Jump to content

Psr Needs To Be Replaced


105 replies to this topic

#41 Kyynele

    Member

  • PipPipPipPipPipPipPip
  • The Vicious
  • The Vicious
  • 973 posts

Posted 29 February 2016 - 06:19 AM

View PostSavage Wolf, on 29 February 2016 - 05:52 AM, said:

Snowballing in 8v8 is even worse as each mech is a bigger percentage of the teams. If you want less snowballing, you need bigger teams.


It's called snowballing because the effect gets bigger the further the stomp goes. And it goes much further with bigger teams.

In 4 vs 4, the worst case scenario is to end up 1 vs 4.
In 8 vs 8, it's 1 vs 8
In 12 vs 12 it's 1 vs 12.

#42 zeves

    Member

  • PipPipPipPipPipPip
  • FP Veteran - Beta 1
  • 282 posts
  • LocationNorway

Posted 29 February 2016 - 06:21 AM

system is working alright at the moment, at some point they will implement Ranked games and all the people that care about that will go into that.
But sure why not spend another 3 years focusing on the matchmaker and reworking mech balances instead of adding more content.

#43 Mawai

    Member

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

Posted 29 February 2016 - 06:29 AM

View PostMrMadguy, on 29 February 2016 - 02:05 AM, said:

Snowball argument is proven to be false, sorry:
Posted Image


LOL. I think you need to do some reading about what "PROOF" and "FALSE" mean.

The statement was that the game design in MWO tends to result in rolls with results of 12: 3,2,1 and 0 being more frequent than more balanced results like 12:11,10,9,8

How can one particularly useless screen shot, in which you did not contribute to the match, in any way either prove or disprove the above statement?

1) ALL match results are possible to any ONE result is utterly and completely meaningless.

2) Rolls do NOT have to do with the contributions from individual players ... they are usually due to the tactical setup where one team literally rolls over or along the other with a focus fire advantage. Hitting the opposing team from the side or a long a line where your side gets a 4 or 5 to one firepower advantage against a series of strung out mehcs tends to be the most common scenario. Also, it usually takes 3 to 4 losses for it to begin to snowball (though there are occasions when my team has come back from down four ... it isn't common and those games usually wind up as close matches).

3) Anyway, all the changes PGI has ever made to the matchmaker have not prevented or eliminated stomps ... as a result, I would say the snowball theory is pretty much established fact. The closer the match is, the longer it takes to snowball, but once a side is down a few mechs, it gives the opponents more opportunities for focus fire (i.e. implicit team work) and the side that is down tends to lose more very quickly.

Finally, on the topic of "skill" and trying to get a "rating" for it. It isn't easy.

Here are some factors that affect a player's performance in a match ..
- mech choice
- loadout choice
- choice of playstyle
- experience of player with that playstyle
- ability to aim
- ability to use the loadout effectively
- ability to cooperate with team mates
- ability to assess the tactical information from the HUD and apply it to choosing your next actions

- selfishness ... do you choose to take actions that win the match for your team but that likely result in your death or do you farm damage from a distance even though you get a better match score but stand a better chance of losing?

The last item is one of the hardest to assess ... The GOAL of MWO is to win the match with your team.

Others seem to think the GOAL is to maximize their own personal score whether they win or lose. However, a high personal score doesn't necessarily contribute to winning. I spectated a Timberwolf that was on my team one time ... he was sniping into a melee, not so good aim, racked up lots of damage spread all over, when the melee broke up he was the last one left. He pulled off 4 kills and a bunch more damage and the remaining opponents tracked him down. He had a great match score and was a completely crap player since his efforts only contributed to losing and not to winning. On the other hand, there are others out there who use R, assess the damaged areas on a mech, have good aim, and eliminate or cripple opponents taking out CT/ST/Leg. These folks do relatively little damage ... maybe 300 to 500 ... but almost all of that was USEFUL. Their match score is often lower but these are often the folks who win the matches for their team.

Any scoring system that is based on damage/kills/assists is going to be broken to some extent since it encourages folks to farm damage to get higher scores. It doesn't actually reflect "skill" at winning the game.

