Jump to content

Tier And Psr Reset?


95 replies to this topic

#41 Dungeon 206

    Member

  • PipPipPipPipPip
  • The Wolf
  • The Wolf
  • 172 posts

Posted 17 December 2017 - 10:08 PM

View PostSunstruck, on 17 December 2017 - 03:20 PM, said:

That would basically be letting T1s get that 12kill acheivement, they would throw a few games to get stuck in the lower Tier and then seal club the hell out of you guys.


this is a nonsense statement. if you allocate tiers according to monthly statistics like match score,
it wouldnt take "just a few matches" to engineer a tier drop. you have to play crap for the entire month or more, and thats easily 150 games or more.

im pushing for a tier reset or a tier reallocation or a whatever as long as we get THEM POTATOES OUT OF TIER ONE AND INTO A TIER THAT SUITS THEM

#42 mistlynx4life

    Member

  • PipPipPipPipPipPip
  • 351 posts

Posted 17 December 2017 - 10:34 PM

Out of curiosity, is there a way to see how many players there are in each tier?

#43 MrMadguy

    Member

  • PipPipPipPipPipPipPipPipPip
  • 2,221 posts

Posted 17 December 2017 - 10:35 PM

Yeah, I agree, that we need to reset PSR and Tiers, but first...WE NEED TO FIX PSR NOT TO BE SO BIASED TOWARDS INCREASING. Because, you know, after just few weeks you will be at the same place, you're now. Another bad thing - I guess, PGI don't even use any MM now, cuz amount of players just not enough and queues would be too long with MM active. So, it's pointless anyway.

#44 Dungeon 206

    Member

  • PipPipPipPipPip
  • The Wolf
  • The Wolf
  • 172 posts

Posted 17 December 2017 - 10:47 PM

its all about improvement.
getting the tier system correct might not solve all the issues for example times where there just arent enough players,
but it will improve the experience during times where there are enough for the MM to actually work accordingly.
and that is an improvement, even if it doesnt solve everything

the best suggestion so far has been allocating tiers based on statistics like average match score or KDR over 1 - 3 months.
this means that if a player is playing consistently well he will be bumped up
if he is consistently poor then he will be sent down. over time the system will balance itself so that players will stay at a tier appropriate to their skill level.

allocating tiers according to stats across 1 - 3 months also eliminates the possibility of engineering a tier drop to "club seals"
have you actually tried to play badly for a whole month or two? as in 150-300 games of playing like rubbish JUST SO you can drop tier?


and even if you somehow manage to drop tier just so you can seal club. in a month you'll be back up to a tier where you get seal clubbed.

PGI.
please take on this suggesting. its ridiculous how many potatoes there are in Tier 1, potatoes who dont even know to press "R"!

#45 Invictus XVII

    Member

  • PipPipPip
  • Survivor
  • 50 posts

Posted 18 December 2017 - 02:09 AM

Id like them to scrap the tier and matchscore system entirely.
Its not working and you cant lose your tier ranking once you reach the top.

Why not have it se to league rankings every quarter, which affect your matchmaking?
And why not base your league on your actual performance? why not have a set number of players in each legue so you have to compete for the upper ranks (could work as a motivator)

Newer players dont want to get steamrolled by veterans and veterans dont want to have to carry potato players. Win win.

#46 Vellron2005

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • The Blood-Eye
  • The Blood-Eye
  • 5,444 posts
  • LocationIn the mechbay, telling the techs to put extra LRM ammo on.

Posted 18 December 2017 - 03:39 AM

Personally, I'd be interested in a full Tier reset, but done like this:

PGI says "we're gonna reset the tiers on XX.XX.2018". Then, when that "event" starts, you get a certain amount of time. In that time frame, your best 100 matches are taken into account, and you are put in a tier according to that. (If you play less than 100 matches, then the rest are taken from your last matches before the event, or you are tier 3 by default, until you play 100 matches and get ranked)

Not the last two years of playing. Just those 100 matches.

And they do this every 3 months.

That way, your PSR Tier would reflect your current skill, not the skill you had 6 months ago, or a year ago..

Ideally, I would go even further and divide the PSR rating by weight class..

I don't know about you, but I don't think I'm equally as good in lights as I am in heavies or assaults.. That should be taken into account when matchmaking..

Edited by Vellron2005, 18 December 2017 - 03:46 AM.


#47 Wolfways

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Mercenary
  • 6,499 posts
  • LocationIn a shutdown overheated mech near you.

Posted 18 December 2017 - 03:49 AM

View PostVellron2005, on 18 December 2017 - 03:39 AM, said:

I don't know about you, but I don't think I'm equally as good in lights as I am in heavies or assaults.. That should be taken into account when matchmaking..

The problem there is that there is a huge gap between good and bad mechs within classes.
PSR/mech would be good imo.

#48 Lily from animove

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • The Devoted
  • The Devoted
  • 13,891 posts
  • LocationOn a dropship to Terra

Posted 18 December 2017 - 04:33 AM

Without PSR change that makes no sense.

