Jump to content

Elo Is For Chess, Not Mwo


198 replies to this topic

#81 Fire and Salt

    Member

  • PipPipPipPipPipPipPip
  • 526 posts
  • LocationFlorida

Posted 09 October 2014 - 07:09 AM

View PostVictor Morson, on 09 October 2014 - 02:11 AM, said:

Been saying this since the start. There are too many random variables in a team game like this to do ELO for anything other than absolutely competitive matches, to be honest.


Random, you say? Ever heard of the law of large numbers?



View PostVictor Morson, on 09 October 2014 - 02:11 AM, said:

Tracking it for individual pilots is absolutely worthless.

An experiment: Go grab a Locust. Go play 20 games. Charge in and die. You have a 50% chance of ending up with a positive ELO from that anyway.


Wrong. It will probably cause your team to lose a little more often than had you participated. The more trials you conduct, the more likely this will show. Again, law of large numbers.

But thanks for ignoring my detailed experiment posted above done using a new account and a trial nova. I won my first 11 games, about 20 out if the first 25, and this was in the solo queue where I was only 1/12th of the team.

Your fictional experiment truly adds value to this thread.





View PostVictor Morson, on 09 October 2014 - 02:11 AM, said:

But it is an absolute joke with 11 other players factoring in - any individual pilot makes up less than 10% of the team.


Nope. There are only 2 reasons why you would feel that you have a small impact as one of the twelve players. Again, see my nova example.

1.) You are average
2.) ELO is working, so you are being placed with similarly skilled players, thus making you seem more average than you actually are.

#82 IceSerpent

    Member

  • PipPipPipPipPipPipPipPipPip
  • 3,044 posts

Posted 09 October 2014 - 07:25 AM

View PostDiablobo, on 08 October 2014 - 08:40 PM, said:

Elo assumes both players have the same basic starting point. Apart from white getting the first move, chess is an extremely balanced game. Both players have the same pieces, and both players' pieces can do the exact same things.


Yes, that's correct.

Quote

Until PGI ditches Elo and uses some sort of Battle Value system, there can be no matchmaker that is anything other than random, luck of the draw crap. The sooner they acknowledge this fact, the sooner we can all get on to enjoying our epic well matched battles we were promised. Instead, we have to be saddled by noobs on our team who think it is fun to pilot an Atlas with a couple of med lasers and some LRMs with single heat sinks and no unlocks, while the other team has fully unlocked and optimized meta builds and pilots who know how to use them.


Your conclusion is wrong though - BV is needed, but it has to work in conjunction with Elo, not replace it. Elo determines player skill, but can't determine quality of equipment being used. BV determines quality of equipment, but can't determine player skill. Those two systems can't replace each other, but both are necessary in order to for MM to work properly.

#83 Jon Phoenix

    Member

  • PipPipPip
  • Mercenary
  • Mercenary
  • 68 posts
  • LocationDraconis Combine

Posted 09 October 2014 - 07:32 AM

View PostKarl Streiger, on 09 October 2014 - 06:44 AM, said:

Lobbby for casuals - that are new to the game - and they won't earn anything? Sounds really bad for a F2P.
The "competive" game on the other hand - with different ranks - is this in any kind different to the current matchmaker - with the difference that you see the other guys before you hit launch?


The new players can play in the low ranked lobbies in the competitive area. That's why there would really need to be a ranking system. At least new players would be able to practice and play with people in the casual lobbies at no risk to their rank, like we can do already with private matches. It makes sense to give no rewards when there's no risk.

Edited by Jon Phoenix, 09 October 2014 - 07:33 AM.


#84 Tastian

    Member

  • PipPipPipPipPipPipPip
  • Elite Founder
  • Elite Founder
  • 768 posts
  • LocationLayton, UT USA

Posted 09 October 2014 - 07:33 AM

I wish I personally had ELO per mech instead of per weight class. When I bought the Vindicator recently with 0 basic unlocks, I felt like I was thrown to the wolves again. Even mastered it was trash. And then subpar mechs like the Dragon I wouldn't be paired up against Timberwolves.

ELO is working somewhat however. I've seen my KDR go from about 7 down to about 1.75 since it was implemented.

