Jump to content

Yet Another Psr Idea


4 replies to this topic

#1 SharDar

    Member

  • PipPip
  • Bridesmaid
  • 38 posts

Posted 30 July 2020 - 05:25 PM

I'm sorry. I'm sure you guys are getting tired of this, but we all want MWO to be the best it can be.

I've been reading and participating in the discussion about the PSR change, and it suddenly hit me: we should use a pilot's Average Match Score for the PSR. While the Match Score is factored into PSR, I think we can simplify the system to just use a pilot's average match score to directly establish his PSR.

The Match Score is already deciding how well you did in a match. The Average Match Score for all pilots forms a distribution, which can be used to establish tiers. You decide what percentage of pilots should fall into each tier. If you want 5% to be considered Tier 1, then the top 5% of pilots ranked by Average Match Score are in Tier 1.

Can you post a distribution of the average match scores for each of the mech classes? I'd be curious to see what that looks like.

This has several benefits:
  • It is simple.
  • It is based directly on the visible score you attain based on your contribution to a match.
  • You have a different skill rating for each mech class, which factors in the different capabilities of each mech class and your ability to pilot each one. You may be excellent in an assault mech but stink in a light mech.The skill rating used by the match maker would be based on the class of the mech you have chosen to pilot for each match.
There is the notion that you should be differently rewarded based on whether your team won or lost. This could be achieved by adding a win/loss factor. Winning team members could get 50 points added to their match score, for example. You could make it more complex so that members of the winning team who did not contribute get a lower bonus; but I don't think that would be necessary.



In this way, members of the losing team with a high match score are not penalized for being on the losing side. They just don't get a bonus. Or maybe they have some points deducted for losing.

Likewise, there is still the issue about having pilots from different tiers playing in the same match. This can be addressed by adjusting a pilot's match score based on the difference between their Skill Rating and the average skill rating. I pilot with a skill rating significantly higher than the average skill rating for a match would have their match score for that match reduced proportionately. Likewise, a player whose skill rating is significantly lower than the average in the match, would have their match score increased to compensate for playing against better players. While this doesn't completely even out the playing field, at least it rewards all pilots relative to the skill of the players they are fighting.

Sorry to keep bringing this topic up, but it's at the core of creating good matches and giving players a sense of pride in their accomplishments. Thanks for your all of your hard work!

Edited by SharDar, 30 July 2020 - 05:42 PM.


#2 martian

    Member

  • PipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 2,198 posts

Posted 30 July 2020 - 08:57 PM

View PostSharDar, on 30 July 2020 - 05:25 PM, said:

There is the notion that you should be differently rewarded based on whether your team won or lost. This could be achieved by adding a win/loss factor. Winning team members could get 50 points added to their match score, for example.

The victory bonus already exists. It is +20 points added to your Match Score if you are on the victorious side.

#3 SharDar

    Member

  • PipPip
  • Bridesmaid
  • 38 posts

Posted 31 July 2020 - 02:53 PM

Cool! Then your ranking by Average Match Score can be directly used as your Pilot Skill Rating. Why make it harder than that?

#4 Biomechtric

    Member

  • PipPipPip
  • Bad Company
  • 81 posts

Posted 01 August 2020 - 06:36 AM

I'll think about it... After we have given the new system a chance to settle in.

#5 SharDar

    Member

  • PipPip
  • Bridesmaid
  • 38 posts

Posted 02 August 2020 - 10:36 AM

I’ve ­­been thinking about my proposal a lot since the start of Season 49. That’s given me the chance to wipe away bad performances of the past and start over with what I’ve learned. I’m currently the 9th ranked Assault pilot by Average Match Score with a score of 449:
Rank Pilot Name Total Wins Total Losses W/L Ratio Total Kills Total Deaths K/D Ratio Games Played Average Match Score
9 SharDar 15 9 1.67 35 14 2.50 24 449

I’m also the 95nd ranked Light pilot by Average Match Score:
Rank Pilot Name Total Wins Total Losses W/L Ratio Total Kills Total Deaths K/D Ratio Games Played Average Match Score
95 SharDar 22 24 0.92 41 35 1.17 46 278

Yes, I know the season has just started so I will likely be moving down.

This illustrates what I’ve been saying about the relative difficulty of getting the same Match Score in a Light mech versus an Assault mech. When I drive a Light mech, I have to compete against Assault mechs for a favorable change to my PSR. A Match Score of 278 (my average as the 95th ranked Light pilot) will almost always cause my PSR to go down. I’m on the cusp of climbing out of tier 4 into tier 3. My PSR is based on mostly playing light mechs.

When I want to know how I compare to other pilots, I look at the leaderboard and sort by Average Match Score. The Match Score is an objective standard that computes your contribution to the match. While it certainly is biased towards damage and kills, most matches are decided by kills which are caused by damage. Yes, the Match Score calculation can be improved; but for now it is the standard by which we assess your contribution to the match, and it is used to calculate your PSR change.

Using a pilot’s Average Match Score is a better assessment of relative ability. It is a better predicter of the contribution a pilot will make to a match than the current PSR.

My Average Match Score for assault mechs is 170 points higher than for light mechs. I believe this reflects how much harder it is to achieve the same Match Score in a light than you could get in an assault. Looking at the leaderboard you see lower average match scores in the distribution as you move down the weight classes.

However, the Match Maker uses a single PSR value to cede me into matches. I think that explains why my win/loss ratio is so much higher for my Assault mechs than my Lights: 1.67 vs 0.92. I don’t make as big of a contribution when I drive my light mechs. Using a mech class-specific value to balance teams would avoid that.

Average Match Score is also a better way to balance teams than a combination of tonnage and PSR. People complain when there are more Assault mechs on the opposite team. People also say that a Light can perform as well as an Assault. A Light mech pilot with Average Match Score of 300 for light mechs is the equivalent of an Assault mech pilot with an Average Match Score of 300 for Assaults. They both can be expected to contribute 300 points worth to the match. All you need to do is ensure that the two teams have nearly the same totals for Average Match Score based on the mechs the pilots have selected.

People are concerned that a focus on Match Score gives people an incentive to ignore the game mode and focus on damage. I made that point about a month ago. I am no longer concerned by that. There is already a win bonus applied to your match score. Also, most matches are decided by kills (which are caused by damage), not by capping. If we want to incent people to pay more attention to the game mode, then we would do that by changing how Match Score is calculated—which is a completely separate topic.

Please let me know if any of this resonates with you. Thanks for listening. You guys rock! Thanks for working to make Mech Warrior Online the best game it can be!

Edited by SharDar, 02 August 2020 - 11:23 AM.






1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users