19 replies to this topic

[Jamun]

I was wondering why the PSR seems to be so dependant on a Win/Loss?

It would seem to be that you can do pretty poorly and get a PSR increase if your team wins and conversely you can do quite well, but get a decrease if your team loses.

It's not particularly encouraging, particularly since the Matchmaker consistently puts you in teams with/against what are demonstrably better players.

Any views/opinions?

[MrJeffers]

Good players win.

Determining what *exactly* a player did to help win is nearly impossible.

Was it because they said 'enemy all in D4' that caused their teammates to shift and focus them down?

Was it because they back stabbed the enemy assaults that got left behind because of NASCAR?

Was it because they soaked 200 damage while their teammates killed the enemy?

Was it because they played the squirrel and drew the enemy forces out for easy kills?

The game can't tell.

[Jamun]

I would position it that good teams win, not individual players. A good player with an under performing team is punished? That doesn't make much sense.

Absolutely the game can't take all tactics into account (acting as a shield as you point out isn't captured).

But the rest of your comments I thought were caught by the game, and are currently rewarded with C-bills?

By the way before someone piles in with any 'Git Gud', L2P or any other ad hominem I don't think I'm a good player, but having the PSR reduced for matches I lose, when I do well (and conversely being rewarded for wins I haven't contributed much towards) strikes me as odd.

[- World Eater -]

I try to do at least 500 dmg per game (win or lose) to guarantee a gain.

[El Bandito]

Jamun, on 29 April 2017 - 07:37 AM, said:

I would position it that good teams win, not individual players. A good player with an under performing team is punished? That doesn't make much sense.

Absolutely the game can't take all tactics into account (acting as a shield as you point out isn't captured).

But the rest of your comments I thought were caught by the game, and are currently rewarded with C-bills?

By the way before someone piles in with any 'Git Gud', L2P or any other ad hominem I don't think I'm a good player, but having the PSR reduced for matches I lose, when I do well (and conversely being rewarded for wins I haven't contributed much towards) strikes me as odd.

A good player will still have more wins in the long run--at least in Solo-Q. As for complaining about PSR loss when you did well in a losing match--your PSR will not drop if you actually have done well. And it is not too hard to do to ensure that.

MWO's PSR is way more lenient than other games' ladder system, cause your contribution softens the blow by a lot. Compare that to Overwatch, or LoL, where if your team loses, even with a godlike score, you will still lose tons of points.

Edited by El Bandito, 29 April 2017 - 10:20 AM.

[Bulletsponge0]

El Bandito, on 29 April 2017 - 10:19 AM, said:

A good player will still have more wins in the long run--at least in Solo-Q. As for complaining about PSR loss when you did well in a losing match--your PSR will not drop if you actually have done well. And it is not too hard to do to ensure that.

MWO's PSR is way more lenient than other games' ladder system, cause your contribution softens the blow by a lot. Compare that to Overwatch, or LoL, where if your team loses, even with a godlike score, you will still lose tons of points.

I've never played Overwatch or LoL, but do they reward players that play awfully but get carried by the team like MWO does?

[El Bandito]

Bulletsponge0, on 29 April 2017 - 11:06 AM, said:

I've never played Overwatch or LoL, but do they reward players that play awfully but get carried by the team like MWO does?

Even more so. In MWO, if you play awfully, your PSR goes up only a little, and if your match score is below 101 points, your PSR will not go up even if your team wins. Compared to that, both Overwatch, and LoL (and other popular games such as Dota2), will increase your rating by sizeable amounts even if you did nothing all game.

[DragoWolf]

On the topic of the PSR system, I have played since closed beta and I used to own a computer that was technically below the specs for running the game but it still worked. I played until I went to college about a year ago at which point I got a new computer that can run MWO quite well. This is all just background, but those years before psr was implemented I had a very low "rank".

I would like to clarify that I understand that the win/loss mechanic of weighting is intended to focus team coordination. This being said, I believe PGI needs to examine and rework their values for the match score. Despite the system's beauty on paper, even dropping with a team where you push for teamwork and play your best to be a team, there is no eye-catching hook to convince your team to do such. Low tiers are riddled with players who aren't "BAD", they are just misinformed on the game mechanics and see no proof of play-style improvement.

The PSR and rewards system should be altered to use in-match "hints" or something sorta like in the Battlefield series. Orders are verbally and visually confirmed to be completed via the game's announcer(not for the visual confirmation). As of now, there are rewards for team-oriented actions, but they are often overlooked being at the bottom of the screen and do not draw attention. There needs to be something like a tactical command screen in the player's cockpit that updates with orders, shows the distance to the region that has the command, or something.


I am still only in tier 4 now and it seems nearly impossible to climb tiers despite the overwhelming "assurance" by PGI that the system is designed for a tier "creep" to get everyone eventually in tier 1. It seems that the PSR is way too heavily win/loss dependent where I have literally felt like I have lost the game, but my team won, so I went up in ranking. I have also experienced situations where I was almost at the top of my team in match points, yet we lost, so I dropped in rank.

