Jump to content

Psr Community Feedback - Round 1


357 replies to this topic

#41 AdmiralAmazing

    Rookie

  • Survivor
  • Survivor
  • 5 posts
  • LocationGermany

Posted 17 June 2020 - 01:58 AM

I vote Core 2B with JayZ's modifications.

ZeroSum is easily implemented and displayed in JayZ's provided reference document. Zero sum is a requirement for total change of PSR values, not for x players move up while also x players move down.
A minor boost in PSR for top performing players on the loosing teams side will overall increase the strive to perform at your best potential at all times and reduce griefing by a substantial amount.

When presented well 2C should be taken into consideration. The community probably only has this chance for now to voice concerns on this fundametal level.


Testing procedure proposal:

PGI could implement versions 1 and 2 alongside each other in the background in addition to the current system and collect data for each. This data can then be evaluated before making a change to the actual system in production based on thorough testing.

Edited by AdmiralAmazing, 17 June 2020 - 02:01 AM.


#42 Aivazovsky

    Member

  • PipPipPipPipPipPipPip
  • The Dragoon
  • The Dragoon
  • 801 posts

Posted 17 June 2020 - 02:00 AM

Let's try 1A. But 2B is also an interesting model.

Edited by Aivazovsky, 17 June 2020 - 02:00 AM.


#43 General Solo

    Member

  • PipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • 3,625 posts

Posted 17 June 2020 - 02:38 AM

View PostCapt Deadpool, on 16 June 2020 - 11:14 PM, said:

This is a good point, Homer, and hopefully addressing the group vs. solo issues will come after, such as having a separate PSR for each player depending on whether they are solo, or in a 2-man, 3-man, or 4-man, or at least a PSR modifier depending on whether someone is grouped or not like Dauntless suggested if someone is in a good vs. bad chassis.

Otherwise, maybe Tier 1 just ends up being the 'grouped tier', which maybe that is fine? Maybe it's bad, I don't know. Hell, we could just rename tier 1 'grouped tier', and tier 2 becomes tier 1, and kudos to solo players who achieve 'grouped tier' lol.

Or, think of it like this, if 1A or 1B is used, that still isn't unmerging groups and solo, so solos will probably play against groups less often if 2A or 2B is used.


My thoughts are theirs the real possiblity we won't get other chances so this change has to have a big impact in leveling the battlefield in order to firstly retain and maybe grow players

Firstly in my experience if groups are the tier 1 and dominate the queue then it dies like every other group dominated que.

I have two points why I disagree with using win/loss, all this assumes a feasable population or expert team construction with low populations.

Two points

1) By using Win/Loss you are effect causing a downward PSR bias simply because a player is solo, not in a group.
Its like before with poor performing on a win bias but a diferent biased, but still biased!

You may say its a team game and your right, but listen. MATCHMAKER's don't care about team game its goal is produce a metric called PILOT SKILL RATING that ranks a pilot against all other pilots, groups are not mentioned.
Its logic not maths.

We never had a proper functioning group MM other wise it wouldn't be a dead queue. So we are using and modifiying the solo one. But any way biased against soloes when its supposed to be a pilot skill metric, groups had another MM that was crap.

Anyway point two

And if a solo pilots PSR does not reflect their skill simply because they are not in a group, IMO by using win loss you are in fact causing MM to forgo using the very soloes that can best counter groups as far as soloes can counter them, when a suitably high PSR group is not available. Giving groups a further advantage.

imo

Edit: In my experience being in a group don't magically make you a better player. You still the same player.

I think the game should track how many group games a player does, so they can be ignored when they solo and then complain on the forums about matchmaker and losing PSR.

Tell me my two points are wrong, cause I can't see it.

Edited by OZHomerOZ, 17 June 2020 - 03:10 AM.


#44 Bistrorider

    Member

  • PipPipPipPipPipPip
  • The Howl
  • The Howl
  • 273 posts

Posted 17 June 2020 - 02:55 AM

Maybe this will help some to make a decision? It was very gg close. I did few simple colored brackets on match scores. Both teams had some very good players. Both teams could win that match. There was even some teamplay on both sides :) Note Vapor Eagle score on red team, this with the highest score from all the players. I recorded that match, so I know how this guy was playing - it was a top play for me. And his score showing that in fair way. Why he shouldn't get any points? I think he should go up. Match scores without brackets are those from grey area ;)


https://photos.app.g...Lqc72t5eh3p1jC8

#45 Sitzheizung

    Rookie

  • 3 posts

Posted 17 June 2020 - 02:56 AM

TLDR: vote fore JAYZ variant 2C
rest is not viable


As far as i understand, option 1A and 1B are only about building Matchscore, which is HEAVILY determined by creating as much DMG output as possible, (not shooting for CT but LT/RT etc....).
For what reason shall i even bother for the games winning objectives? If you implement either one of this methods you can delete all other gamemodes but Skirmish from the game. They will be obsolet...
IF ANY, i suggest a total overhaul of the matchscore boosters, reducing the DMG value and a strong increase for serving the winning conditions (capping in assault and concquest f.e.)

