Jump to content

Do A Complete Reset Of The Psr?


34 replies to this topic

Poll: Do you want a reset of the PSR in hopes of getting more equal matches? (22 member(s) have cast votes)

Do you want a reset of the PSR in hopes of getting more equal matches?

  1. Voted Yes (11 votes [50.00%] - View)

    Percentage of vote: 50.00%

  2. Yes, because... (text below) (3 votes [13.64%] - View)

    Percentage of vote: 13.64%

  3. No (6 votes [27.27%] - View)

    Percentage of vote: 27.27%

  4. No, because... (text below) (2 votes [9.09%] - View)

    Percentage of vote: 9.09%

Vote Guests cannot vote

#21 Cichol Balor

    Member

  • PipPipPipPipPipPip
  • The God
  • The God
  • 354 posts

Posted 05 January 2019 - 09:52 AM

sitting in the back line firing lrms getting loads of unfocused damage is not helping the team. charging in and leading the push leading to a much lower score is helping the team. because teams are random and you will always be mixed with different people if you are regularly doing bad you are not going to be winning much at best it will only be 50% of the time but since your team will always be at a disadvantage having you it will be a far lower win rate than that.

Edited by Cichol Balor, 05 January 2019 - 02:24 PM.


#22 General Solo

    Member

  • PipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • 3,625 posts

Posted 05 January 2019 - 04:53 PM

View PostCichol Balor, on 05 January 2019 - 09:52 AM, said:


sitting in the back line firing lrms getting loads of unfocused damage is not helping the team. charging in and leading the push leading to a much lower score is helping the team



Damage is damage it contributes to wins, besides if the back field lermer dudes team collapses they wont have a great match score compared to if their team wins. So IMO this is wrong.

Also charging in and leading a push isn't the only play style so why should it be favored over other play styles, such as trading, flanking, wolf packing, lerm/.narcing etc.

If your regularly doing bad your going to have a bad Match score regularly as well, win or lose.

Being on the winning team does not make you a winner in terms of personal performance.
PSR should reflect this.

Boosting a players PSR on a win when they perform badly just causes problems as this player now is fighting opponents more skilled than themselves.

Which leads to disappointment and eventually maybe even leaving the game.

The upward bias creates a larger and larger skill gap in a matches as time goes on.

Edited by OZHomerOZ, 05 January 2019 - 04:58 PM.


#23 Cichol Balor

    Member

  • PipPipPipPipPipPip
  • The God
  • The God
  • 354 posts

Posted 06 January 2019 - 10:15 AM

it doesn't just favor one play style it favors winning playstyles

damage is not damage. when you CT and kill a mech any damage dealt to the rest of it was simply waisted damage.

you can have a high match score with plenty of methods that don't generally lead to a team winning. taking your flea to the enemy spawn and farming DCs is another one. You can't however regularly win matches if you yourself are playing badly. sure will you sometimes be on teams that carry you? yes, however over hundreds of matches this effect becomes greatly lessoned. And I have said personnel score should be taken into account but to a much lower degree than win/loss ratio.

The current problem with the existing PSR is mainly how hard it is to lower your ranking because PGI was more interested in using it as a leveling system than a skill rating.

#24 General Solo

    Member

  • PipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • 3,625 posts

Posted 07 January 2019 - 05:15 AM

Teams win or Lose
Players, play good or bad
PSR is intended to weigh up the player not the team.

#25 Cichol Balor

    Member

  • PipPipPipPipPipPip
  • The God
  • The God
  • 354 posts

Posted 07 January 2019 - 07:47 PM

YOU ARE PART OF THE TEAM. how well you play directly contributes to how well your team does. a guy who literally never touches his keyboard is going to lose far more matches than someone actually trying. by factoring in wins/loss you do weigh up the player not the team. your point would only be true if teams stayed the same but they don't

#26 General Solo

    Member

  • PipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • 3,625 posts

Posted 08 January 2019 - 04:49 AM

So the guy who never touches the keyboard, what happens when they get carried.
Their PSR gets an artifical boost.

So even if they played bad on a win, their PSR goes up or remains unchanged. Logic?
Posted Image

I agree to disagree with you.

#27 General Solo

    Member

  • PipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • 3,625 posts

Posted 08 January 2019 - 05:03 AM

Lets have a look at the example I posted earlier

Posted Image

Have a look at the mech on the winning team who got 21 damage and 75 match score.

FACT: They have the worst performance metrics out of all 24 players, including both winning and losing teams.

In my mind they deserve a PSR drop

However under the current system their PSR remains unchanged just because they are on the winning team (got carried) and due to biased PSR change conditions for winning team.

