103 replies to this topic

[Skipmagnet]

You can also do almost 400 dmg, 5 kills, 3 assists and go down in psr on a loss. I can't even math this.

[Savage Wolf]

eyeballs, on 01 November 2017 - 12:21 PM, said:

You can also do almost 400 dmg, 5 kills, 3 assists and go down in psr on a loss. I can't even math this.

I have done less and not have my PSR change. It's not hard to get the required matchscore to not drop and you can optimize your playstyle to maximize matchscore rather than try to win. Just stay in the back, use long range, strip, not kill mechs.

[Gas Guzzler]

Savage Wolf, on 01 November 2017 - 12:20 PM, said:

If you focus on damage, chances are you will only ever increase your PSR, because then even if you lose, you don't fall. If you focus on winning then there will be more matches where you lose and didn't have the damage to offset the loss and even if you win, you might only have enough damage to rise a little.
In conclusion, the system is optimized for damage, not win/loss.

Well, I personally prefer destroying mechs as a primary objective because its the sure fire way to win. But like I said "optimized for damage" is a stretch and is largely a subjective statement. You can literally do almost zero damage and win, and not have your PSR decrease, and that is ********. So where you are drawing the line for whether or not it is optimized for damage or win/loss, doesn't seem like the right place.

I wonder if there is a positive correlation with high Average Match Score and High KDR... would that be shocking to you, that it might just be that people who focus on doing a **** ton of damage also happen to win more than people who just "play to win"?

Edited by Gas Guzzler, 01 November 2017 - 12:32 PM.

[Savage Wolf]

Gas Guzzler, on 01 November 2017 - 12:27 PM, said:

Well, I personally prefer destroying mechs as a primary objective because its the sure fire way to win. But like I said "optimized for damage" is a stretch and is largely a subjective statement. You can literally do almost zero damage and win, and not have your PSR decrease, and that is ********. So where you are drawing the line for whether or not it is optimized for damage or win/loss, doesn't seem like the right place.

Killing mechs inefficiently is rewarded higher than efficiency. If all you do is take the core of a mech, you are not rewarded much, but it is better for winning. And you also highlight another problem, it further enforces that the best way to win is through kills, making it even harder to make gamemodes play differently than skirmish.
And the fact that I can win a match and not rise in PSR just further shows how much the focus is on damage. Simply, if you must win, then it must be by doing massive damage. And it's not subjective which way fills you xp bar the fastest. One way is the fastest, but I do not have the required data to find the objective answer.

[Gas Guzzler]

Savage Wolf, on 01 November 2017 - 12:46 PM, said:

Killing mechs inefficiently is rewarded higher than efficiency. If all you do is take the core of a mech, you are not rewarded much, but it is better for winning. And you also highlight another problem, it further enforces that the best way to win is through kills, making it even harder to make gamemodes play differently than skirmish.

Well, yeah. Its PVP MechWarrior, a 'Mech slugfest should be what this game is about.

Do you really find shooting a quasi-stationary Atlas until it dies because the other team decided to not even try to defend it fun? Its spectacularly boring, I'd quit the game if all this was was doing BS "objectives" like that.

Savage Wolf, on 01 November 2017 - 12:46 PM, said:

And the fact that I can win a match and not rise in PSR just further shows how much the focus is on damage.

Not really, there are players that don't do anything to help win the match, and they should not get a rise in PSR for doing nothing.

[Savage Wolf]

Gas Guzzler, on 01 November 2017 - 12:50 PM, said:

Well, yeah. Its PVP MechWarrior, a 'Mech slugfest should be what this game is about.

Do you really find shooting a quasi-stationary Atlas until it dies because the other team decided to not even try to defend it fun? Its spectacularly boring, I'd quit the game if all this was was doing BS "objectives" like that.