The much maligned Elo actually assessed your contribution towards winning. On top of that it was conservative (when working properly) so that some would go up and some down. It also didn't take that many matches to get a decent idea of how much you contributed to winning. However, it needed more resolved data ... separate numbers for group and solo queue ... maybe separate numbers for each mech ... use a weight class number until there are enough statistics for a specific mech then switch to that.

PSR on the other hand is not conservative ... PSR is added to the system almost every match ... even changing that now won't change anything since the ones who play the most using damage farming techniques are already tier 1 while the ones who play less frequently are probably still in tiers 2,3,4. The only way to change things up would be to recalculate rating from the beginning based on the recorded match results when you apply a new formula ... that would reallydrive the folks who "worked" for tier 1 around the bend and generate a huge amount of player frustration,especially when folks in tier 1 find that isn't where they belong anyway.

So ... I expect the current quality of matchmaking to continue for better or worse ... since PGI has a awesome PR nightmare on their hands if they make any substantial changes to player ranking at this point ... and PGI likes to ignore anything that will rock the boat too much.

#44 Lugh

    Member

  • PipPipPipPipPipPipPipPipPip
  • The Widow Maker
  • The Widow Maker
  • 3,910 posts

Posted 29 February 2016 - 06:30 AM

View PostThunderbird Anthares, on 29 February 2016 - 02:41 AM, said:

match win/loss factors WAY too much into PSR to ever be considered a "skill" rating

its just a win/loss rating that is pretending to be something else - everyone hates it and everyone is attempting to avoid higher tiers to dont get tangled up with the tryhards and cheaters

This just in, Focus Fire and teamwork = cheating. Newsreel at 11 with 'proof'.

#45 DjPush

    Member

  • PipPipPipPipPipPipPipPip
  • Overlord
  • Overlord
  • 1,964 posts

Posted 29 February 2016 - 06:37 AM

Oooorrrrrr..... They could build a game that attracts more players.

#46 Thunderbird Anthares

    Member

  • PipPipPipPipPipPip
  • 392 posts

Posted 29 February 2016 - 06:40 AM

View PostLugh, on 29 February 2016 - 06:30 AM, said:

This just in, Focus Fire and teamwork = cheating. Newsreel at 11 with 'proof'.


dude, can you even read? Posted Image

#47 Crockdaddy

    Member

  • PipPipPipPipPipPipPipPip
  • Elite Founder
  • Elite Founder
  • 1,684 posts
  • Twitch: Link
  • LocationSaint Louis

Posted 29 February 2016 - 06:56 AM

View Posthimself, on 28 February 2016 - 09:32 PM, said:

The current implementation of PSR is lazy.

Not only is it NOT a good gauge for how much skill a player has, it only serves to segregate the new players and the ultra baddies from the old players and the tryhards. It needs to be removed and replace with a system that doesn't pretend to be keeping new players safe.

New Idea:

PXI [pik-see] (Pilot Experience Index): Works exactly like PSR, but doesn't pretend to be a skill rating. It serves as a decent indicator that a player has learned the basic mechanics of the game and can apply them with some modicum of efficiency.

Instead of locking tiers out, use PXI to balance the teams.

Current:

[5-3][4-2][3-1] Locked Tier MM

You're segregating your already limited player base.

Suggested:

Use PXI to ensure that both teams have a near-equal amount of newer players and experienced players to avoid total steam rolls.

Congratulations, now you also aren't screwing over people bringing in their friends by pitting them against tryhard groups every time they want to play with you, because they're PSR5 and you're PSR2.
You're welcome.



This post hurts my brain. I ... am not even sure the point of it. I am afraid to read more.

#48 CreativeAnarchy

    Member

  • PipPipPip
  • Bad Company
  • 62 posts

Posted 29 February 2016 - 07:04 AM

If the stats were not bugged, they would be more useful to determine a players skill, but since the game thinks I've only shot an AC/20 in 4 matches, my accuracy with that weapon, I don't believe, is accurate as well as others that I've used.