Making the team-oriented actions part of the immersion will not only make lower tier players aware of their actions and incentives to play in a team-oriented manner, but it will give a more visible interface for orders besides just the command-map that exists. It will also give visual affirmation for low tier players that following orders and team-play is better than being a "PUG". More people will use the command roles and the current split between player commands and rewards will prevent farming of the system. Players will be rewarded for their teamwork and WILL NOTICE IT. The player commands will be suggestions with noticeable weight due to the reward system being separate, but still giving confirmation that i.e. : sending alpha lance to scout your team's flanks isn't a waste of "damage dealers".

[Bishop Steiner]

MrJeffers, on 29 April 2017 - 07:15 AM, said:

Good players win.

Determining what *exactly* a player did to help win is nearly impossible.

Was it because they said 'enemy all in D4' that caused their teammates to shift and focus them down?

Was it because they back stabbed the enemy assaults that got left behind because of NASCAR?

Was it because they soaked 200 damage while their teammates killed the enemy?

Was it because they played the squirrel and drew the enemy forces out for easy kills?

The game can't tell.

Good players, actually don't necessarily win. They will likely win more. But even a tier 1 MetaLord can't guarantee a win pulling 11 potatoes. In fact basically, if the other team has jsut 2-3 decent players, said metalord will almost certainly still lose. (especially if said metalord is using his team as a meatshield to pad his stats to get above the PSR drop threshold, which is often the case)

Far too much weight is placed on W/L. Especially since many mediocre players get pulled along to wins with subpar performances simply by being on a team. W/L should be a part of it, but we should likely have separate PSR tiers for Group Play and Solo.

Honestly, even damage gets way too much weight. Hit% (and not you got 1 tick of damage per laser hit % crap, but you fired X laser this many times and score Y damage, meaning your REAL accuracy was.....) and Component Hit % accuracy can and should be huge factors. Kills Most Damage Dealt and Solo Kills. Those tell far more than Win and Loss. 4oo dmg and 4 Solo Kills (doable, especially for a backstabber) and you still go down in PSR on a loss, I believe?

And far too much is placed on pure combat.

PSR (and XP and Cbills) should have modifiers by Class/Role. An Assault should get a bonus for damage dealt and tanked, and possible a negative multiplier for not hitting certain thresholds of damage. A Raven, on the other hand should get huge bonuses for Spotting, UAV assists and use, TAG dmg and Kills, NARC, and other Scout/Spotter related activities that take considerable skill to do well, and CAN impact team performance, and yet get ignored in all facets of this games rewards and ranking system.

Basically this whole aspect of the game is long overdue for a massive overhaul.

Edited by Bishop Steiner, 29 April 2017 - 11:41 AM.

[MrJeffers]

Bishop Steiner, on 29 April 2017 - 11:40 AM, said:

Good players, actually don't necessarily win. They will likely win more. But even a tier 1 MetaLord can't guarantee a win pulling 11 potatoes. In fact basically, if the other team has jsut 2-3 decent players, said metalord will almost certainly still lose. (especially if said metalord is using his team as a meatshield to pad his stats to get above the PSR drop threshold, which is often the case)

Take your favorite sports athlete at the peak of their career down to a local field/arena/whatever. Have them play 100 pick up games of their sport with random people there to play that game.

You're fooling yourself if you think they are not going to win a lot more games than they lose.

[Bishop Steiner]

MrJeffers, on 29 April 2017 - 11:47 AM, said:

Take your favorite sports athlete at the peak of their career down to a local field/arena/whatever. Have them play 100 pick up games of their sport with random people there to play that game.

You're fooling yourself if you think they are not going to win a lot more games than they lose.

You're also fooling yourself if you think this is an accurate and valid comparison

A closer, though still flawed comparison is to use professional sports. Michael Jordan and LeBron James are (were) transcendent players. Without Other talented help, they couldn't win championships.

A lone good player will (as I already noted) still win more often than a lone average to bad player. But not to a degree high enough to actually determine "Personal Skill Rank", as there is far more than just one person's contribution that determines a win or a loss in this game.

[MrJeffers]

Bishop Steiner, on 29 April 2017 - 11:57 AM, said:

You're also fooling yourself if you think this is an accurate and valid comparison

A closer, though still flawed comparison is to use professional sports. Michael Jordan and LeBron James are (were) transcendent players. Without Other talented help, they couldn't win championships.

A lone good player will (as I already noted) still win more often than a lone average to bad player. But not to a degree high enough to actually determine "Personal Skill Rank", as there is far more than just one person's contribution that determines a win or a loss in this game.

I never said that PSR was good, or a representation of skill. I said winning is, above anything else.