I will admit that PGI are really really bad at making alternative game modes and making objectives interesting. But so many other games have done it so it is possible to have interesting objectives that doesn't distract from the combat, but rather mix with it.
The best that they have done in my opinion is Conquest in FW where is it becomes quite the tug of war and it matters where you fight, but you are fighting none the less.
So using other games in the genre, yes, objectives are fun and adds depth when done right. But even when PGI does something barely decent, PSR shows up and ruins it again.

Gas Guzzler, on 01 November 2017 - 12:50 PM, said:

Not really, there are players that don't do anything to help win the match, and they should not get a rise in PSR for doing nothing.

And instead we have people also do badly when it comes to winning, but if they just did lots of damage, they might even get rewarded for it. I've seen way too many assault racking up damage from the back, but not sharing armor or help push.

It's probably why Polar and Frozen are so popular. It's a trading map with lots of opportunity to rack up lots of damage without doing anything to win.

So in the end, I would rather carry people doing 0 damage than I want to be meatshields for people optimizing for matchscore and rewards and see them rise in PSR, while I risk falling.

[Vxheous]

Davegt27, on 31 October 2017 - 03:35 PM, said:

one of the devs said in a pod cast that when they tightened up the MM

that resulted in more stomps

OP PGIs computer controls are at the core of the problem

That's working as intended. Higher skilled players tend to punish mistakes far more frequently than lower skilled players. The moment someone on one side makes a mistake and gets killed, all of a sudden, it's 12vs11. Unless the 11 side can somehow immediately capitalize on a mistake that the 12 side makes, the numerical difference between the teams will eventually create a snowball effect. You see this a lot in comp matches, where an initial close game at 0-0 will suddenly snowball to being 8-1, 8-2.

[Gas Guzzler]

Savage Wolf, on 01 November 2017 - 01:05 PM, said:

And instead we have people also do badly when it comes to winning, but if they just did lots of damage, they might even get rewarded for it. I've seen way too many assault racking up damage from the back, but not sharing armor or help push.

It's probably why Polar and Frozen are so popular. It's a trading map with lots of opportunity to rack up lots of damage without doing anything to win.

So in the end, I would rather carry people doing 0 damage than I want to be meatshields for people optimizing for matchscore and rewards and see them rise in PSR, while I risk falling.

Well, I disagree with the notion that at an assault racking up damage isn't helping the team win. Sharing armor is a slippery goal, as so often I see people clamoring for others to share armor while they basically just walk straight into the enemy teams firing line. If I am in a Dire (will usually be a 5-600m alpha build), the way I help my team win is not by sharing armor, but by decimating any target that presents itself to me. That is the mech's strength. A Dire that aims to "share armor" folds pretty easily. Without even trying to "share armor" I will end up taking plenty of fire. Dire's make great targets. Any direct fire mech will end up taking
damage and "sharing armor" by default. Its really the LRM boats that you see hiding and not sharing armor, and that I will agree does contribute to a loss.

On the other hand, it is NEVER in the teams best interest to lumber out into the open in an Annihilator and then die with 120 damage and whine about others not sharing armor.

[Savage Wolf]

Gas Guzzler, on 01 November 2017 - 01:16 PM, said:

Well, I disagree with the notion that at an assault racking up damage isn't helping the team win. Sharing armor is a slippery goal, as so often I see people clamoring for others to share armor while they basically just walk straight into the enemy teams firing line. If I am in a Dire (will usually be a 5-600m alpha build), the way I help my team win is not by sharing armor, but by decimating any target that presents itself to me. That is the mech's strength. A Dire that aims to "share armor" folds pretty easily. Without even trying to "share armor" I will end up taking plenty of fire. Dire's make great targets. Any direct fire mech will end up taking
damage and "sharing armor" by default. Its really the LRM boats that you see hiding and not sharing armor, and that I will agree does contribute to a loss.

On the other hand, it is NEVER in the teams best interest to lumber out into the open in an Annihilator and then die with 120 damage and whine about others not sharing armor.