#49 MrMadguy

    Member

  • PipPipPipPipPipPipPipPipPip
  • 2,221 posts

Posted 29 February 2016 - 07:40 AM

I'm not against snowball effect theory - I'm against using it to justify terrible matches and to shield terrible matchmaker.

#50 Xmith

    Member

  • PipPipPipPipPipPipPipPip
  • The Ironclad
  • The Ironclad
  • 1,099 posts
  • LocationUSA

Posted 29 February 2016 - 07:51 AM

View Postc0de4014, on 28 February 2016 - 11:55 PM, said:

Pilot skill rating is crap, it ONLY goes up if your team wins... how does that judge an individuals skill level!? it is ridiculous...
i have a screenshot of a spreadsheet that shows in detail how ridiculous it is. the rating ONLY goes up if your TEAM wins a match...

https://gyazo.com/19...912ed91107d05c0

You PSR can rise when you lose. It just depends on how skilled you are.

http://imgur.com/GjFs8Rp

Edited by Xmith, 29 February 2016 - 07:54 AM.


#51 Sandpit

    Member

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

Posted 29 February 2016 - 09:00 AM

View PostMrMadguy, on 29 February 2016 - 02:05 AM, said:

Snowball argument is proven to be false, sorry:
Posted Image

uhm what exactly is this screen shot supposed to prove?
That your team carried you to a victory because you might as well not even been there based on your score? lol

#52 MrMadguy

    Member

  • PipPipPipPipPipPipPipPipPip
  • 2,221 posts

Posted 29 February 2016 - 09:20 AM

View PostSandpit, on 29 February 2016 - 09:00 AM, said:

uhm what exactly is this screen shot supposed to prove?
That your team carried you to a victory because you might as well not even been there based on your score? lol

Snowball theory adepts claim, that two team are perfectly balanced at the beginning of the match and that balance breaks, when first 1-2 'Mechs are being killed, which causes chain reaction due to losing team being outnumbered by winning one.

I claim, that stomps happen only due to matchmaker not working properly and building initially unbalanced matches.

What this screenshot proves? That initially match wasn't even near balanced. Enemy team had initial two 'Mechs advantage and according to snowball theory it's my team, that should have been stomped. But no! Enemy team was stomped instead! Conclusion is obvious.

And yes. Also this screenshot proves, that you can be carried - i.e. that there is no correlation between W/L and your skill. W/L = 1 may be easily achieved, even if your will be AFK in 100% matches. That's why ELO and any other W/L based MM - is inappropriate for anything else, than 1vs1 or constant team vs constant team matchmaking with static pool of players.

Edited by MrMadguy, 29 February 2016 - 09:23 AM.


#53 Savage Wolf

    Member

  • PipPipPipPipPipPipPipPip
  • The Wolf
  • The Wolf
  • 1,323 posts
  • LocationÅrhus, Denmark

Posted 29 February 2016 - 09:28 AM

View PostMrMadguy, on 29 February 2016 - 07:40 AM, said:

I'm not against snowball effect theory - I'm against using it to justify terrible matches and to shield terrible matchmaker.


I'm not trying to say the matchmaker can't be bad. I just wanted to tell people they shouldn't expect stomps to go away if the matchmaker gets good.

View PostXmith, on 29 February 2016 - 07:51 AM, said:

You PSR can rise when you lose. It just depends on how skilled you are.

Skilled at the things that PSR can measure and deems important that is.

#54 JC Daxion

    Member

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

Posted 29 February 2016 - 09:34 AM

Russ just stated in the town hall they are willing to look at it and have no issue tweaking it..

the problem to me, is your PSR should not go up in groups, only solo play.. Your premade group should be all about the highest PSR. This would solve a lot of issues IMO


Though honestly, i see many tier 1 and 2's that just are bad at the solo chaos. Maybe they are aces in match play, but often it does not translate. Solo play shows how well you can adapt, and if you really are a good player.. Because the good ones, show up and play well.