View PostBilbo999, on 15 December 2017 - 11:57 PM, said:

I'd be happy if the tiers were reset according to historical data, as it was done when the tiers were introduced. I suggest you find a unit though, because resetting tiers isn't going to make you happy until you get out of QP altogether.


that would make us end with the same rating as before. So you reset and recalculate.

#49 MortallyTransparentCupcake

    Member

  • PipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 75 posts
  • LocationCanada

Posted 18 December 2017 - 04:41 AM

I could get behind a Tier reset, but I'm not so sure about the stats wipe. We were all new once. Sure, some people made a new account when they "got good," but that's their own insecurity.
I was surprised when I found out that the Tiers don't reset once every 6 months or so. Most games do a similar function, and I am a bit shocked this game doesn't do this as well.

#50 Vellron2005

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • The Blood-Eye
  • The Blood-Eye
  • 5,444 posts
  • LocationIn the mechbay, telling the techs to put extra LRM ammo on.

Posted 18 December 2017 - 04:49 AM

View PostWolfways, on 18 December 2017 - 03:49 AM, said:

The problem there is that there is a huge gap between good and bad mechs within classes.
PSR/mech would be good imo.


I agree, taking again last 100 matches into account.. but that seems like more work PGI is willing to do than just by weight class..

Just classes would be a start, while per mech would be a better solution..

#51 GabrielSun

    Member

  • PipPipPipPipPip
  • Philanthropist
  • Philanthropist
  • 171 posts

Posted 18 December 2017 - 09:29 AM

As long as PSR bonus is tied directly to win/loss of the team you will have it just be an experience bar. Have it tied strictly to individual average match score and you will see true separation of skill ranks.

I get 700 dmg and the team loses I get equal PSR. I get 700 dmg and the team wins I get a huge increase. Wrong method of assuring people get into the correct ranks.

#52 Humpday

    Member

  • PipPipPipPipPipPipPipPip
  • The Pharaoh
  • The Pharaoh
  • 1,463 posts

Posted 18 December 2017 - 10:37 AM

Ranks within tiers themselves "could" help this situation of if the population was high enough, but alas...nope.
I will say, I don't know what was going on this weekend or what people were doing but, this weekend in particular was really bad.

Some strange stuff going on there.

#53 mistlynx4life

    Member

  • PipPipPipPipPipPip
  • 351 posts

Posted 18 December 2017 - 11:56 AM

View PostGabrielSun, on 18 December 2017 - 09:29 AM, said:

As long as PSR bonus is tied directly to win/loss of the team you will have it just be an experience bar. Have it tied strictly to individual average match score and you will see true separation of skill ranks.

I get 700 dmg and the team loses I get equal PSR. I get 700 dmg and the team wins I get a huge increase. Wrong method of assuring people get into the correct ranks.

That's not a tie to W/L, that's a tie to damage and that's the primary problem with the PSR - it's weighted to favor damage above anything else. You have people who get consistently high damage in every match, regardless of teamwork or outcome, and they rise to T1. This is why you have people who say 'I'm in T1 but I'm no pro player...?' It's because they aren't. They are consistently high damage-dealers. Conversely, you have people who don't get that but who are excellent pilots and they're still in T4 or T3 after years. It's because they are favoring builds and tactics that don't produce consistently high damage. This doesn't mean there aren't truly skilled players in T1 or truly derpy folks in the lower tiers, only that the PSR is not unbiased in what factors determine advancement.

Resetting a broken system doesn't fix things, no matter how often you reset it - it's just a challenge to get back to the top. In fact, I bet you'd see a lot less Light pilots in the higher tiers if you just reset the rankings every so often because Lights don't typically (there are exceptions to everything, of course) put out the requisite consistently high damage to produce consistently high match scores. It's a weird system. Giving equal weight to all factors (damage, kills, caps, spots, stealth TAGs, etc.) except Win/Loss (let that weigh heavier) might result in a 'Whatever you're doing, you're good at it and it wins your matches' kind of dynamic but I'm not a mathematician so I can't begin to imagine what that might look like.

#54 AzureRathalos

    Member

  • PipPipPipPipPip
  • The Deadly
  • The Deadly
  • 185 posts

Posted 18 December 2017 - 02:54 PM

There are so many other factors that can mess with results. Making a new system or refining the current one will not be as easy as some people make it out to be.

We've had plenty of examples of a good player getting stuck in a team that collapses and how team win/loss should not be counted for an individual's rankings. What about the other way around?

Let's say a player goes down very early with very little damage and score. However, they lead the team to victory from beyond the grave with excellent calling over voice chat. In a system where win/loss doesn't count, the player's rank would drop like a rock. In this case, wouldn't it be worth counting in the team's win?

#55 Telemachus -Salt Wife Salt Life-

    Member

  • PipPipPipPipPipPip
  • The Deadly
  • The Deadly
  • 364 posts
  • LocationTerra

Posted 18 December 2017 - 03:12 PM

I didn't read this whole thread but I am completely behind the OPs idea as long as I get to start in Tier 5 again!