Imho it has to be a 2A, 2B OR 2C method of calculation.
And Zero Sum is very well possible to achieve at those methods...
And Zero Sum shall reference to the amount of matchscore that is added or substracted in total and not on how many players will gain or lose matchscore (f.e.: 20 players gain each 1 MS and 4 players loose each 5 MS = zero Sum <- everything perfect)

I like JayZ`s variant the most, its logic, its rewarding, it can support all gamemodes the best way.

edit: i like the idea from Axys Rageborn to add matchscore for "surviving", this would help Assault Mech Tanks a bit and encourages passive long range players to join the fight a bit more...

edit: actually i switch to variant JayZ 2C. its best

Edited by Keldomet, 17 June 2020 - 08:52 PM.


#46 Bud Crue

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Rage
  • Rage
  • 9,943 posts
  • LocationOn the farm in central Minnesota

Posted 17 June 2020 - 03:14 AM

I'm in Zeph's/Gagis's camps:

On the one hand, I have come to agree that personal performance, especially the "I carried on a loss, I should get a +" and "why should a poor performance on a win get a +" circumstances, just doesn't matter. But regardless of whether or not a given system provides for this sort of thing, just pick one system, and get on with it.

#47 AdmiralAmazing

    Rookie

  • Survivor
  • Survivor
  • 5 posts
  • LocationGermany

Posted 17 June 2020 - 03:16 AM

View PostBistrorider, on 17 June 2020 - 02:55 AM, said:

[...] Why he shouldn't get any points? I think he should go up. [...]


He definetely should. A well chosen example list.

#48 East Indy

    Member

  • PipPipPipPipPipPipPipPip
  • The Hammer
  • The Hammer
  • 1,244 posts
  • LocationPacifica Training School, waiting for BakPhar shares to rise

Posted 17 June 2020 - 03:27 AM

View PostZephrym, on 16 June 2020 - 10:38 PM, said:

Personally this is now a case of ........get on with it! We've essentially already discussed the pros and cons for the two core options and whilst both have their supporters and detractors, we are in danger of covering old ground.

Paul, you have identified your preferred system and been open to trying the other core options over time. This is great.

Agreed. No need to dither any longer. Start with 1A ASAP, observe and assess.

#49 Ragedog4

    Member

  • PipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 118 posts

Posted 17 June 2020 - 04:12 AM

My vote is this, 1A but the match score points towards more team support actions are greatly increased. K/D and solo kill is reduced to a little score, and damage done is about in the middle (can't be high because then tons of LRM boats would just sit in the back), this discourages people from just wanting to go solo on fighting, sitting in the back and not Sharing armor, and working as a team for score.

Punish those who use thier team as meat shields to just get match score, reward those who help the team. Perhaps rewards for more attacking the same target as a teammate, for flanking, for taking more damage as a tank up front. Reward heavy the team effort without rewarding the "win" where a bad team or match doesn't affect your individual score.

#50 Knight Captain Morgan

    Member

  • PipPipPipPipPipPip
  • Bad Company
  • 340 posts

Posted 17 June 2020 - 04:36 AM

Discussing how to calculate PSR/ELO/whatever is all fine & dandy but won’t affect match quality unless and until the matchmaker starts making good use of that number.

#51 Belorion

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 5,469 posts
  • LocationEast Coast

Posted 17 June 2020 - 05:55 AM

W/L can always be accounted for in the MS adjustment, personally I think the level now is fine.

#52 Krasnopesky

    Member

  • PipPipPipPipPipPip
  • CS 2022 Gold Champ
  • CS 2022 Gold Champ
  • 217 posts

Posted 17 June 2020 - 06:30 AM

I vote Core 2B with JayZ's modifications.