I never blame the match maker.. Most of my matches are close till about half way, and then one pulls out.. Rarer matches are the ones that i truly love, the ones were we come back 2-4 mechs down and win, Or the ones that come down to the wire.. My rarest matches are on the loosing side of a total roll. Not to say they never happen, but for me, they really are the rarer matches.. Outside of a random loosing streak.. but these days they typically end after a couple rolls for me.

Over all most of my matches in a night are in the 6-8 to 12 matches.. Maybe one of these days, ill go and track 20 matches over a night and post a chart.. But anyone can take out a screenie of a bad roll. or +1k damage, or 8 kills,, or what ever awesome match you had.. after a 2k+ matches everyone has um, they are just not that impressive for the most part, even though i have posted a few great games i have had, but it's more about.. whoopie.. i had a great match! Now back to your regularly scheduled programming.



View PostMrMadguy, on 29 February 2016 - 09:20 AM, said:



I claim, that stomps happen only due to matchmaker not working properly and building initially unbalanced matches.




total crap.. For proof all you need to do is go group up and play private matches.. exact same teams for a few hours, and you will see, even then people can have stomps, close games, comebacks.. ect..

You can see the exact point a stomp stats more often than not... Yet stop it with a better strat going forward. AKA "Stop the Stomp game play" which often means taking on a brand new strat, and communication.. I had 3 matches the other day, that we pulled back and won, by some great game play. Glame the match maker all you wan't, but people never want to admit that they could of won by doing a few simple changes.. I see matches that we could of easily rolled, and then all a sudden we are down 4 mechs, because of a simple thing, and its game over.. How is that the match makers fault?

Edited by JC Daxion, 29 February 2016 - 09:41 AM.


#55 Idealsuspect

    Member

  • PipPipPipPipPipPipPipPipPip
  • The 1 Percent
  • The 1 Percent
  • 2,127 posts

Posted 29 February 2016 - 09:40 AM

" PSR need to be replaced " with a totally new futur epic fail system?


In fact this game need dedicated servers for newbies,
Then when they are vets put them randomly in each team.
BECOSE CHAOS IS SMARTER THAN PGI ( and chaos knew this game better than them )

>>> fair system, some stomps rolls will appear of course but it will be only chaos's fault :)

And concerning CW, people should be able to join only WHEN they have 4 mechs totally mastered for build a proper dropdeck ( and new people will less left the game with some goals like this ).

Oh yea this post will be never read by people in charge also whatever xD

#56 Savage Wolf

    Member

  • PipPipPipPipPipPipPipPip
  • The Wolf
  • The Wolf
  • 1,323 posts
  • LocationÅrhus, Denmark

Posted 29 February 2016 - 10:08 AM

View PostMrMadguy, on 29 February 2016 - 09:20 AM, said:

Snowball theory adepts claim, that two team are perfectly balanced at the beginning of the match and that balance breaks, when first 1-2 'Mechs are being killed, which causes chain reaction due to losing team being outnumbered by winning one.


We claim that the game is prone to it. Doesn't have to be the first 2 mechs that does it, but if the gap becomes too great, it will escalate. Even perfect matchmaking will not change this and if this is what you are waiting for, then you will be disappointed.

View PostMrMadguy, on 29 February 2016 - 09:20 AM, said:

I claim, that stomps happen only due to matchmaker not working properly and building initially unbalanced matches.

Making you prone to confirmation bias. So far you have simply concluded that all stomps are the work of the matchmaker and that all non-stomps do be good matchmaking. You only seek to prove your theory and find your truth.

View PostMrMadguy, on 29 February 2016 - 09:20 AM, said:

What this screenshot proves? That initially match wasn't even near balanced. Enemy team had initial two 'Mechs advantage and according to snowball theory it's my team, that should have been stomped. But no! Enemy team was stomped instead! Conclusion is obvious.

Again, it's only one data point and the fact that you jump directly to your desired conclusion suggests that you know nothing about how to avoid bias or how to read data and the pitfalls their in.