And that player is PART OF THE WINNING TEAM.
Did they contribute to the win?

I Dunno you tell me.

I say no and PSR should reflect this.

Now Mr 25 damage will be allocated to the same skill pool instead of a skill pool of players more his own skill level.

Can't be fun for Mr 25 damage seal playing with sharks.

Edited by OZHomerOZ, 08 January 2019 - 05:07 AM.


#28 Cichol Balor

    Member

  • PipPipPipPipPipPip
  • The God
  • The God
  • 354 posts

Posted 08 January 2019 - 06:00 AM

but... you literally can't tell from the score if they contributed or not. that urbi could very well have squirreled off a portion of the enemy for a good deal and considering they managed to at least ping 6 mechs that's probably the case. getting a lower score being away from protected and information ticks probably didn't help.

or he could have just run through the enemy firing mg and died right away
the scoring system we have simply isn't an accurate representation of how much you did or did not contribute.


your adverage win rate however is a pretty good metric for how offten you were a contributing factor

Edited by Cichol Balor, 08 January 2019 - 06:02 AM.


#29 General Solo

    Member

  • PipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • 3,625 posts

Posted 09 January 2019 - 03:52 AM

What you say may or may not be the case, lets say it is true.

How do you measure that? Currently their is no way to measure that, currently their is no contribution metric we can measure except Match score and other stats.

So I'm just going by things we can measure.

And 25 damage 6 assists guys metric is rated last from all 24 players and is quite a bit below the 2nd last guy of 125 damage and 5 assists.

So in my mind and possibly others 25 damage guy contributed the least.

And even if he drew half the enemy team away he got the win for his troubles.

And if he did Jack Sheet, he got immunity from PSR drop coz he got carried.
And thats just wrong!

Edited by OZHomerOZ, 09 January 2019 - 03:52 AM.


#30 Cichol Balor

    Member

  • PipPipPipPipPipPip
  • The God
  • The God
  • 354 posts

Posted 09 January 2019 - 07:20 AM

I know a metric that is easy to measure and isn't arbitrary like the current scoring system.

your win/loss rate

mine promotes doing what you can to win in order to increase your rank and will ultimetly wind up placing players against other players of similar skill. its going to be near impossible to maintain the same win/loss rate of players much better than you are.


your promots gaming a scoring system that to your own admition cant actually tell how much you contributed to your team winning. and because this is can be achieved either by gaming or playing well you wind up with players being matched with a wider variety in skill levels. I mean FFS one of the most important metrics to measuring how well you play isn't even tracked by the scoring system that being how well you spread damage and use armor

Edited by Cichol Balor, 09 January 2019 - 07:25 AM.


#31 General Solo

    Member

  • PipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • 3,625 posts

Posted 09 January 2019 - 12:49 PM

Yes if you measure win rate alone and nothing else.

Agreed.

Edit: IMO win rate still is tainted by team performance especially if queues are not separated.

Also my way is minimal effort.
Simply change a if statment like iF Between X and Y statment or Vlook up table and thus easy to do.

Edited by OZHomerOZ, 11 January 2019 - 03:17 AM.


#32 Liveish

    Member

  • PipPipPipPipPipPipPip
  • CS 2022 Referee
  • CS 2022 Referee
  • 843 posts
  • LocationDarwin

Posted 09 January 2019 - 02:59 PM

A Reset with no changes is pointless and we just end up back to where we started

#33 Mr Andersson

    Member

  • PipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 217 posts

Posted 09 January 2019 - 04:54 PM

Yes, I would like to see a reset.

But more importantly, the metrics behind the ranking need to be changed into something that, more accurately, measures actual skill.

#34 Warning incoming Humble Dexterer

    Member

  • PipPipPipPipPipPipPipPip
  • Bridesmaid
  • 1,077 posts

Posted 12 January 2019 - 08:50 AM

I'd like a PSR reset ~once per year.

But also a change to the PSR mechanic, where a balance of ~20% active players per PSR level is maintained.

Like once per year everyone is reset to level 3, but after a minimum amount of ~10 matches you are teleported straight to your current PSR level, with only one PSR level update per day required (or even less).

Then add cockpit items that display your current PSR level, now that it means something.

Example of a ranking formula : KILLS / (1+(DEATHS * WINRATE))

#35 JRcam4643

    Member

  • PipPipPipPipPipPip
  • 216 posts
  • Twitter: Link
  • LocationArizona, USA

Posted 26 January 2019 - 11:36 PM

If you're not going to change how PSR is calculated then I don't see what you're gaining. Getting rid of PSR all together would be best I think but I don't think that will happen.





3 user(s) are reading this topic

0 members, 3 guests, 0 anonymous users