#53 Dogmeat1

    Member

  • PipPipPipPipPip
  • CS 2020 Gold Champ
  • CS 2020 Gold Champ
  • 123 posts
  • LocationButte Hold

Posted 17 June 2020 - 06:46 AM

I'm voting for the Core 2B win/loss system with revisions to the matchscore formula to make it more balanced. Separating win/loss into PSR, will make it easier to focus on making matchscore just represent in-game performance/contributions.

It should be pointed out that core 1B and core 2B are actually zero-sum if that is Jay Z's proposal you are referring to.

#54 Jay Z

    Member

  • PipPipPipPipPipPip
  • Deadset Legend
  • Deadset Legend
  • 436 posts
  • Twitch: Link
  • LocationAustralia

Posted 17 June 2020 - 06:51 AM

v v v v v v v v v v v v v v v v v v v v v v

https://docs.google....dit?usp=sharing
^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^
Righto you lot. Here it is. Current vs Russ Post vs 1A vs 1B vs 2A vs 2B vs 2C vs Somaru vs Kami vs Kami Tweaked.

I can post the maths for each but essentially I have collated a bunch of suggestions so you can see how they work. Even better YOU can put any MS values from your real matches or even synthetic results in the Orange Cells with White text.

2C is my favourite and adds a C value which is a weighted average factor of Team AVG MS and Match AVG MS. 1B has an inherent C value of 0 since it looks at the entire Match AVG MS and does not care about winning or losing at all. 2B has an inherent C value of 1 since it looks at the Team AVG MS but does not look at the other team's results. 2C uses a weighted average by the factor of C (0.6) which gives the best of both worlds in rewarding winning while punishing losing while smoothing the boost effect of performing strongly on the losing team. I am happy to answer any questions.

For those that won't click the link. Here is a screenshot using some values from a real match. Note the big numbers in the light blue row. ALL suggested solutions are Zero Sum unlike the current system.

Posted Image
There you go. Have fun!!!!

One final thing to mention. The maths behind 2C is identical to 1B and 2B, just different values set. Want to check? OK. Change the 2C variables to (X=5, C=1) and you will have 2B. Change the 2C variables to (X=0 C=0) and you will have 1B. :D

Edited by Jay Z, 17 June 2020 - 07:14 AM.


#55 Jay Z

    Member

  • PipPipPipPipPipPip
  • Deadset Legend
  • Deadset Legend
  • 436 posts
  • Twitch: Link
  • LocationAustralia

Posted 17 June 2020 - 07:00 AM

...but wait....there's more....

Appendix A: Divergence Control Measure

Each and every solution presented will cause the entire playerbase to continually spread out until T3 is less populated and pilots settle towards T1 and T5. This would mean new players, those who play less often and those who change in skill will take a long time to be seen accordingly by the MM after a few months of ANY of these systems. So....let's address that with an easy fix.

Thought of an idea to address the uncontrolled divergence in any in-scope PSR system. What about a regular "Re-scale" of all player PSR ratings on say, a 3 monthly basis to line up with Solaris Reset.

For easy numbers, let's say:
T5: 0-1000 PSR
T4: 1000-2000 PSR
T3: 2000-3000 PSR
T2: 3000-4000 PSR
T1: 4000-5000 PSR

OK so every 3 months scale around middle PSR (2500) the proportion they are away from it (40%).

Maths:

Chosen Variables:
PSRmid = 2500
Scaleback = 0.4

Rescale:
PSRnew = PSRold - Scaleback * (PSRold - PSRmid)
PSRnew = PSRold - 0.4 * (PSRold - 2500)

What does this do? This gives PGI the ability to run a very simple command through the database at the same time as Solaris Reset which will reign in the divergence and pre-seed everyone for the following 3 months. This ensures the reset day match quality is not random where in a full reset match quality would be poor for weeks.

Example:

Pilot "GGpro" is max T1 at PSR 5000. At end of season, calc becomes:
PSRnew = 5000 - 0.4 * (5000 - 2500)
PSRnew = 4000 or borderline T1/T2

Pilot "KillRocks" is bottom T5 at PSR 0. At end of season, calc becomes:
PSRnew = 0 - 0.4 * (0 - 2500)
PSRnew = 1000 or borderline T4/T5

Pilot "BerryMouse" is low T2 at PSR 3179. At end of season, calc becomes:
PSRnew = 3179 - 0.4 * (3179 - 2500)
PSRnew = 2907 or high T3

There you go. This effectively empties T1 and T5 and squeezes the playerbase back to centre in a fair manner within T2-T4 primed for T1 and T5 to refill. Strong pilots can be in T1 immediately separating them from T4 players and below. This should meet PGI's set purpose of PSR in separating new from experienced players.