View PostMrMadguy, on 29 February 2016 - 09:20 AM, said:

And yes. Also this screenshot proves, that you can be carried - i.e. that there is no correlation between W/L and your skill. W/L = 1 may be easily achieved, even if your will be AFK in 100% matches. That's why ELO and any other W/L based MM - is inappropriate for anything else, than 1vs1 or constant team vs constant team matchmaking with static pool of players.

But this game is a constant team battle. We all have played plenty of matches. That's why ELO does work in the long run. You will always be able to be carried, but with time the bad players will be filtered out. So it's not ideal, it's not efficient, but at least it's accurate contrary to PSR.

Edited by Savage Wolf, 29 February 2016 - 10:42 AM.


#57 Napoleon_Blownapart

    Member

  • PipPipPipPipPipPipPipPip
  • Shredder
  • Shredder
  • 1,169 posts

Posted 29 February 2016 - 10:25 AM

you will be graded on performance in this match
Posted Image

all systems nominal

Posted Image

carry harder

#58 Sandpit

    Member

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

Posted 29 February 2016 - 10:33 AM

View PostMrMadguy, on 29 February 2016 - 09:20 AM, said:

Snowball theory adepts claim, that two team are perfectly balanced at the beginning of the match and that balance breaks, when first 1-2 'Mechs are being killed, which causes chain reaction due to losing team being outnumbered by winning one.

I claim, that stomps happen only due to matchmaker not working properly and building initially unbalanced matches.

What this screenshot proves? That initially match wasn't even near balanced. Enemy team had initial two 'Mechs advantage and according to snowball theory it's my team, that should have been stomped. But no! Enemy team was stomped instead! Conclusion is obvious.

And yes. Also this screenshot proves, that you can be carried - i.e. that there is no correlation between W/L and your skill. W/L = 1 may be easily achieved, even if your will be AFK in 100% matches. That's why ELO and any other W/L based MM - is inappropriate for anything else, than 1vs1 or constant team vs constant team matchmaking with static pool of players.

This is why you can't take the vast majority of "stats" and "evidence" provided on the forums as proof of anything.

You just tried to prove your theory and disprove another based on a single match screenshot

#59 Aleksandr Sergeyevich Kerensky

    Member

  • PipPipPipPipPipPipPipPip
  • 1,187 posts

Posted 29 February 2016 - 10:51 AM

I personally thought the PSR system was just used to seperate players from those who need to learn the basics of mech piloting, those who have learned basics, and those who mastered the basics?

Going by those standards, I think PGI has actually done a pretty good job (please keep in mind- I generally complain a lot). I have experienced many more enjoyable games when i que solo latelly, much better then before at least.

For group play- I generally expect stomps to swing hard, but thats to be expected

#60 MrMadguy

    Member

  • PipPipPipPipPipPipPipPipPip
  • 2,221 posts

Posted 29 February 2016 - 10:53 AM

View PostSandpit, on 29 February 2016 - 10:33 AM, said:

This is why you can't take the vast majority of "stats" and "evidence" provided on the forums as proof of anything.

You just tried to prove your theory and disprove another based on a single match screenshot

No, my stats, I provided in this thread too, prove my theory too. My performance - is 200dmg and 600XP per match. My survival rating - is just 17%. I die very early in most matches and don't contribute anything towards victory. But my W/L is still the same 1, as it was during Open Beta, when all my matches were like this:
Spoiler

I was carrying matches in the past and I'm being carried now and it's obvious. And if I was carrying then and I'm being carried now - then W/L has nothing to do with skill and performance. And if W/L has nothing to do with skill and performance - then PSR is complete mess, that isn't better, then random matchmaking. And if PSR is almost random matchmaking - then all obviously unbalanced games are caused by it.

You may continue ignoring clear evidence for as long, as you want. Truth won't turn into false, if you will ignore it. I really hope, I will finally quit this game on this Wednesday. And I hope, that this time I'll never return.

Edited by MrMadguy, 29 February 2016 - 10:54 AM.






3 user(s) are reading this topic

0 members, 3 guests, 0 anonymous users