Jump to content

The Way To Fix Psr


  • You cannot reply to this topic
24 replies to this topic

#21 Sandpit

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • Veteran Founder
  • Veteran Founder
  • 17,419 posts
  • Facebook: Link
  • Twitter: Link
  • LocationArkansas

Posted 25 December 2015 - 06:17 PM

One thing I've noticed regarding PSR and advancing up the ladder.

It's not so much slanted on wins, losses, kills, and such as it is match score and your bonuses for filling roles. I started noticing this the other night when I switched to my Spider to free up some tonnage for our PUG. I would rarely break 150 and/or get a kill, although I'd get 3-6 assists on average.
My match score would be 250+ though because of all the scouting, uav, ecm counterhit and runs, protection bonuses, etc. that is boosted my score enough to many times get an = even on a loss and a rare + so that all in all it was about 60/40 to either stay even/increase or decrease with decreasing being the 60.

I think people are putting way too much stock into certain things that don't matter as much as they think they do, taking out of context soundbytes from Paul's post, and using anecdotal evidence at best if they even truly bothered tracking stats. Yes, my evidence is anecdotal as well, but I don't claim it to be the factual representation of how PSR works, merely ADDING my anecdotal experience to the pool.
There's a big difference in submitting information and making sweeping declarations or definitive statements based on anecdotal information.

The way I understood Paul on this, is that players naturally progress up and winning obviously increases the points over a loss or draw, but that doesn't mean it's "heavily slanted" for that.
Paul specifically stated that a player, as they progressed, would begin to do well not so much based on individual scores, but on how well they filled a role on their team. That has nothing to do with match score, kdr, kills, wins, or losses (although I'm sure they all factor into the overall formula), and more to do with how you perform in supporting your team.

I'm still testing on assault mechs but I'm going to see is by dealing out lots of damage with lurms, grabbing tons of assists, and spotting bonuses if I seem to advance more often as opposed to doing less damage, fewer assists, and more kills.

I can't presume to factually know PGI's true formula, but I have a theory. It's based more on weight class, speed rating, and firepower rating.
If you look at your mech in the mech lab, it has that diagram that shows how highly it places in those areas such as heat and firepower. It then rolls all those numbers up and gives you a "role class".

If you're a light mech moving at 136/kph, with ECM, and only 2 MPL, you'll be weighted more on spotting, scouting, etc.

If you're an assault mech with a bunch of lurms and move 55/kph, you'll be weighted more on doing lots of damage, spotting, components destroyed, etc.

So on and so forth. Obviously that's just theorycrafting, but it's in part based on how I interpreted Paul's response regarding the PSR and how it worked.

He essentially says, players who learn to use teamwork, coordination, and effectively fill roles will advance at a much higher rate.
I have discovered that since I got my "feet wet" again, and found my groove, and adjusted to the current maps and play styles, my PSR started climbing through Tier 3 at a much higher rate. The odd thing that made me take notice was that my W/L wasn't really going up because honestly, we as individuals have little to no real effect on the W/L of a match sometimes, so I began to wonder how I had changed my style and adapted.

It wasn't so much that my "skill" improved, it was more that my tactical knowledge of the map, best positioning for me based on loadout, anticipating the enemy movements, watching the mini map to ensure I was caught in a bad spot much less of the time.
Trying to fill a role whether that be a brawler, sniper, support, scout, etc. and adapt to getting bonuses based on those roles. Anyhow, that's my .02 on PSR and how it works in my best guess.

#22 JC Daxion

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Liquid Metal
  • Liquid Metal
  • 5,230 posts

Posted 25 December 2015 - 06:26 PM

i agree with much of what has been said.....


that said.. As a guy that just plays pugs.. To me it's just not an XP bar.. sure, a new account getting out of the lower tier was easy.. But nearly topping out at top of tier 4 after 25 games.. it has gotton slower.. I am still at under 50 games, and just a shade bellow tier 3...


My IS account, on the other side, i have played 100's of matches and my tier has not budged at all.

I also only play pugs... Often play less than ideal mechs, mainly ones i just like.. Sorry if playing a commando 3A is not ideal, but man it can be fun! I also like to play a bit more aggressive, verse the peak and poke play so many enjoy. That all can nd probably is a contributer.. but as my clan side i have just been playing my Timber.. and plan on grabbing a nova, and then just getting 2 more of each, and playing those for a while.. just to see how it goes..

give me a few more months and i will have a much better handle on it.. I might even skip the nova for a while, just to see if playing a great mech, leveling up does feel like an XP bar..

#23 Araevin Teshurr

    Member

  • PipPipPipPipPipPip
  • The Warden
  • The Warden
  • 372 posts
  • LocationIn your base, eating your food!

Posted 25 December 2015 - 06:30 PM

So if you are in a match, do 1000+ damage, but your team gets wiped out, you take a loss to your standing?

#24 Votanin FleshRender

    Member

  • PipPipPipPipPipPipPip
  • Wrath
  • Wrath
  • 518 posts
  • Location3rd rock from the Sun

Posted 25 December 2015 - 06:46 PM

Paul (or Russ) said that the more people play, the better they get... ergo a slanted PSR system. Again, that's stupid. Doing something over and over again does not automatically imply skill improvement. One of my part time employees has been with me for over 10 years and is still an idiot. But he's a friend, too, so /shrug. And the people that DO get better over time would rise just as fine with a equally weighted PSR up/down system, rather than biasing it towards an increase. Having a bias only leads to people rising higher than they should.

#25 Simbacca

    Member

  • PipPipPipPipPipPipPip
  • Overlord
  • Overlord
  • 797 posts
  • LocationCanada

Posted 25 December 2015 - 07:17 PM

View PostVotanin FleshRender, on 25 December 2015 - 06:46 PM, said:

Paul (or Russ) said that the more people play, the better they get... ergo a slanted PSR system. Again, that's stupid. Doing something over and over again does not automatically imply skill improvement. One of my part time employees has been with me for over 10 years and is still an idiot. But he's a friend, too, so /shrug. And the people that DO get better over time would rise just as fine with a equally weighted PSR up/down system, rather than biasing it towards an increase. Having a bias only leads to people rising higher than they should.

Plus some players such as myself reach their max ability in a given game - where improvement is no longer possible. I am a solid Tier 3 player. Yet, because I was carried to T2 (plus during my break I played A LOT) as soon as I take heavy or assault mechs MM expects me to carry teams. Hence why my win-loss ratio for assaults and heavies are all 0.5 to 0.65 W/L -- effectively meaning I have only a 30% chance to win or less.





1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users