#56 Thebackson

    Member

  • PipPip
  • The People's Hero
  • The People
  • 23 posts

Posted 17 June 2020 - 07:16 AM

Core1a is my vote. Best pure option for determining PSR.

As there is already a ms kicker of 25 points for winning , that is plenty of bias already built in.

Core1a

Edited by Thebackson, 17 June 2020 - 07:20 AM.


#57 Fenrir-1

    Member

  • PipPipPip
  • The Wolf
  • The Wolf
  • 82 posts

Posted 17 June 2020 - 07:19 AM

I'm leaning 1A. I'm against tying wins/losses heavily to PSR movement, especially when it comes to solo queue. In many matches I have played over this past year, I've found the teamwork element in game is a large variable. You can try communicating with some solo queue teams, and they will often ignore any input you have, or any warnings - such as that firing line they're about to walk into. And then you end up losing the match because you had a lance+ that just did their own thing, got killed, and the rest of us aren't in META/carry mechs.

I mean, sure. There are PUG/solo teams that come together and play as a team. When that happens, it's glorious. But with the current PSR system based on winning, and with a number of players who do not find incentive in coming together and winning - to continue with the same premise would be nonsensical. If it were 12 vs. 12/ 8 vs. 8 Group Play (remember those days?), yes. I would be in favor. But it is not, and there are too many different variables in Solo Queue as to how the team will interact (or not) with each other, and how it will contribute to a win or loss. If individual progress is being weighted in a solo play match, then the individual's actions and performance should be given more weight in determining PSR movement. In an environment where an evaluation is based on a team's win or loss, given that solo or quick is the only real option available to us and we cannot control who we play with, PSR movement results will be effectively as it is now - a crap shoot.

Edited by Gunfighter-1, 17 June 2020 - 08:39 AM.


#58 Sjorpha

    Member

  • PipPipPipPipPipPipPipPipPip
  • Philanthropist
  • Philanthropist
  • 4,477 posts
  • LocationSweden

Posted 17 June 2020 - 07:27 AM

I appreciate that you are trying to solve the divergence issue, but it seems to me that a psr formula that needs periodical adjustment of this kind has some fundamental issues.

There has to be a way to make a psr calculation where players stabilize at their relative performance within the player base.

The predictive accuracy of psr should increase over time, not decrease. It doesn't make any sense to have a predictive formula whose predictive power decays the more data you feed it, at that point you have to accept that the formula doesn't do the job and start over.

Edited by Sjorpha, 17 June 2020 - 07:31 AM.


#59 Nightbird

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • The God of Death
  • The God of Death
  • 7,518 posts

Posted 17 June 2020 - 07:47 AM

The predictive accuracy of these models increase over time. Modelling that show resets being beneficial are due to some false assumptions.

None of the suggestions will improve quality of matches (reduce stomps), so base all preferences on how you prefer to see PSR move.

All of the suggestions still reward farming Match Score. Even if you have 1 WLR, if you farm MS, then you will lose 40% on a loss and gain 60% on a win, allowing PSR to climb. If anything, all the players around 1WLR will be forced to farm MS like crazy because those who don't farm will gain 40% on a win and lose 60% on a loss...

Edited by Nightbird, 17 June 2020 - 07:57 AM.


#60 OneTeamPlayer

    Member

  • PipPipPipPipPipPip
  • Ace Of Spades
  • 399 posts

Posted 17 June 2020 - 07:49 AM

Vote: Core 2B

Also reiterating that either of the Core 1 choices that are designed around players directly "competing" against their own team and will highly incentivize negative in-game behavior.

If you think that the following are noted issues that come up often in our current setup:

- Players won't push
- Assaults won't tank armor
- People like to hide and use others locks for LRMs
- Lights won't scout
- Players stay in cover till armor is exposed to get KMDD and actual damage points
- Players are willing to core friendlies in an attempt to get a kill confirmation
- and dozen other issues

...anyone who has ever played a competitive game will tell you that if the system is based on individual versus all these activities will increase dozens of times over.

Ironically the shift in gameplay will lead towards pushing teamplay to the bottom of pilot consideration, heck if i'm just competing against everyone i don't even need to bother with putting on my headset for callouts as it's actively against my best interest to coordinate and ensure everyone on my team does well.

Put simply

Core 1 "Team" gameplay in 1 image

Posted Image


Edited by OneTeamPlayer, 17 June 2020 - 07:51 AM.






1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users