Currently it's more Clan ER-LL boats that often do this. Stay far behind the group and just blasts from a distance. Assaults don't have to be at the front to share armor so your example sounds fine. But then there are the times when a push is suggested, a good number of team mates acknowledge and then go for it. Meanwhile the rest of the team doesn't follow up and totally waste the momentum, but still pick up damage from a distance.

And I've tried picking up those builds as well, and it is really easy to rack up the damage, but it's spread all over the enemy team and all over their mechs and you are too far behind to support a push. But according to PSR, I did better than my teammates that actually did most of the kills and took the targets that mattered. Sometimes I just these builds out to reap the rewards and then go back to my fun mechs afterwards.

In the end, my most winning mechs and most effective mechs are not the most rewarded mechs in terms of matchscore or c-bills. And that is just wrong.

[Gas Guzzler]

Savage Wolf, on 01 November 2017 - 01:31 PM, said:

Currently it's more Clan ER-LL boats that often do this.

That's possible, but a good player in a 6 ER LL SNV-1 can literally pick a couple targets before the fight even gets going (provided it get's a favorable map like Frozen City). That can make winning a lot easier. You also have to remember that even if they don't get a kill, they could have completely cored the mech and then had some closer range build go in there to snag the killing blow.

It almost seems like you are talking about bad players that do Clan ER LL boats, but then they always get less damage than good players, so ... meh. IMO the PSR system just needs to be more punitive. Its way too easy to go up and not hard enough to go down. But... both the win/loss and the personal performance SHOULD factor into it.

Edited by Gas Guzzler, 01 November 2017 - 01:39 PM.

[Savage Wolf]

Gas Guzzler, on 01 November 2017 - 01:37 PM, said:

That's possible, but a good player in a 6 ER LL SNV-1 can literally pick a couple targets before the fight even gets going (provided it get's a favorable map like Frozen City). That can make winning a lot easier. You also have to remember that even if they don't get a kill, they could have completely cored the mech and then had some closer range build go in there to snag the killing blow.

It almost seems like you are talking about bad players that do Clan ER LL boats, but then they always get less damage than good players, so ... meh. IMO the PSR system just needs to be more punitive. Its way too easy to go up and not hard enough to go down. But... both the win/loss and the personal performance SHOULD factor into it.

There are other builds as well. Machine gun boats are also good at racking up matchscore because they destroy many components. And I'm not saying that you cannot do well with these builds, because you can, and I'm fine with that, but you can also just use them to farm the system and contribute little to victory. And it's also sort of wrong that if you do well in these builds you are rewarded more than in equally good builds. It just shows that PSR does not measure your contribution to victory.

And personally I'm fine with the concept of measuring a players contribution to victory. Problem is, there really is only one reliable way to do so and that's just simply win/loss. No matter what else you try to measure, it can just be farmed and optimized for without having to win more. So to me it's a fools errand and it just cultivates bad behavior. If winning was the only way to rise, you'd have to optimize for winning. And it would work equally well in any gamemode.

And luckily for you, since all gamemodes are most easily won by killing the enemy team, that would still be what they optimize for. Until the day they create a good objective mode by accident.

[Gas Guzzler]

Savage Wolf, on 01 November 2017 - 01:51 PM, said:

There are other builds as well. Machine gun boats are also good at racking up matchscore because they destroy many components. And I'm not saying that you cannot do well with these builds, because you can, and I'm fine with that, but you can also just use them to farm the system and contribute little to victory. And it's also sort of wrong that if you do well in these builds you are rewarded more than in equally good builds. It just shows that PSR does not measure your contribution to victory.

And personally I'm fine with the concept of measuring a players contribution to victory. Problem is, there really is only one reliable way to do so and that's just simply win/loss. No matter what else you try to measure, it can just be farmed and optimized for without having to win more. So to me it's a fools errand and it just cultivates bad behavior. If winning was the only way to rise, you'd have to optimize for winning. And it would work equally well in any gamemode.

And luckily for you, since all gamemodes are most easily won by killing the enemy team, that would still be what they optimize for. Until the day they create a good objective mode by accident.

