Jump to content

Tier Distribution


44 replies to this topic

#41 Mawai

    Member

  • PipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 3,495 posts

Posted 22 January 2016 - 12:20 PM

View PostKHETTI, on 22 January 2016 - 05:22 AM, said:

Tier system is nothing more than a distraction from the fact that MWO has absolutely no match making in place at all, its currently a f*cking joke.
There is no skill separation at all, low skill players are mixed with high skill players dis-proportionally, evident from the fact that i regularly have nights where getting a win is not even possible.
Having consistent teams made up of players who game after game can't scrape more than 50-60 damage, i'm outscoring them easily in deliberate crap mechs with crap loadouts.

I'll be bluntly honest here, noobs ruin this game for me, i don't want them in my games.


The problem is having a reliable metric for deciding "good" or "skilled" players.

The previous system used was Elo which some folks hated. It was based on your team winning matches. You are the only common element in every game you play (unless in group play where there should have been a separate Elo rating) so over enough matches, your contribution to winning can be extracted and used for matchmaking. There was a separate Elo for each weight class.

The current system is called PSR and is related to your match score though it has a different formula and may have different weighting. On a win your PSR goes up or stays equal if your match score is low enough (something like less than 100 or 80). On a loss your PSR goes down unless your match score is over 250 in which case it stays even and if it is really high your PSR could go up. However, averaged over time for an average player with more or less even numbers of wins and losses ... PSR is more of an XP bar than anything else. The developer has said that everyone should end up in tier1 or 2 if they play long enough.

However, the biggest issue with any of these systems is not the ranking but the players ..
- players do not play the same in every match ... sometimes they are playing well and other times not. A matchmaker can never account for this.
- players play different mechs, different weight classes, different loadouts and with different playstyles. The PSR matchmaker does not take any of this into account ... and it can't really unless you want a perfect match about once a century. The more constrained the matchmaking the longer it takes to make a match.

Anyway, one of the better ways to advance in PSR is actually using LRM boats to inflict decent damage on lots of targets and collect assists and other things to boost the match score ... however, many "good" players complain about LRMs and claim that they are nub and unskilled weapons. Thus, a lot of the LRM "nubs" out there are probably at tier 1 just due to the mechanics of PSR and grinding out sufficient games.

The bottom line is that matchmaking can't fix any of these problems ... it doesn't know the players mood and it has a hard time judging if the player is using a decent build, joke build or meta build ... all of which will affect their performance in the match but are more or less un-quantifiable by the matchmaker without some significant work on rating systems like battle value for mechs.

#42 1Grimbane

    Member

  • PipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 1,123 posts
  • Locationsafe. . . . . you'll never get me in my hidey hole.

Posted 22 January 2016 - 12:23 PM

for a while i really put in effort to move down to tier 4... waht ev's i like wiping noobs but nyway.. that was a trial in patience that never payed out, still tier three after 2 solid months of the worst non TK play i could muster. i mean real asinine gameplay to try to lower my tiers....... i have moved farther into tier 3 than before lol



if you ever saw me trying to take out dire's head on in my commandos .... ^

Edited by 1Grimbane, 22 January 2016 - 12:25 PM.


#43 SHD2

    Member

  • Pip
  • CS 2024 Top 12
  • CS 2024 Top 12
  • 19 posts
  • LocationFrance, Toulouse

Posted 03 March 2018 - 03:22 PM

plot twist i know some peoples who play long time and dont end up T1 :
"A" player is T2 with [color=#000000]

5542 matches during the past 20 seasons (0.92 W/L 170 avg match score)

[/color]

[color=#000000]

"B" player stay T5 : 395 matches (0.77 W/L 128 avg match score)

[/color]


#44 Exilyth

    Member

  • PipPipPipPipPipPipPipPipPip
  • Bridesmaid
  • 2,100 posts
  • LocationTerra

Posted 03 March 2018 - 04:21 PM

I'm playing since open beta and just reached tier 2 and I'm pretty sure it wasn't my skill getting me there.

*goes and reads more about tactics and the so called 'meta' (whatever that is) on the forums*

#45 PocketYoda

    Member

  • PipPipPipPipPipPipPipPipPip
  • Shredder
  • Shredder
  • 4,147 posts
  • LocationAustralia

Posted 03 March 2018 - 05:06 PM

View PostXX Sulla XX, on 22 January 2016 - 04:37 AM, said:

Except players that can not keep a positive winning percentage when they move up in tier.

Not true i lose far more than i win (some on purpose) and the bar still rises, you cannot go backwards unless you lose every match permanently and i'm not sure even that would make it go backwards.

I hate PSR to the point i'd rather just die in matches to make it go backwards.

Edited by Samial, 03 March 2018 - 05:07 PM.






1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users