55 replies to this topic

[PhoenixFire55]

Karl Streiger, on 29 September 2016 - 02:02 AM, said:

Your so-called individual skill is an illusion.

LOL

Is there really any point in reading beyond this?

[Karl Streiger]

PhoenixFire55, on 30 September 2016 - 12:34 AM, said:

LOL

Is there really any point in reading beyond this?

No, not when you think I'm wrong. Of course you might convince me that I'm wrong.
Tell me what is this thing - skill in MWO.

[DovisKhan]

Karl Streiger, on 30 September 2016 - 01:12 AM, said:

No, not when you think I'm wrong. Of course you might convince me that I'm wrong.
Tell me what is this thing - skill in MWO.

Dude, you explained that yourself in your own post and now you're contradicting yourself.

"[color=#959595]so when you play MWO in solo QP and in 2 out of 3 games you think you did carry the team with >1000dmg. This doesn't happen because you are so good - it happens because your team is bad as well as the other team.[/color]"


You mentioned that a good player never carries the game, it's just that other are so bad...

Which is an oxymoronic statement, because that's the point - an experienced player is worth several potatoes, so if there is 23 potatoes and one good player, he will have a considerably higher than average potato win rate


If you can't fathom that you're contradicting yourself, there is little point in a discussion with you.


Individual skill matter a huge deal, mech choice matter a lot, team work is of course essential


But that's part of skill, a skilled pilot will be able to coordinate with a team and put his mech to good use, a potato won't

[PhoenixFire55]

Karl Streiger, on 30 September 2016 - 01:12 AM, said:

No, not when you think I'm wrong. Of course you might convince me that I'm wrong.
Tell me what is this thing - skill in MWO.

Same as everywhere ... being able to do smth better than others. In this case - being able to play MWO better, and if we are talking about "individual skill", then its being able to play MWO better while in solo mode.

Still, if you think that individual skill is an illusion you'll have no problem duelling someone like me for example and proving *me* wrong by winning.

So?

[mogs01gt]

The illusion of PSR:
3 kills, 500+dmg in a loss and my PSR stayed the same...WTF do they want from us in a loss!!

[Ghogiel]

mogs01gt, on 30 September 2016 - 06:17 AM, said:

The illusion of PSR:
3 kills, 500+dmg in a loss and my PSR stayed the same...WTF do they want from us in a loss!!

I think they want a 300 match score or something. Which is 600dmg. iirc it's just 200 match score/400dmg to not lose PSR on a loss.

[Jman5]

What really confuses me about the PSR is that there are players who have a negative KDR, negative WLR, an average matchscore below 250 and yet they are Tier 1. Having an average matchscore above 250 would mean you would inexorably climb tiers, because your losses aren't usually dragging you down. However, to be below that with a negative WLR should mean that your PSR should be trending downward. Your wins would bump you slightly, but your losses would bump you down doubly. If you're losing more than you win than shouldn't that mean you're trending downward? At the very least it shouldn't be moving.

It makes me wonder things.

Edited by Jman5, 30 September 2016 - 11:24 AM.

[C E Dwyer]

Jman5, on 30 September 2016 - 11:22 AM, said:

What really confuses me about the PSR is that there are players who have a negative KDR, negative WLR, an average matchscore below 250 and yet they are Tier 1. Having an average matchscore above 250 would mean you would inexorably climb tiers, because your losses aren't usually dragging you down. However, to be below that with a negative WLR should mean that your PSR should be trending downward. Your wins would bump you slightly, but your losses would bump you down doubly. If you're losing more than you win than shouldn't that mean you're trending downward? At the very least it shouldn't be moving.

It makes me wonder things.

The math is wrong that people use then, because people clearly are in that position.

Also if you score big on matches you lose you don't drop, and unless you get under 100 damage in a win game, you'll always go up

[Ghogiel]

Jman5, on 30 September 2016 - 11:22 AM, said:

What really confuses me about the PSR is that there are players who have a negative KDR, negative WLR, an average matchscore below 250 and yet they are Tier 1. Having an average matchscore above 250 would mean you would inexorably climb tiers, because your losses aren't usually dragging you down. However, to be below that with a negative WLR should mean that your PSR should be trending downward. Your wins would bump you slightly, but your losses would bump you down doubly. If you're losing more than you win than shouldn't that mean you're trending downward? At the very least it shouldn't be moving.

It makes me wonder things.

It's because you lose less PSR on losses than you gain on wins. Which is indicated in the crappy graphic chart posted about PSR. Since it's not isolated cases of this, it's the norm so basically flushes out statistical abnormality down the toilet and suggests that's working as intended.

It's probably something like you have to scrub out with sub 100dmg on the losses more than once to counter a single over 100 dmg win. Perhaps if you win and do 600dmg you need at two sub 100 losses to counteract that and break even or 3 losses over 100dmg. Which makes it very possible for something around a 0.6 W/L to remain steady/slowly increase PSR.

Edited by Ghogiel, 30 September 2016 - 02:42 PM.