Some loadouts are better at inflating numbers than others, but I think you will still see a positive correlation between KDR/Avg Match Score and WLR if you look at the data as a whole.

[Savage Wolf]

Gas Guzzler, on 01 November 2017 - 01:54 PM, said:

Some loadouts are better at inflating numbers than others, but I think you will still see a positive correlation between KDR/Avg Match Score and WLR if you look at the data as a whole.

But we are not talking about the average player, we are talking about the ones that the system fails to measure correctly. And if I had access to the data I'm positive I could find these players.

It seems more that we differ in who we find it unacceptable that the system might reward above their contribution. You don't want to reward the potato, I don't want to reward the farmer.

And another thing is that since the system rewards high damage numbers and so some players optimize for this, then your best chance of winning is actually to play along with them, making their behavior scew the numbers. They have no incentive to help you win more unless it's by doing it their way.

[Trissila]

Savage Wolf, on 01 November 2017 - 12:46 PM, said:

Killing mechs inefficiently is rewarded higher than efficiency. If all you do is take the core of a mech, you are not rewarded much, but it is better for winning. And you also highlight another problem, it further enforces that the best way to win is through kills, making it even harder to make gamemodes play differently than skirmish.
And the fact that I can win a match and not rise in PSR just further shows how much the focus is on damage. Simply, if you must win, then it must be by doing massive damage. And it's not subjective which way fills you xp bar the fastest. One way is the fastest, but I do not have the required data to find the objective answer.

If you win, and your PSR is stagnant for that match, that means you got less than 100 match score.

You did not contribute meaningfully to your team's victory in that case. You could maybe argue conquest, if you were doing nothing but capping and it came down to the wire on points (rather than the usual blowout where one team kills all of the other and wins by default because they're ahead on caps), or Domination if for some reason the enemy refused to get in the circle at all and you were the one holding it for your team.

Even in the objective-oriented modes, even in the ones that don't auto-win you for killing the entire enemy team, doing damage to the enemy is a key component of victory.

This is not to say that I believe kills/damage are the only things that should matter -- I wish objectives were rewarded more so that people would actually play them as more than an afterthought -- but sub-100 MS is indicative of an incredibly passive match where you didn't do much of anything but run around.

Edited by Trissila, 01 November 2017 - 02:33 PM.

[Gas Guzzler]

Savage Wolf, on 01 November 2017 - 02:10 PM, said:

But we are not talking about the average player, we are talking about the ones that the system fails to measure correctly. And if I had access to the data I'm positive I could find these players.

It seems more that we differ in who we find it unacceptable that the system might reward above their contribution. You don't want to reward the potato, I don't want to reward the farmer.

And another thing is that since the system rewards high damage numbers and so some players optimize for this, then your best chance of winning is actually to play along with them, making their behavior scew the numbers. They have no incentive to help you win more unless it's by doing it their way.

I think I just disagree that "farming" is an epidemic. Most players shoot to kill. I will take STs if it is on a mech with large STs who happen to be twisting their CT away, because I am more confident that I will get my entire shot on that component which is ultimately more debilitating than spreading damage around, but that's not because I am farming. I am just making a decision on what will make the enemy less dangerous faster.

Gaming PSR is pretty dumb, and if you don't win that means you might die, and the kind of person who would try to game PSR probably wants to have a good KDR, so that is self defeating. Basically... I don't think players intentionally trying to work the PSR system is a common thing.

[Angel of Annihilation]

Tarogato, on 01 November 2017 - 10:55 AM, said:

Maybe I'm not explaining myself clearly.

I would like to see a "Tier 0" that is just a label. We understand that there aren't even enough T1 players to build T1 matches, the matchmaker has to draw from also T2 and T3 just to build a match in a timely manner at all.

I just want a T0 that shows up on your forum badge, and your in-game progress bar. It doesn't necessarily have to be used for matchmaking, I just want something you can look at that says "this player is rated a lot higher than your typical T1 player", so that for instance Proton doesn't appear to be in the same tier as JoeSchmoTierOne who isn't even in the 50th percentile for average match score.

