Jump to content

Psr Update And Changes - Jun 2020


490 replies to this topic

#21 Keiron Halcyon

    Rookie

  • The Warrior
  • The Warrior
  • 8 posts

Posted 04 June 2020 - 10:49 AM

View PostKrasnopesky, on 04 June 2020 - 10:17 AM, said:

Excellent decision to reset the tiers.

However the changes to PSR in regards to zero sum completely miss the mark.

I have one question for you:

Why should a player who wins a match with a matchscore of 0 receive the same PSR change as a player who loses with a matchscore of 600?

THIS! THIS RIGHT HERE!!!!

#22 pvt Hudsoff

    Rookie

  • Bad Company
  • 6 posts

Posted 04 June 2020 - 10:50 AM

Thanks PGI for this decision! Finaly we're getting somewhere.

View PostDogmeat1, on 04 June 2020 - 10:23 AM, said:


Player LOSES:
IF Matchscore 401 * (PlayerPSR/MedianPSR) THEN no PSR change
ELSE Matchscore 251 * (PlayerPSR/MedianPSR) THEN PSR goes down by -1
ELSE Matchscore 101 * (PlayerPSR/MedianPSR) THEN PSR goes down by -3
ELSE PSR goes down by -5

Player WINS:
IF Matchscore 401*(PlayerPSR/MedianPSR) THEN PSR goes up by +5
ELSE Matchscore 251*(PlayerPSR/MedianPSR) THEN PSR goes up by +3
ELSE Matchscore 101*(PlayerPSR/MedianPSR) THEN PSR goes up by +1
ELSE PSR doesn't change



This is a good idea, but I would rather slightly change the numbers to:

Player LOSES:
IF Matchscore 401 * (PlayerPSR/MedianPSR) THEN PSR +1
ELSE Matchscore 251 * (PlayerPSR/MedianPSR) THEN PSR no change
ELSE Matchscore 101 * (PlayerPSR/MedianPSR) THEN PSR goes down by -1 (or -3?)
ELSE PSR goes down by -3 (or -5)

Player WINS:
IF Matchscore 401*(PlayerPSR/MedianPSR) THEN PSR goes up by +3 (or +5)
ELSE Matchscore 251*(PlayerPSR/MedianPSR) THEN PSR goes up by +1 (or +3)
ELSE Matchscore 101*(PlayerPSR/MedianPSR) THEN PSR no change
ELSE PSR -1 (or -3)

Just to reward players who actually done something for the team to win.

Edited by pvt Hudsoff, 04 June 2020 - 10:55 AM.


#23 TheLaserTherapist

    Rookie

  • Bridesmaid
  • Bridesmaid
  • 2 posts

Posted 04 June 2020 - 10:51 AM

So: like the idea of a zero sum PSR. Not quite like this though. As others have said, a really good matchscore in a losing game should influence PSR positively.

As laid out in the current post, the PSR would not become a representation of player skill, but of "frequency of overall good teams". Except for extreme cases, a single player can not win a match by themselves.

Here is a score screen of a match I was just in that I'd say felt "good" Posted Image

Notice something? Every single players MS on the losing team is under 400. Just under, but under. I refuse to believe that every single one of those 12 players was the reason the team lost. Notice something else? Every single player on the winning team has a matchscore over 150, nearly all of them over 200.

Allowing a PSR decrease on win for MS < 100 and an increase for loss with a high matchscore (doesn't have to be 400, could be higher - those numbers are always tweakable) would make for a somewhat fairer matchmaking, which could be kept zero sum by tweaking the exact numbers around a bit.

#24 Firewired

    Member

  • PipPip
  • Trinary Star Captain
  • 31 posts

Posted 04 June 2020 - 10:54 AM

Thanks for the follow-up and background information, much appreciated.

Couple of things...

1. If I were to make a suggestion, base your statistical analysis on a larger sample size than just 24-hours.

Human beings, like it or not, live on a week-to-week schedule. We ran into challenges at work with statistical analysis pre-vs-post implementation because the Architect (who was NOT an Engineer) based his results analysis on a Monday result vs. a Tuesday result. He had not collected sufficient pre-change data to compare versus post-change. We insisted the changes be rolled back (they were because we ARE Engineers and we have authority over Architects).

The data was then gathered for a full 7-day period, then changes re-implemented and the data gathered to be compared day-vs-day (as in Monday pre- vs Monday post-implementation, same for Tue-Tue, Wed-Wed, and so on).

If you do anything at the end or at the start of a new month, then only compare that data to those same dates next month, as human beings do chaotic things near the end and start of a month that can throw data analysis off compared to other dates of the month.