A Battle Value system would be nice but you can still give two persons the exact same mech and one pilot be 100x better then the other pilot.

#85 Fire and Salt

    Member

  • PipPipPipPipPipPipPip
  • 526 posts
  • LocationFlorida

Posted 09 October 2014 - 07:34 AM

Battle value won't magically fix anything.


Smart players will just maximize their
Effectiveness / battle value
ratio.


Currently, they maximize their
Effectiveness / tonnage
ratio instead.




Side note: The T-Rex is a min-maxer.

#86 Jon Phoenix

    Member

  • PipPipPip
  • Mercenary
  • Mercenary
  • 68 posts
  • LocationDraconis Combine

Posted 09 October 2014 - 07:37 AM

View PostTastian, on 09 October 2014 - 07:33 AM, said:

A Battle Value system would be nice but you can still give two persons the exact same mech and one pilot be 100x better then the other pilot.


That's exactly why players should be ranked rather than mechs. BV can help balance teams so it might have some value, but it's not a fix for a match making system.

-Jon (wants to get rid of the MM)

#87 Glythe

    Member

  • PipPipPipPipPipPipPipPip
  • 1,566 posts

Posted 09 October 2014 - 07:38 AM

There's more than 1 element here when you get right down to the core of the issue.

1. Build value is completely ignored...... Stalker=Atlas=Dire wolf (lol... more like Stalker+Atlas).
2. Elo of the team is a quite stupid way to rank individual performance. You lost but do massive damage+assist every game. Your Elo goes down because you're a "bad" player. This makes no sense.


Build value would be a lot more viable as a method of balance than Elo. Why? Well if you're rocking double Gauss in a clan mech that's a lot of points. You could probably have just shy of 2 Atlases for that cost (give the DW a kitfox). And that fight would probably be pretty close. This would be a better way of balancing clan mechs in the first place. Oh you like clan mechs? well I hope you didn't want to bring more than about 1-2 heavies/assaults. As it stands look at 2 assaults or heavies clan vs IS and there is zero comparison because the TW and DW rule their brackets with an iron fist.

If you really want to start being fair then we should start having weight handicaps for build value inequalities. In the old days of 8 vs 8 sometimes you had 6 assaults and they still lost to good mediums and lights. In this format you really need to do something to make the IS assaults equal to the clan assaults. There's already a solution here with build points. I don't care who you are but if you are in a terrible mech you aren't going to contribute even if you are the world's best player.

Elo is handholding for bad players and bad matchmaking for good players.



I see a great number of games where one team has 4-5 players who score under 100 damage (most of them under 50). If the MM wasn't working this wouldn't be the case. I typically get about 100 match score with 400 damage (in crappy IS mechs).

I remember this game before Elo and I know it was better without it. The one concession I will make is that new players need protected match making.


Also.... just to be clear there is a really easy way to cheat in all the tournaments. Tank your score about every 6 weeks. Let your kids play, let your dog play... whatever. This assumes that you are good however.

Edited by Glythe, 09 October 2014 - 07:46 AM.


#88 Bront

    Member

  • PipPipPipPipPipPipPipPipPip
  • Urban Commando
  • Urban Commando
  • 4,212 posts
  • LocationInternet

Posted 09 October 2014 - 07:42 AM

The problems with Battle Value:

1) Battle Value is balanced for Table Top. Translation to MWO would be problematic, as weapons and gear are different in MWO (the AC5 and AC5 are much more useful here, for example, and most weapons stats are different from TT).

2) Battle Value does not account for the pilot like ELO does.

These 2 flaws are hard to overcome. Sure, you could rebalance BV for MWO, but there are so many non-TT factors (Ghost Heat, the effectiveness of weapon groups, how weapons complement a mech's role, chassis variations in hit boxes/size/speed, etc) to get it easily right, and that still doesn't include the pilot.

You can complain about ELO, but it's a system that many games use because it allows a general mathematical calculation to try to balance matches and has clear and impartial ways it can be influenced. And just like in Chess, ELO isn't everything. Plenty of players with lower ELO beat players with higher ELOs.

#89 RetroActive

    Member

  • PipPipPipPipPipPip
  • 405 posts
  • LocationFL, USA

Posted 09 October 2014 - 07:47 AM

View PostStormwolf, on 09 October 2014 - 04:10 AM, said:

BV is the way too go, players with min maxed builds will always have higher BV due to all the expensive components.