This is kind of an example of what I think is going on. Your typical Tier 1 player can range from someone who is just a good play and someone like Spriggin that is truly an Elite player that only few players can match up with. The problem is or at least the problem I think I am having is that I am sitting at the "Just Good enough to be Tier 1" level and am commonly fighting against people who are elite ranked. This is why I feel like I am getting spanked cause unfortunately I am. Look at it this way, lets say Tier 1 encompasses the top 20% of players. Then lets say I am maybe ranked in the 19% percentile, well I am still a good player. However if during the time I regularly play there is a inordinate amount of top 5% players are in the ques, then I am not going to be able to compete because there is a huge difference in the skill level of a 19th percentile player and a 5th percentile player. Tier 0 in your example would encompass maybe only those top 5% players.

Vxheous, on 01 November 2017 - 01:14 PM, said:

That's working as intended. Higher skilled players tend to punish mistakes far more frequently than lower skilled players. The moment someone on one side makes a mistake and gets killed, all of a sudden, it's 12vs11. Unless the 11 side can somehow immediately capitalize on a mistake that the 12 side makes, the numerical difference between the teams will eventually create a snowball effect. You see this a lot in comp matches, where an initial close game at 0-0 will suddenly snowball to being 8-1, 8-2.

Yeah I think this has alot to do with it as well. It only takes one mistake at a higher competitive level of gameplay for things to fall apart and it is almost impossible to recover from it.

[Shard Phoenix]

I feel it boils down to this...

From the 'Taterist of Taters to the Hardest of Tryhards, we all (should) want a balanced game play experience where we're pitted against opponents who best fit our abilities.

The current PSR system does not provide that in the way it should.

I feel that comp players don't really want to bore themselves by farming the potato patch any more than the patch wants to be farmed. A player "rating" system which can be advanced in how MWO's PSR can, does not provide such a thing. The joke of "Welcome to Tier 1" exists for a reason.

Personally, I don't want to end up in a PSR bracket where I don't belong. I admit that I suck whale crap and that's fine. Keep me away form the players who choose to dedicate the time to become gods. I'm not on their level and I don't want to be food for them. Put me amongst my kind and keep me there.

[SFC174]

Shard Phoenix, on 01 November 2017 - 08:03 PM, said:

Personally, I don't want to end up in a PSR bracket where I don't belong. I admit that I suck whale crap and that's fine. Keep me away form the players who choose to dedicate the time to become gods. I'm not on their level and I don't want to be food for them. Put me amongst my kind and keep me there.

Well, you're going to have to play with a few of them. We should be playing against people somewhat better than us, at least a few of them. This is true for several reasons: (1) You don't get better without facing better opponents (2) The player population is too small to keep the mechgods to themselves.

I've proposed it before, but I believe the key is to balance the talent of players on opposing teams. If we had a few more tier levels (determined by bell curve distribution of the entire player base and with zero sum movement between tiers), you could end up with 1 Tier0 (the best), 3 Tier1 and 8 Tier2 on each team in the match (for a top Tier game). You could theoretically have more than 1 Tier0 if enough were queued, but it has to be balanced. However, this only works if the Tier system actually represents player skill. Right now it assuredly does not (or if it does, the skill spread from top to bottom of Tier1 is waaayyyy too large).

[Shard Phoenix]

Don't read too far into what I said. I get that to get better you must fight better.
My point was that there needs to be balance.
However, there are some people who just want to f-around after work and don't give a rat's butt about being awesome at a video game. Those people don't necessarily want to be thrown into anything that resembles a shark tank.
I'm kind of in between. I like a challenge, but at the same time don't find any value in being farmed.

Edited by Shard Phoenix, 01 November 2017 - 08:27 PM.

[The Cyberserker]

Xmith, on 31 October 2017 - 10:02 PM, said:

You have to do better than this. This is not funny.

It wasn't supposed to be. It was a serious question.