#56 HammerMaster

    Member

  • PipPipPipPipPipPipPipPipPip
  • The Benefactor
  • The Benefactor
  • 2,516 posts
  • LocationNew Hampshire, USA

Posted 18 December 2017 - 03:55 PM

Reset? Not needed. Maybe a readjustment?
Wipe? No. The stats are the numbers we can use.
But something has to change. I give 2 rats about comp. I play to win but I have a full time job and overtime that's NOT MWO.
I'm tier 1 because I played enough. So match maker puts me in with player "Tier up" who loves meta and carries a 5.0 KDR or W/L or something. Is this fair to either of us? No I don't belong in tiers 3,4, or 5. This guy performs theoretically 5 times better but the metric is not that granular.

Edited by HammerMaster, 18 December 2017 - 04:10 PM.


#57 Brain Cancer

    Member

  • PipPipPipPipPipPipPipPipPip
  • The 1 Percent
  • 3,851 posts

Posted 18 December 2017 - 04:13 PM

Quote

I'm tier 1 because I played enough. So match maker puts me in with player "Tier up" who loves meta and carries a 5.0 KDR or W/L or something. Is this fair to either of us? No I don't belong in tiers 3,4, or 5. This guy performs theoretically 5 times better but the metric is not that granular.


it's perfectly fair for "Tier up". He's got a constant supply of marginally skilled (compared to him) target practice to slaughter and rarely has to wait long for a game, as PGI will not only shove endless targets into his rank, it'll even go down two tiers to make sure there's plenty more. The massive positive bias means as many players as possible are put into the same tank as the sharks, regardless of actual fairness. This means the people most likely to feed money hang around- long term players sit in a pool that is refreshed with newbies, regardless of if they learn anything or not.

PGI would much rather sacrifice them than the tryhards buying that $200 set of new robots with every release.

#58 Xmith

    Member

  • PipPipPipPipPipPipPipPip
  • The Ironclad
  • The Ironclad
  • 1,099 posts
  • LocationUSA

Posted 18 December 2017 - 04:21 PM

View PostNovember11th, on 15 December 2017 - 11:29 PM, said:

Greetings Mechwarriors.

How would all of you feel if there was a PSR/Teir wipe, in which everyone is given a clean slate and then given the opportunity to again climb their way through the rankings once again?

I've been giving some thoughts as to how to cleans the higher rankings of potatos and putting legacy (133t) potatos back where they belong with the rest of the spuds. That may sound incredibly elitist, but hear me out; I just want to make any given random pug unit assembled by the match maker in QP some better cohesion by allowing players of actual skill level to be actually placed among their peers.

I'm really tired of being dropped into matches where 2-5 of the players on my side are either suicidal lone wolves (looking at you cheeta players) or stationary missile/laser/gauss platforms. (if your going to use LRMS, at least do it right and stop giving the rest of us a bad rep. FFS!)... and dont get me started on griefers/tkers that work as teams on both sides.

I like being part of a well coordinated pack of hunters and/or traders, as I'm sure most of you do as well. I feel as if this is our best solution to fixing PSR without actually having to overhaul anything or create major game changes. Let the spuds be spuds, and let the rest of us actually play to win.

What do all of you think? Is it a feasible idea? Is it even a good idea?

Wow, everyone starts out at the same level. This would be a great opportunity to pad my stats. It would a lot more fun than it is now. New players will be farmed like crazy. Yeah, let's do it.

#59 HammerMaster

    Member

  • PipPipPipPipPipPipPipPipPip
  • The Benefactor
  • The Benefactor
  • 2,516 posts
  • LocationNew Hampshire, USA

Posted 18 December 2017 - 04:24 PM

View PostBrain Cancer, on 18 December 2017 - 04:13 PM, said:


it's perfectly fair for "Tier up". He's got a constant supply of marginally skilled (compared to him) target practice to slaughter and rarely has to wait long for a game, as PGI will not only shove endless targets into his rank, it'll even go down two tiers to make sure there's plenty more. The massive positive bias means as many players as possible are put into the same tank as the sharks, regardless of actual fairness. This means the people most likely to feed money hang around- long term players sit in a pool that is refreshed with newbies, regardless of if they learn anything or not.

PGI would much rather sacrifice them than the tryhards buying that $200 set of new robots with every release.


More than these try hards spend $. If my wife knew how much...

#60 mistlynx4life

    Member

  • PipPipPipPipPipPip
  • 351 posts

Posted 18 December 2017 - 04:28 PM

View PostBrain Cancer, on 18 December 2017 - 04:13 PM, said:

it's perfectly fair for "Tier up". He's got a constant supply of marginally skilled (compared to him) target practice to slaughter and rarely has to wait long for a game, as PGI will not only shove endless targets into his rank, it'll even go down two tiers to make sure there's plenty more. The massive positive bias means as many players as possible are put into the same tank as the sharks, regardless of actual fairness. This means the people most likely to feed money hang around- long term players sit in a pool that is refreshed with newbies, regardless of if they learn anything or not.

PGI would much rather sacrifice them than the tryhards buying that $200 set of new robots with every release.


I've never thought about it like that but wow is that an astute observation.





21 user(s) are reading this topic

0 members, 21 guests, 0 anonymous users