BV is an OK way to match drop decs, but it doesn't take into pilot skill into account AT ALL. A brand new player can throw down some MC for a min/maxed Mad Cat.

#90 Dracol

    Member

  • PipPipPipPipPipPipPipPipPip
  • The Steadfast
  • The Steadfast
  • 2,539 posts
  • LocationSW Florida

Posted 09 October 2014 - 07:47 AM

View Postlartfor, on 09 October 2014 - 05:20 AM, said:

ELO and excessive match making are designed for the hand hold generation of gamers and "participation" trophy kiddies. Sadly egomania has become the norm with today's younger gamers and apparently must be catered to at all costs.

Would you approve of white belts being forced to take on black belts?

Without ELO/Matchmaking, why would a new player stick around if they dropped against veterans a couple of times?

Although I agree the hand hold generation is a little ridiculous, but in regards to MM, it is required to make a game appealing to new players. If people are not matched up by skill, then only veterans will be left and the game would slowly die away.

A biz requires new players. Oh, and although the tutorials need a reworking, no matter how good they got, a new player would still never stand a chance against a veteran.

Edited by Dracol, 09 October 2014 - 07:49 AM.


#91 Jon Phoenix

    Member

  • PipPipPip
  • Mercenary
  • Mercenary
  • 68 posts
  • LocationDraconis Combine

Posted 09 October 2014 - 07:48 AM

Is this still how Elo is calculated?

#92 Tolkien

    Member

  • PipPipPipPipPipPipPipPip
  • Giant Helper
  • Giant Helper
  • 1,118 posts

Posted 09 October 2014 - 07:49 AM

Even putting Elo aside, here are problems with the current system:

Is a locust equal to a firestarter? not even close. But the matchmaker calls them both 'lights' and will give the same pilot in each the same value. Regardless of who the pilot is, they will do better in the latter than the former.

Is a hunchback equal to a shadowhawk? not even close. But the matchmaker calls them both 'mediums' and will give the same pilot in each the same value. Regardless of who the pilot is, they will likely do better in the latter than the former.

Is a Quickdraw equal to a madcat? not even close. But the matchmaker calls them both 'heavies' and will give the same pilot in each the same value. Regardless of who the pilot is, they will likely do better in the latter than the former.

Is an awesome equal to an atlas? not even close. But the matchmaker calls them both 'assaults' and will give the same pilot in each the same value. Regardless of who the pilot is, they will very likely do better in the latter than the former.

Is an atlas bristling with machine guns and flamers equivalent to an atlas with real weapons? No, but the matchmaker calls them both 'assaults' and will give the same pilot in each the same value. Regardless of who the pilot is, they will do better in the latter than the former.

Is an atlas with real weapons but no modules equivalent to an atlas with real weapons, artillery, cool shot, two good mech modules, and two good level 5 weapon modules? No, but the matchmaker calls them both 'assaults' and will give the same pilot in each the same value. Regardless of who the pilot is, they will do better in the latter than the former.

Is any mech with fully unlocked skills equivalent to the identical mech with nothing unlocked? No, but the matchmaker calls them identical and will give the same pilot in each the same value. Regardless of who the pilot is, they will do better in the latter than the former.

There's a very strong argument to convert the matchmaker over to use Battlevalue and treat our individual skill level as just another BV item like it is treated in tabletop - hell keep the Elo scoring for the pilot just scale it to BV for the purposes of balancing matches.

Edited by Tolkien, 09 October 2014 - 07:52 AM.


#93 IceSerpent

    Member

  • PipPipPipPipPipPipPipPipPip
  • 3,044 posts

Posted 09 October 2014 - 07:54 AM

View PostFire and Salt, on 09 October 2014 - 07:34 AM, said:

Battle value won't magically fix anything.

Smart players will just maximize their
Effectiveness / battle value
ratio.


BV is a measure of build effectiveness, "maximizing" effectiveness/BV ratio is only possible when BV values are not quite right, and this can be fixed via correcting those values.

