75 replies to this topic

[Cnaiur]

I like the idea of tweaking the MM to take average match score into account, because as somebody pointed out earlier, there are vast disparities in skill between the top 2% or so and the rest of Tier 1. According to the Jarl's List, I'm a top 10% player which seems preposterous to me; my only skill that's even vaguely above average is aiming. And on those occasions I get a top player in my matches and get to spectate them, they're visibly much, much better than me in every way. And yet we're both Tier 1 players and worth the same for match making purposes. I can understand and accept that I need to be grouped with those guys to create adequate numbers for the match maker, but the skill disparity is real and should be factored in.

This isn't even addressing the people that have reached T1 and are below average players in almost every way.

[C E Dwyer]

*shrugs* Systems broken but doesn't really matter.

If all the high T3 and T2's get dropped out of T1

The T1's will complain about wait times, and the gates will get opened more quickly, letting the people that dropped back in.

The active population is about half the size it needs to be for any system to work, and not need a book waiting for games to form.

[Racerxintegra2k]

Sounds like we need a another tier above 1 for the truly great players. I suggest the Epeen Tier for the top 2% of the game. 1 Epeen tier player is worth at least 5 potatoes.

[Cnaiur]

For matchmaking purposes there are going to have to be some mediocre players mixing it up with top 2% guys because low population niche game, but perhaps it could be mitigated by the match maker not treating all Tier 1 players as fungible widgets.

[Leone]

Dee Eight, on 11 January 2018 - 11:54 AM, said:

Random conspiracy question... did the forum get the search feature broken at about the same time that new 3rd party ranking app was being written/implemented ?

Nope! was already borked by then. I just use google.

~Leone.

[Tahawus]

Tier is a useful, but flawed construct. (Not a surprise).

I've suggested several more complex ways of handling it, but here's a simple one...
Tiers should be assigned dynamically based on the player's elo in relation to the total population's Eli distribution. I.e. The top 20% of pilots are tier 1, and so on.

[DaFrog]

lol at that leaderboard, My t2 alt account is higher than my t1 main account.
That says a lot about leaderboards.

[Asym]

Davegt27, on 10 January 2018 - 03:37 PM, said:

I was looking at Jarl's list

https://leaderboard.isengrim.org/

and my overall rank is 26259

26259! I don't think you can get much lower then that
it also means if you see me on your team your chances of winning just
went down I mean come on why am I even in tier 1 it makes no sense

its obvious the current PSR system is not working
also PGI will never listen to me no matter how much money I spend
since my low scores mean in there eyes it qualifies me as low like scum

can any of you try hards talk some sense into PGI
the PSR is not working
GLHF

OP, first, you need to have a reason for a PSR.... FP is dead. QP is there but isn't a priority; and, with all of the alternate accounts, you can't keep T1's out of QP anyway so they aren't really messing it up.... Solaris doesn't need a PSR since the entire concept is based on chassis weight class and match winnings (a specific measure versus a general measure of capability).... PSR's aren't about your skill (capabilities), PSR's are about what you have done (your history.)

Being in any tier isn't hurting the game. What is hurting the game is that the game doesn't know what it wants to be and is not suceeding at anything since they clearly lack a strategic vision that makes any sense as it drives towards a specific goal or customer sarisfaction.... They (PGI) really doesn't know what it wants to be so, how can we?

[Fleeb the Mad]

Guys, you need to have some more realistic views of the actual size of the various player pools.

Tier 1 isn't for the top 1% competitive folks and it never was. Those people are in it, but creating a matchmaking bucket for such a small segment of the player base by itself is pointless. People complain now about the MM being unable to fill a game with tier 1 and 2 folks, when it's reportedly stuffed full of people who don't belong there.

Even if it represents the top 10%, the competitive folks in there will still be vastly outnumbered by people with a much more mundane, even if competent, skillset.

Also keep in mind the longer anyone spends in the higher tiers, the less impressive their stats will look. WLR and KDR don't differentiate between how tough the competition is.