[Scout Derek]

PhoenixFire55, on 30 September 2016 - 01:51 AM, said:

Same as everywhere ... being able to do smth better than others. In this case - being able to play MWO better, and if we are talking about "individual skill", then its being able to play MWO better while in solo mode.

Still, if you think that individual skill is an illusion you'll have no problem duelling someone like me for example and proving *me* wrong by winning.

So?

(in my mind how'd this little argument would play if Karl had responded earlier)

*Karl Refuses*

Phoenix begins to go Malthus on him.



You can't help but laugh at what you're arguing about.

Gotta have humor or else you'll go crazy.

[PhoenixFire55]

Scout Derek, on 30 September 2016 - 02:46 PM, said:

(in my mind how'd this little argument would play if Karl had responded earlier)

*Karl Refuses*

Phoenix begins to go Malthus on him.

You can't help but laugh at what you're arguing about.

Gotta have humor or else you'll go crazy.

Not much of a Malthus in here. "Those" guys are always the same, all talk with nothing to actually back it up.
I agree on the humor part tho, can only laugh at a guy who seriously says smth like "personal skill is an illusion".

[M T]

People basically get a few things wrong, to name a few:

- Higher damage doesnt mean better. 1000 damage 8 kills is better than 2000 damage 4 kills, or 3, or even worse. The less damage (required) and more solo kills, the better you aim

- How can you conclude the PSR is not broke when, especially on the high end tier, you clearly see 8 players being teamed up consistently, against total pugs, and with expected 12-0 3 minute games? We're talking competitive players here.

Yes, it is to a large extend just an experience bar, and highly likely that everyone playing on a consistent basis will eventually reach Tier 1.

The PSR, or any statistic in this game, that the PSR or any subsystem in MWO uses can not conclude or analyze which players do well or not, with the exception for weapon accuracy assuming that mechanic even works.

PSR was a fail to begin with, I think they should just base everything on proper accuracy statistics as that seems to be the only somewhat measurable skill, how much of your weapons damage is aimed towards a single component for instance, or do you spread it around like a newb.

Lets get rid of PSR and do it all over again, PGI should ask themselves again how would they go about measuring a players skill? Because thats the only thing the PSR's function should be. Make equal and balanced matches with equally skilled players, where possible.

Edited by M T, 01 October 2016 - 07:30 PM.

[El Bandito]

Jman5, on 30 September 2016 - 11:22 AM, said:

What really confuses me about the PSR is that there are players who have a negative KDR, negative WLR, an average matchscore below 250 and yet they are Tier 1. Having an average matchscore above 250 would mean you would inexorably climb tiers, because your losses aren't usually dragging you down. However, to be below that with a negative WLR should mean that your PSR should be trending downward. Your wins would bump you slightly, but your losses would bump you down doubly. If you're losing more than you win than shouldn't that mean you're trending downward? At the very least it shouldn't be moving.

It makes me wonder things.

Are you talking about seasonal Leaderboard stats, or overall stats? Leaderboard stat tracks only a month worth of matches.

[Jman5]

El Bandito, on 01 October 2016 - 07:30 PM, said:

Are you talking about seasonal Leaderboard stats, or overall stats? Leaderboard stat tracks only a month worth of matches.

The seasonal leaderboards that are archived. You can go back and check seasons 1-4, which gives you over 3 months of recent stats.

In the solo queue, I think PSR works alright up to tier 1. That's where it's been fraying at the edges and it's only getting worse with time. We need someway for Matchmaker to differentiate between a guy like Proton and someone who played 1000+ games to slowly inch his way From tier 2 to 1.

Edited by Jman5, 01 October 2016 - 08:47 PM.

[Scout Derek]

Jman5, on 01 October 2016 - 08:46 PM, said:

The seasonal leaderboards that are archived. You can go back and check seasons 1-4, which gives you over 3 months of recent stats.

In the solo queue, I think PSR works alright up to tier 1. That's where it's been fraying at the edges and it's only getting worse with time. We need someway for Matchmaker to differentiate between a guy like Proton and someone who played 1000+ games to slowly inch his way From tier 2 to 1.

This I can agree with. Jman, you and I have been in many different matches with many different teams. Suffice to say we've both witnessed separately and together the unbalance of team skill can be.

[El Bandito]

Jman5, on 01 October 2016 - 08:46 PM, said:

The seasonal leaderboards that are archived. You can go back and check seasons 1-4, which gives you over 3 months of recent stats.

In the solo queue, I think PSR works alright up to tier 1. That's where it's been fraying at the edges and it's only getting worse with time. We need someway for Matchmaker to differentiate between a guy like Proton and someone who played 1000+ games to slowly inch his way From tier 2 to 1.

PSR can do a better job than the previous Elo system, as long as PGI increases penalty on matches lost, and at least halves the current effect damage dealt score has on the match score (meaning damage only counts as 1/4 of the value on match score). Then PGI can do a soft reset on everyone's PSR (let's say lower everyone's PSR by two ranks) and allow the good players to climb up, while keeping the mediocre ones down below.