View PostBront, on 09 October 2014 - 07:42 AM, said:

1) Battle Value is balanced for Table Top. Translation to MWO would be problematic, as weapons and gear are different in MWO (the AC5 and AC5 are much more useful here, for example, and most weapons stats are different from TT).


We're not talking about using TT values, but about creating new schema.

Quote

2) Battle Value does not account for the pilot like ELO does.


It's not supposed to - it's a metric of mech "quality", not player skill. As such, it has to be used in conjunction with Elo (essentially it would replace 3/3/3/3 rule).

#94 Haipyng

    Member

  • PipPipPipPipPipPipPip
  • The Grizzly
  • The Grizzly
  • 593 posts
  • LocationIn Transit

Posted 09 October 2014 - 07:55 AM

It's a valid point, ELO does not work well with MWO because not everything else is equal. There are a huge number of variables and builds to consider.

For the large group size it does not consider multipliers such as cooperative teams or comm usage simply because there is no way a computer can know what that looks like.

Some mechs work better on some maps...whether that is because it is a hot/cold, long/short range, tall/short obstruction, objective/deathmatch, or flat/hilly map. The mech you drop will do better or worse in those conditions depending on how you have it built.

#95 Killstorm999999

    Member

  • PipPipPipPipPip
  • 196 posts

Posted 09 October 2014 - 07:59 AM

What is the purpose of perfectly matching player skill, so all matches end up being an even affair? Of course we should avoid stomps, but there should be some asymmetry for the purpose of making things interesting.

Edited by Deltron Zero, 09 October 2014 - 08:05 AM.


#96 IceSerpent

    Member

  • PipPipPipPipPipPipPipPipPip
  • 3,044 posts

Posted 09 October 2014 - 08:09 AM

View PostFire and Salt, on 09 October 2014 - 07:09 AM, said:

Random, you say? Ever heard of the law of large numbers?


It doesn't work in this case, as match outcome doesn't always adjust one's rating. You get into a match against an equal Elo team, your team fights at a disadvantage due to having a mix of low Elo + high Elo players, you lose, your Elo drops. Next match you go against a team with slightly lower Elo, other team fights at a disadvantage due to having a mix of low Elo + high Elo players, you win, but your Elo remains unchanged. Lawof large numbers kicked in, but your rating is still off.

#97 Tastian

    Member

  • PipPipPipPipPipPipPip
  • Elite Founder
  • Elite Founder
  • 768 posts
  • LocationLayton, UT USA

Posted 09 October 2014 - 08:12 AM

Another problem with Battle Value is that it would have to be updated every time a weapon or system is buffed/nerfed.

#98 Killstorm999999

    Member

  • PipPipPipPipPip
  • 196 posts

Posted 09 October 2014 - 08:15 AM

I am for a lobby system. Let the players organize them selves, and also let them choose things like map and game mode, and number of players.

#99 Ghogiel

    Member

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

Posted 09 October 2014 - 08:15 AM

View PostDeltron Zero, on 09 October 2014 - 08:15 AM, said:

I am for a lobby system. Let the players organize them selves, and also let them choose things like map and game mode, and number of players.

So is PGI. That's why they put exactly that in the game.

Edited by Ghogiel, 09 October 2014 - 08:16 AM.


#100 Tolkien

    Member

  • PipPipPipPipPipPipPipPip
  • Giant Helper
  • Giant Helper
  • 1,118 posts

Posted 09 October 2014 - 08:18 AM

View PostDeltron Zero, on 09 October 2014 - 08:15 AM, said:

I am for a lobby system. Let the players organize them selves, and also let them choose things like map and game mode, and number of players.


When people say this, I remember how great multiplayer battletech 3025 was...
Posted Image

-Lobbies where you saw the other team, chose your mechs and only dropped when both sides agreed it was a good match. If a side was winning every match, they would reduce their tonnage to stay sporting and taunt the other team. e.g. people would ton down from jenners into urbanmechs.
-Every match had a meaning - if you won, you took 1% of the planet. If the other team declined to drop against you, you took 0.1% after 60 seconds.
-The more territory your house controlled, the higher your salary was.
-This was all in in beta.

Edited by Tolkien, 09 October 2014 - 08:18 AM.






55 user(s) are reading this topic

0 members, 55 guests, 0 anonymous users