[Revis Volek]

Fleeb the Mad, on 11 January 2018 - 06:51 PM, said:

Guys, you need to have some more realistic views of the actual size of the various player pools.

Tier 1 isn't for the top 1% competitive folks and it never was. Those people are in it, but creating a matchmaking bucket for such a small segment of the player base by itself is pointless. People complain now about the MM being unable to fill a game with tier 1 and 2 folks, when it's reportedly stuffed full of people who don't belong there.

Even if it represents the top 10%, the competitive folks in there will still be vastly outnumbered by people with a much more mundane, even if competent, skillset.

Also keep in mind the longer anyone spends in the higher tiers, the less impressive their stats will look. WLR and KDR don't differentiate between how tough the competition is.

Did you happen to notice where the Op ranked?

He almost right in the middle of the total monthly avg player population, so you are saying its ok that half the game is T1?

Edited by Revis Volek, 11 January 2018 - 08:29 PM.

[Shifty McSwift]

I feel the same way Dave. I know I shouldn't be in the top ranking system of the game, but I am, along with many others who perform way below me too. Kudos to you for being able to admit it <o

[Davegt27]

a few points
1) from this http://mwomercs.com/...28#entry3185728

PGI's vision

Quote

[color=orange]Our Vision for Player Experience:[/color]

• Drop into a match and play against a team of reasonably equal skilled players and equally balanced lance builds.
  
  • The ideal lance build would consist of 1 Assault, 1 Heavy, 1 Medium and 1 Light ‘Mech to bring the full breadth of play types to the battlefield. We do understand that this isn’t the best lance build within a company since you’d probably want to have your scout lance, skirmish lance, support lance and assault lance. This 1/1/1/1 build however would bring lances together that made a company build a lot more strategically balanced.
• Teams should not be mixed with multiple groups on 1 team and solo players on the other. An equal balance must be maintained due to the groups having much more communication and can affect how a team performs.
  
• As a 12-player group, we want the player to be able to choose who they fight against. This is a major feature for the competitive groups and league players alike.

Here are our steps involved with solving these issues in a way that keeps our vision integrity intact and at the same time address some of the hot topics within the community.

point 2)
PGI's goal is have game play consisting of

"Drop into a match and play against a team of reasonably equal skilled players and equally balanced lance builds."

point 3)
flaws in thinking equals flaws in game

a player can be defined as (PSR) add to that (build)

ok we know what PSR is what is the definition of "build" well a build in essence is a weight class

"The ideal lance build would consist of 1 Assault, 1 Heavy, 1 Medium and 1 Light ‘Mech to bring the full breadth of play types to the battlefield."

PGIs major flaw is using weight class and not factoring in "fire power"

"Fire Power" in its most simplistic form fire power would be raw fire power a given mech has at any given time

in a more complex format Fire Power could be defined as BV battle value or combat effectiveness rating,

any system that does not factor in the fire power of a Mech is a flawed system

IMHO
David

[Shifty McSwift]

Davegt27, on 12 January 2018 - 02:32 AM, said:

a few points
1) from this http://mwomercs.com/...28#entry3185728

PGI's vision



point 2)
PGI's goal is have game play consisting of

"Drop into a match and play against a team of reasonably equal skilled players and equally balanced lance builds."

point 3)
flaws in thinking equals flaws in game

a player can be defined as (PSR) add to that (build)

ok we know what PSR is what is the definition of "build" well a build in essence is a weight class

"The ideal lance build would consist of 1 Assault, 1 Heavy, 1 Medium and 1 Light ‘Mech to bring the full breadth of play types to the battlefield."