[Bishop Steiner]

MrJeffers, on 29 April 2017 - 12:00 PM, said:

I never said that PSR was good, or a representation of skill. I said winning is, above anything else.

And I pretty thoroughly debunked that in my post, though You ignored most of it. Since, again, winning is too easily affected by things outside your individual control, so over reliance on W/L is meaningless (I know quite a few people who play almost entirely team only, with good W/L despite this, despite being, at best, average players, on a good day).

Also, the OP, and thus, this discussion, is about PSR.

[Nameless King]

MrJeffers, on 29 April 2017 - 12:00 PM, said:

I never said that PSR was good, or a representation of skill. I said winning is, above anything else.

Winning is not everything

[MrJeffers]

Bishop Steiner, on 29 April 2017 - 12:05 PM, said:

And I pretty thoroughly debunked that in my post, though You ignored most of it. Since, again, winning is too easily affected by things outside your individual control, so over reliance on W/L is meaningless (I know quite a few people who play almost entirely team only, with good W/L despite this, despite being, at best, average players, on a good day).

Also, the OP, and thus, this discussion, is about PSR.

You didn't debunk anything - show me all the game skill rating systems that don't use Wins as a key criteria. You can't "game" wins (outside of cheating/fixing) you can game most/all of the other metrics in this game.

Any skill system that doesn't use wins as a key indicator is a failed and inaccurate system. Which is why every skill tracking system used today has it as a key performance indicator.

King Alen, on 29 April 2017 - 12:05 PM, said:

Winning is not everything

And I never said that either.

Good players win.

That simple. The best indication of skill is winning.

[Bishop Steiner]

MrJeffers, on 29 April 2017 - 12:08 PM, said:

You didn't debunk anything - show me all the game skill rating systems that don't use Wins as a key criteria. You can't "game" wins (outside of cheating/fixing) you can game most/all of the other metrics in this game.

Any skill system that doesn't use wins as a key indicator is a failed and inaccurate system. Which is why every skill tracking system used today has it as a key performance indicator.

actually, I just showed how you could, by simply being carried by better players on a team.

So since I just debunked that comment without even having to try?

Again, your real accuracy, KMDD, Solo Kills, and actual fulfillment of Roles, tell more about actual skill than wins. Because none of those are gameable outside of hax. Winning, by contrast is. As it is also far more affected by simple random vagaries like the MM, than any of those other aspects.

MrJeffers, on 29 April 2017 - 12:09 PM, said:

And I never said that either.

Good players win.

That simple. The best indication of skill is winning.

In a solo play format this is true. 12v12, it's much more diluted, and much less accurate an indicator. When I was part of HARDCorp still, a NON comp team, my W/L was far higher than it is now, at over 2 to 1. As a solo player of average skill to on a good day, high average? I believe I am just slightly above 1 to 1, currently. My skill has not changed. But my W/L has.

But keep preaching bumper sticker slogans as fact if it makes you feel superior.

[MrJeffers]

Bishop Steiner, on 29 April 2017 - 12:14 PM, said:

actually, I just showed how you could, by simply being carried by better players on a team.

That doesn't debunk anything. Yes, a player can be carried. For a single match. Or for a few matches. They cant *always* be carried. Their skill deficit is going to show up as they are matched against better and better opponents.

Bishop Steiner, on 29 April 2017 - 12:14 PM, said:

But keep preaching bumper sticker slogans as fact if it makes you feel superior.

And you keep on feeling superior and resorting to witless quips.

Edit: And ignoring all the questions posed to you that you can't answer or back up.

Show me all the skill rating systems in games that don't use win. Link them for me, enlighten me and get me away from my "bumper sticker slogan" because obviously its wrong. Prove it.

Edited by MrJeffers, 29 April 2017 - 12:27 PM.

[Nameless King]

MrJeffers, on 29 April 2017 - 12:09 PM, said:

And I never said that either.

Good players win.

That simple. The best indication of skill is winning.

Bad players get dragged along in those wins all the time so again winning is not everything.

Edited by King Alen, 29 April 2017 - 12:45 PM.

[James Argent]

Saying 'good players win' is not the same as saying 'wins indicate good players.' The fact that people often get carried to a win doesn't change the fact that good players will tend to more often win than lose. No, 'winning isn't everything,' but it IS a thing that can't simply be ignored.

[El Bandito]

King Alen, on 29 April 2017 - 12:43 PM, said:

Bad players get dragged along in those wins all the time so again winning is not everything.

Not unless they drop with a good team all the time in GQ and CW. And even then, good players in that team have higher WLR than the baddies in that team. Xavier for example, has higher WLR than me cause he is better than me, even though we are both MS and we drop a lot together.

The majority of MWO players play Solo-Q, and good players will always have higher WLR than baddies in the long run there.

Edited by El Bandito, 29 April 2017 - 06:25 PM.