10 replies to this topic

[MysticLink]

So let's say you get 990 damage, 6 kills and a win. As opposed to 500 damage, 1 kill, and a win.

Does your ranking go higher with more damage, or is the more damage only useful if you lose in that it keeps you going higher despite losing?

I checked FAQ and couldn't find exactly how this ranking system works.

[Ghogiel]

PSR goes up more the better your match scores are for sure. a 1500dmg 6kill game might not be much better than a 1200 dmg 3 kill game if at all> Likely they have a maxium cap of PSR gain per match. But doubt anyone knows what match score you'd need to hit it.

But in your example between 1kdmg and 500dmg, there is definitely a difference.

Edited by Ghogiel, 26 January 2016 - 10:47 AM.

[XtremWarrior]

https://mwomercs.com...-tiers-and-psr/

This is what you're looking for.

[MysticLink]

Thank you.

[Sandpit]

The short answer?

Based on your individual match score and performance, you are given one of three adjustments to PSR after your match
- go down in PSR
+ go up in PSR
= no change in PSR

Usually I've found that on a loss, if I do about 350+ damage, I'll wind up with an =, anything less usually results in a -, while getting into the 500 damage range and a kill or two will often result in a + even for a loss.


Personally I think one change needs to be made to PSR. It needs to be based on a bell curve based on your entire team's performance. If you're on a team where 6 players fail to break 100 damage, you get 350 damage and a kill, then you did the best you could with the team you had to work with.

This would give you a closer PSR to true individual skill level. Until the MM system uses a system that takes things like this into account you're never going to get a true skill level for individual players.

Yes, this is a team based game, but when teams are built by individual players dropping solo and using random team generation mechanics, you cannot judge an individual player on the overall team performance.

The exception to that is the group queue in quickplay. If you enter group queue, then yes, you should have a more team based average PSR rating to use for MM purposes as that would help offset the "skill handicap" placed on teams.

A better system would be have 2 separate ranking systems. One for solo dropping and one for group drops. Give players a "group PSR" based on how they do in the group queue and na "solo PSR" based on how they do in solo drops.

[MoonUnitBeta]

Yep.

Basically the higher the match score, the more likely you will be able to go up. Match Score isn't all about damage, or kills, but they do carry their weight in the equation.
Winning or losing is the biggest factor that decides how much you go up in PSR. If you get 350 match score and lose, you will be going down in PSR. If you get 350 match score and win, you'll be going up in PSR.

[Dread Render]

Win / Loss should not be a factor... it is to dependent on luck.
Seems to me you have a 50/50 chance to win any game.
It should be based on your Score and your score alone.

[Sandpit]

Render, on 26 January 2016 - 11:01 AM, said:

Win / Loss should not be a factor... it is to dependent on luck.
Seems to me you have a 50/50 chance to win any game.
It should be based on your Score and your score alone.

The problem with that is that it doesn't take in the very real factor of having to carry a team sometimes. Just like the example above, if I only manage 250 damage and 2 assists, but that's the highest match score on my team, wouldn't it make more sense to reward that individual player?
Especially if 4-6 players just derped out and died early on.

A sliding scale takes into account more of the individual's impact on the team performance overall.

[Dread Render]

Sandpit, on 26 January 2016 - 11:18 AM, said:

The problem with that is that it doesn't take in the very real factor of having to carry a team sometimes. Just like the example above, if I only manage 250 damage and 2 assists, but that's the highest match score on my team, wouldn't it make more sense to reward that individual player?
Especially if 4-6 players just derped out and died early on.

A sliding scale takes into account more of the individual's impact on the team performance overall.

Yes, Even Better

But Take a look at my Sig though... My team lost but was that my fault? No...
Win or loss should count for much less.

p.s. that was a screen shot from the Old score system, not as it is now.

Edited by Render, 26 January 2016 - 11:27 AM.

[Sandpit]

Render, on 26 January 2016 - 11:24 AM, said:

Yes, Even Better

But Take a look at my Sig though... My team lost but was that my fault? No...
Win or loss should count for much less.

p.s. that was a screen shot from the Old score system, not as it is now.

I think W/L should still play a factor into the formula. It's still part of the overall factor of your individual skill levels and a metric that can be used. It just shouldn't be the only metric used. One of the issues with MM and all of the ranking systems they've used thus far are purely based on individual stats and don't take into account the factors I mentioned above.

[Dingo Battler]

If W/L didn't factor into PSR, there'll be no point to conquest and assault. Also, the objective is to win, not to deal damage. The team that has the initiative to push into base-cap should be rewarded, and not just be ignored. Also, without that, it destroys role warfare, as one vital role of a distraction-light is to cap.

Either way, WLR should be around 1.0, slightly higher if you're good. So if you're dealing damage, half the time you'll be getting "=", and the other half, increases, which should lead to a high PSR.

In the case of wipes, no matter how much people blame for "crap" teams, all the blame should fall equally on everyone. Its like assault players complaining they get NASCAR-ed to death. Their team is crap, true, but you can turn around and deal at least a couple of hundred damage before you die to soften the enemy, pray for a win, or in case of a loss, get a "=" rating.