PGIs major flaw is using weight class and not factoring in "fire power"

"Fire Power" in its most simplistic form fire power would be raw fire power a given mech has at any given time

in a more complex format Fire Power could be defined as BV battle value or combat effectiveness rating,

any system that does not factor in the fire power of a Mech is a flawed system

IMHO
David

Experience with mechs/mech classes should be an important factor in placement from game to game also, the point I commonly raise in relation to this is that I am probably a t2 (1 by the current definition) assault/heavy pilot, but likely a t5/4 light pilot.

If the MM weighs my tier as 1 when I drop in a light I am just adding a false rating to my teams "tier weight" if there is such a thing.

[DaZur]

IMHO...

The biggest problem with the existing system is its predicated on a team win or loss.

The premise that "I" as an individual am being ranked on the win/loss of a team is just asinine.

Yes, of course you can still rank up in a loss dependant upon your damage accrument but that's a cheap shoot too... You can get a good damage score and still be a burden on your team if you don't kill off anyone.

Long and short, PSR should be a composite of personal performance and not some arbitrary outcome predicated by your teams performance.

[Zergling]

I know that my own enjoyment of the game dropped drastically when I hit Tier 1.

TBH, I'd prefer if PSR Tier was removed entirely, so the game has matchmaker that doesn't bother trying to adjust for player skill at all.

[Mcgral18]

DaFrog, on 11 January 2018 - 03:28 PM, said:

lol at that leaderboard, My t2 alt account is higher than my t1 main account.
That says a lot about leaderboards.

Which makes sense


Your Alt would face lesser opponents, meaning you more easily score higher

There is no Tier separation, it's just a collection of leaderboard stats.

[blood4blood]

Just get rid of Tiers, PSR, ELO, and the matchmaker for everything but weight class. I've encountered at least four different founders, and innumerable team members (i.e. playing on competitive teams) in my solo QP games tonight. ....and I'm playing my Tier 4 alt account I created a week ago. MM ain't nuthin'. As in, it ain't doin' nuthin' so just get rid of it.

[Brain Cancer]

Tiering was an illusionary separation by skill when it began (because initial Tier was based on the old ranking system), but that just has been steadily undermined with time, given PSR's constant shoves upwards.

Worse, since more and more average-to-worse players end up in T2+, it's actually more likely for mediocre players to end up in a pile of similar ones, sprinkled with good players desperate to carry above the derp and hoping against hope that their side didn't get all the taters in the field this time. I've seen games where I'm sitting there with multiple well-known death machines in QP and I know before the match even starts that the reds are running without lube and getting reamed.

The last one I checked, there was three T3 players on the red side and at least as many T1 players on our side with obscene W/L and K/D ratios on a 12-0 crushing that the only issue was a target lasting long enough for me to shoot at it. If that's how wide a gap there is with the current system getting matched up, something has gone horribly wrong.

[BTGbullseye]

Davegt27, on 10 January 2018 - 03:37 PM, said:

I was looking at Jarl's list

https://leaderboard.isengrim.org/

and my overall rank is 26259

26259! I don't think you can get much lower then that
it also means if you see me on your team your chances of winning just
went down I mean come on why am I even in tier 1 it makes no sense

its obvious the current PSR system is not working
also PGI will never listen to me no matter how much money I spend
since my low scores mean in there eyes it qualifies me as low like scum

can any of you try hards talk some sense into PGI
the PSR is not working

GLHF

You should see mine... I placed 406th place in season 14... (and bottom of the barrel for seasons 11-13, and 17) Fully T5, and didn't make it out of T5 until season 17. Season 18 has me in the top 4500 again, but I'm not using LRM assaults this season. (I was using the MCII-2/4 for 70x LRM with 2000+ ammo and backup ERMLs, got MVP in 95% of my matches because of it) Now I'm using ballistics, trying to get used to them.

Edited by BTGbullseye, 14 January 2018 - 01:36 AM.

[Dogstar]

I started a new account just to see how the tiering works from new and I notice that a new account is placed quite high up in tier 5 so it only takes a few good games to get to tier 4.

Which means that effectively there are only four tiers anyway, because to stay in tier 5 you would have to actively play to lose.