This method generally collects and compares reliable data consistently.

Yes, it takes longer to gather and compare the data, but the more reliable data results are worth the time, and it does not require any more work effort than shorter duration data comparisons if automated.

2. I think (I may be wrong) people were hoping zero-sum would mean Win vs. Loss would be completely ignored in PSR movement per-match.

While your proposed changes technically fit into the definition of zero-sum, they still reward a player for performing poorly and winning, and penalize a player for performing well and losing.

MWO already has a system for rewarding poor performance and winning: Cbills. They get more Cbills if they win regardless of their performance. That should really be enough.

PSR is a Pilot Skill Rating of an individual Pilot's performance. It is not a Pilot Win Rating.

If you feel that you HAVE to reward players for winning then add the Pilot Win Rating to their profile page and remove the Tier Rating, that should make them happy.

#25 Pseudo98

    Member

  • PipPip
  • Ace Of Spades
  • Ace Of Spades
  • 49 posts
  • Twitch: Link
  • LocationUK

Posted 04 June 2020 - 10:59 AM

I would say a token gesture of Match Score: 600+ goes up by PSR+1 on a loss would go a long way.

That's 1000+ dmg and a few kills, that has to be rewarded on a loss imo.

Edited by Pseudo98, 04 June 2020 - 11:00 AM.


#26 Dionnsai

    Member

  • PipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 469 posts

Posted 04 June 2020 - 11:02 AM

Too many people lost in the details. This is hands down a massive improvement over the current system, flawed or not.

Edited by Dionnsai, 04 June 2020 - 11:02 AM.


#27 PrometheusTNO

    Member

  • PipPip
  • Star Captain
  • Star Captain
  • 43 posts
  • LocationUSA

Posted 04 June 2020 - 11:04 AM

Hi Paul,

The change is a positive direction to be sure. What if, instead of raw match score thresholds (401/251/101), it was a sliding scale based on the average match score of the players on your team? This will bring the PSR adjustment in line with external match factors (i.e. losing team was crazy stomped, or high-tonnage matches are always higher scoring because of all the armor you have to chew through, etc).

Thanks for trying these different things lately to keep MW:O alive and kicking.

#28 Paul Meyers DEST

    Member

  • PipPipPipPipPipPipPip
  • Warrior - Point 3
  • Warrior - Point 3
  • 543 posts

Posted 04 June 2020 - 11:04 AM

View PostDionnsai, on 04 June 2020 - 11:02 AM, said:

Too many people lost in the details. This is hands down a massive improvement over the current system, flawed or not.


True, but still the weirdo-wrong-way. Bad players get easy carried in higher tiers.

#29 Arjohan

    Member

  • PipPipPip
  • Cadet
  • Cadet
  • 94 posts

Posted 04 June 2020 - 11:06 AM

Some idea. This could address having a terribly bad match in a loss where you just can't do much while also addressing the "no-help-to-the-team people loses by not contributing" in a win. Altho, this does add a 5th level... But should be good enough to what we get for. Open to opinions as well

Win:
0   ~ 99	  - -1
100 ~ 199	  - 0
200 ~ 324	  - +1
325 ~ 399	  - +3
400 ~ infinite	  - +5




Lose:
0   ~ 99	  - -5
100 ~ 199	  - -3
200 ~ 324	  - -1
325 ~ 399	  - 0
400 ~ infinite	  - +1



Edit: fixed formatting, I hope.

Edited by Arjohan, 04 June 2020 - 11:10 AM.


#30 Zanotam

    Member

  • Pip
  • Legendary Founder
  • Legendary Founder
  • 16 posts

Posted 04 June 2020 - 11:12 AM

Not ot be devil's advocate or anything, but I don't think the people commenting in here are familiar with how MMR/ELO/PSR works in OTHER games AND HOW IT WAS DESIGNED TO WORK FROM THE BEGINNING WHEN ORIGINALLY INVENTED. It's a tool to measure natively just pure wins/losses but instead of a w/l ratio it gives you a number based upon how you were rated when you beat people at the ratings they were at at the time. It's a real-world skill ranking designed for 1v1 games. However, it scales up just fine afaik into 5v5 or 8v8 or presumably 100v100 let alone 12v12 because in the end hte only consistent statistical element between games is the one player and that player is further known to be given a 'relatively fair' match via matchmaking otherwise you couldn't even due it based upon match score because hey what if a tier 5 gets matched with a tier1 of course they won't do as well.

