first. If you keep ending up in games with terrible players, I have some bad news for you....

#94 Yokaiko

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Bad Company
  • Bad Company
  • 6,775 posts

Posted 26 February 2013 - 07:56 AM

View PostVasces Diablo, on 26 February 2013 - 07:51 AM, said:


2. Yes, it will currently put some low ELO players in matches with high ones to achieve an equal average. However, it will always try and match equal players first. If you keep ending up in games with terrible players, I have some bad news for you....


it also depends on how many people are in the queue. Figure that for the 2 minute window allowed for matching, there is 6-9 minutes of in game time per person. That means the majority (like 65%) of the player online at a given time are in a match.

That means out of perhaps 50 players queuing, they have to find 16 in your ELO bracket....assuming that the server populating is between 2000-3000.

The odds there start getting a little crazy, that is why the weight matching went bye bye as well I'd wager.


(no calculators where harmed making this post, its rounding and general figuring I didn't crunch it out)

#95 Darzok

    Member

  • PipPipPipPipPipPip
  • The Scythe
  • The Scythe
  • 255 posts

Posted 26 February 2013 - 07:57 AM

I think its been a great change most battles are now much more close to the point you would say its 100% balanced yes now and then the odd match is a massive stomp for one side but its quite rare.

#96 Yokaiko

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Bad Company
  • Bad Company
  • 6,775 posts

Posted 26 February 2013 - 07:59 AM

View PostDarzok, on 26 February 2013 - 07:57 AM, said:

I think its been a great change most battles are now much more close to the point you would say its 100% balanced yes now and then the odd match is a massive stomp for one side but its quite rare.



Depends on when you play. I play late EU usually, pretty sure that is a low population during the week. The games get more challenging if I stay awake into US time, which is rare I usually go to bed around 2-3 EST.

In the morning all bets are off I see some faces out of the 8 man queue, but for every one of those there is a guy in a trial mech.

Still landslides are as common as not.

A lot of the time the first team to 5 kills wins, that tends to be critical mass.....three kills if your entire team is relatively undamaged.

Edited by Yokaiko, 26 February 2013 - 08:00 AM.


#97 Vasces Diablo

    Member

  • PipPipPipPipPipPipPip
  • 875 posts
  • LocationOmaha,NE

Posted 26 February 2013 - 07:59 AM

View PostYokaiko, on 26 February 2013 - 07:56 AM, said:


it also depends on how many people are in the queue. Figure that for the 2 minute window allowed for matching, there is 6-9 minutes of in game time per person. That means the majority (like 65%) of the player online at a given time are in a match.

That means out of perhaps 50 players queuing, they have to find 16 in your ELO bracket....assuming that the server populating is between 2000-3000.

The odds there start getting a little crazy, that is why the weight matching went bye bye as well I'd wager.


(no calculators where harmed making this post, its rounding and general figuring I didn't crunch it out)


With out a doubt, lots of factor at play here. Playing at peak times vs not for example. All this means people need to chill out on the ELO critique for now. Though I'm guessing last weeks tourney was a huge data grab.

Edit: also, loved the disclaimer at the end

Edited by Vasces Diablo, 26 February 2013 - 08:00 AM.


#98 Yokaiko

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Bad Company
  • Bad Company
  • 6,775 posts

Posted 26 February 2013 - 08:02 AM

View PostVasces Diablo, on 26 February 2013 - 07:59 AM, said:

With out a doubt, lots of factor at play here. Playing at peak times vs not for example. All this means people need to chill out on the ELO critique for now. Though I'm guessing last weeks tourney was a huge data grab.

Edit: also, loved the disclaimer at the end



Yeah had to have the disclaimer or someone with a couple math classes WOULD grind it out and tell me to the nearest 0.0001% where I screwed it up :)


My assumption was that the tournament was a data grab as well, as soon as I saw "solo" and the math they were using.


My W/L was 2:1 across all four weight classes and my KDR actually went up 0.03 (no laughing matter ~2700 kills). So I figure that I landed a little south of where I should be and am working up to it. I solo about 1/2 of the time anyway (Yank on not Yank time schedule). That wasn't terribly surprising.

Edited by Yokaiko, 26 February 2013 - 08:04 AM.


#99 Joseph Mallan

    ForumWarrior

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

Posted 26 February 2013 - 08:09 AM

View PostPhoenixFire55, on 26 February 2013 - 06:43 AM, said:

Best players in game have all rights to expect that 8 out of 10 games will end up in their favor, worst players in game have no rights to expect that that less then 8 out of 10 games will end up not in their favor. With ELO both types will have 5 out of 10 games ending up in their favor. I fail to see how ELO is fair. Just as I fail to see how people who bring up arguments to show that ELO isn't fair are bad.
No the best players don't have that right. The best players in the game should expect to be in the top 3 places at the end of every match. Win or lose. Individual performance does not insure team victory.

#100 The Amazing Atomic Spaniel

    Member

  • PipPipPipPipPipPipPip
  • 932 posts
  • LocationBath, UK

Posted 26 February 2013 - 08:13 AM

View PostPhoenixFire55, on 26 February 2013 - 06:43 AM, said:

Best players in game have all rights to expect that 8 out of 10 games will end up in their favor, worst players in game have no rights to expect that that less then 8 out of 10 games will end up not in their favor.


Whoa there! That's a big assumption.

A far better way to ensure the longevity of the game would be to make sure that even new and/or not-very-good players (I was the former, I'm now the latter) can still have fun with the game and don't get turned away by crushing defeat after crushing defeat.





4 user(s) are reading this topic

0 members, 4 guests, 0 anonymous users