If anything, I think the focus on your match score is questionable - as mentioned AMS and LRM mechs are a great way to farm that number and I probably will spend some time focusing on my LRM plus my best performing non-LRM mech as the only mechs I play initially after the reset to get a relevant skill rating for the most part considering match score apparently matters which is already much more 'personal' than a normal system and really all that should be expected. Letting someone go up when they lose just ***** the system and turns it back into an exp bar.

#31 Magic Pain Glove

    Member

  • PipPipPipPipPip
  • CS 2023 Gold Champ
  • CS 2023 Gold Champ
  • 152 posts

Posted 04 June 2020 - 11:17 AM

I also think the addition of 5th level is needed for more accurate ranking / PSR .
Low performing winners should be penalized slightly , and good performers on a loss should be rewarded . The addition of 5th level helps with that like described in Arjoe's post .

If you do not wish to add the 5th level (Not enough ,time , effort , money or will). With only 4 levels I would do

Quote

(Win) -1 +1 +3 +5 (Loss) -5 -3 -1 +1

even tho it eliminates neutral ( 0 ) no gain no loss of PSR , I still think it would be better than what was proposed . Which is still a system which is somewhat upwards biased ... again ...

Bad performers on a winning team need to be punished and good performs on a bad team should be rewarded .


View PostArjohan, on 04 June 2020 - 11:06 AM, said:

Some idea. This could address having a terribly bad match in a loss where you just can't do much while also addressing the "no-help-to-the-team people loses by not contributing" in a win. Altho, this does add a 5th level... But should be good enough to what we get for. Open to opinions as well

Win:
0   ~ 99	  - -1
100 ~ 199	  - 0
200 ~ 324	  - +1
325 ~ 399	  - +3
400 ~ infinite	  - +5




Lose:
0   ~ 99	  - -5
100 ~ 199	  - -3
200 ~ 324	  - -1
325 ~ 399	  - 0
400 ~ infinite	 - +1



Edit: fixed formatting, I hope.


#32 MODOK69

    Member

  • Pip
  • The Corsair
  • The Corsair
  • 11 posts

Posted 04 June 2020 - 11:17 AM

I believe the mark has been missed on resetting the PSR! Match score not W/L should be the standard. Re-look what goes into a match score if need be, but if you want a definitive PSR versus pilot skill win/loss rating. The current method described in this forum means i can drop,have the team carry me and still advance, not cool!. Using the match score means i actually have to actually have to perform!! Please re-look this before you pull the trigger otherwise not sure that you accomplished what you set out to do. Oh, how does a group drop work with the new PSR? Everybody on the in the group has to have the same PSR tier 3 to drop in a tier 3 match? Or is that going out the window like solo que just to support group play?

Edited by MODOK69, 04 June 2020 - 01:03 PM.


#33 Dionnsai

    Member

  • PipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 469 posts

Posted 04 June 2020 - 11:19 AM

View PostPaul Meyers DEST, on 04 June 2020 - 11:04 AM, said:


True, but still the weirdo-wrong-way. Bad players get easy carried in higher tiers.


Maybe here and there, but bad players also pull their team down. You don't see truly bad players with higher than 1.0 win/loss. They make their teams lose, which will mean they net negative PSR. You're not going to see a bad pilot make it to tier 1 by being carried there over 100's or 1000's of games.

#34 PrometheusTNO

    Member

  • PipPip
  • Star Captain
  • Star Captain
  • 43 posts
  • LocationUSA

Posted 04 June 2020 - 11:21 AM

View PostMODOK69, on 04 June 2020 - 11:17 AM, said:

The current method described in this forum means i can drop, power down, have the team carry me and still advance, not cool!


You can do nothing and get >100 match score?

New method:



Quote

Player WINS:

Match Score: 0-100 does not move.


#35 Joeseph Pierce

    Member

  • PipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 107 posts
  • LocationTerra

Posted 04 June 2020 - 11:22 AM

I have to agree with the general theme of the thread so far. If PSR is supposed to truly represent pilot skill, then there has to be the ability to gain PSR on a loss and lose PSR on a win. At least in my experience, 400+ MS is almost always a sign of someone who did well, and <100 MS is almost always a sign of someone who did poorly.

Winning and losing is incredibly team dependent and doesn't typically reflect on each individual player. A player who can pull 400+ MS on a loss is probably pretty decent at the game, more so than their teammates. Similarly, a player who manages <100 MS on a win probably did absolutely nothing to contribute for it. I've done it a few times personally, and in my experience when I manage to score less than 100 on a win, I did absolutely nothing to determine the outcome of the game. Even when I'm so slow I get eaten by the NASCAR in a 45kph assault I get more than 100 MS.

One counter I can immediately think of is "what if someone helps their team win by capping instead of fighting, in assault or conquest for example?" Does that really reflect on that pilot's skill in a game that is primarily about stompy robots blowing each other up? In my experience most maps are small enough that one or two players capping instead of fighting will probably lose the game for their team, because despite the appearance of different game modes most play out as "skirmish with a secondary win condition". Conquest at least typically delays the main engagement by a minute or two as each side makes sure that they own their two "home" bases, but generally 2 people capping means your team is fighting 10v12 and will probably lose the fight because of you. On a couple of maps, namely Polar and Alpine, capping is able to win a game simply because the map is so large that by the time the engagement is won the winning team doesn't have time to secure the objectives. To my mind, that doesn't mean the capping players are good at the video game, they just didn't get punished for leaving their team to die.

I've seen many suggestions for how to bracket the PSR benefits, and a common thread between them is <100 MS on a win loses PSR, and >400 MS on a loss gains PSR, and I think that's a very good way of rewarding players who did well despite losing the game, and for punishing players who didn't carry their weight on a win.

#36 Woopass

    Member

  • PipPipPipPipPipPip
  • Fire
  • Fire
  • 209 posts
  • LocationTacoma, WA

Posted 04 June 2020 - 11:27 AM

Dang, I just finally got to tier 1. I don't have a lot of time to play due to school and work. Oh well, this actually might be productive to match maker so lets see what happens :)

#37 Uakari

    Member

  • PipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 116 posts

Posted 04 June 2020 - 11:33 AM

View PostDionnsai, on 04 June 2020 - 11:02 AM, said:

Too many people lost in the details. This is hands down a massive improvement over the current system, flawed or not.


I too think that with a great matchscore on a loosing team you should move up more, but it s not really important.
Fact is: You dont move up with no skill. Until now you could hit enemies 4 times and die, and you d still move up.

Even Lrm boats require some skill concerning positioning and getting locks, and AMS is helpful so high lrm/ams matchscores are no problem i think.

#38 Dionnsai

    Member

  • PipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 469 posts

Posted 04 June 2020 - 11:34 AM

View PostJoeseph Pierce, on 04 June 2020 - 11:22 AM, said:

I have to agree with the general theme of the thread so far. If PSR is supposed to truly represent pilot skill, then there has to be the ability to gain PSR on a loss and lose PSR on a win. At least in my experience, 400+ MS is almost always a sign of someone who did well, and <100 MS is almost always a sign of someone who did poorly.


I think there is a slight EGO element in all of this. People are too worried about PSR gain/loss in individual matches. Across large sets of games, that just isn't going to matter.

Think of it instead in this way. Each pilot has an impact on their match:

Pilot A has a higher win/loss ranking
Pilot B has a lower win/loss ranking
Pilot C has a neutral win/loss ranking

Yes, maybe across 1-10 matches those numbers mean very little, you have bad days and good days. But across 100's of matches and 1000's of matches they mean alot.

If a pilot's mere presence tends to cause their team to win more than lose, then they should rise in psr.
If they have a neutral effect, then they're probably rated where they should be.
If the reverse is true, then they should drop in psr.

This is not something that you can measure in one individual game, it is the net result of scoring across many many games.

Don't worry about whether you did 600 match score in a loss and didn't get anything. If you're scoring 400-600 match score in most games, you're also going to be winning more games than you lose and you will have net + psr.

I also want to reiterate, NO ONE is going to carry a truly awful pilot out of their tier. You can't discount the amount of drag they apply to their team.

Edited by Dionnsai, 04 June 2020 - 11:36 AM.


#39 OneTeamPlayer

    Member

  • PipPipPipPipPipPip
  • Ace Of Spades
  • 399 posts

Posted 04 June 2020 - 11:38 AM

1. I'd use this time to tweak out the best mechs you can muster.

2. Since we're going to do the PSR reset with group queue still pushed into solo queue the best option for solo droppers is to be forced into finding a unit (suboptimal) or simply not playing for about a week starting the day the PSR reset occurs to avoid the first wave of super-units smashing through the queue (suboptimal).

If you're going to reset PSR and attempt to reseed everyone with an individual PSR Please turn off group queue for the week and make ranking such that one cannot finish the placement matches in a group drop.

#40 Nearly Dead

    Member

  • PipPipPipPipPipPip
  • Philanthropist
  • 274 posts

Posted 04 June 2020 - 11:38 AM

Teamwork = win

Who is guaranteed to be working as a team? Premades and units.

Win = PSR +

With all of the suggestions made here in the forum, this is what they do? It is like there is some kind of overriding game ideology here that PGI can't let go of, like they are living in some kind of other reality.







4 user(s) are reading this topic

0 members, 4 guests, 0 anonymous users