Jump to content

Balance Discussion - Aug 2018 - Post Podcast Feedback

Balance

605 replies to this topic

#461 Sable Dove

    Member

  • PipPipPipPipPipPipPipPip
  • Knight Errant
  • 1,005 posts

Posted 12 August 2018 - 12:37 AM

View PostChris Lowrey, on 10 August 2018 - 03:50 PM, said:


I never understood that myself since even in TT, each turn is supposed to be a rough equivalent of a 10 second time period, and you can usually focus down a 'Mech in 4-5 turns depending on the 'Mechs involved with "Average" Pilots. Even less once Advanced / Clan tech or elite / Ace pilots start to get involved.


Is there actually official documentation that says this? I'm not super familiar with the tabletop game, but I've seen claims varying from 10 seconds to 6 minutes (based on weapon ranges vs movement speeds). So focusing a mech down in 50-60 seconds is quite different from focusing a mech down in 25-30 minutes.

Personally, I don't think balance should have much to do with the tabletop, but longer TTK would be more fun, for the most part. Especially in QP, no one wants to push because the first two mechs or so that come out usually get obliterated because five or more mechs are focusing them down at once, so TTK in a 1-on-1 being 50-60 seconds sounds okay, but when the first mechs out are facing 5 or 6 mechs worth of fire, TTK is about 10 seconds, which is less than ideal when you spent 2 minutes finding a match, 3 minutes in the lobby, 2 minutes walking to the fight, and get 10 seconds of fighting. So as a result everyone is timid, and no one wants to be the first ones to push because they're almost guaranteed not to have a good time.

#462 justcallme A S H

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • CS 2020 Referee
  • CS 2020 Referee
  • 8,987 posts
  • LocationMelbourne, AU

Posted 12 August 2018 - 01:21 AM

View PostKhobai, on 11 August 2018 - 11:07 PM, said:

Adding a functional matchmaker is no longer a feasible solution.

Because for a matchmaker to function properly a game needs a consistently high population of players to draw from. That way the game can match skilled players against other skilled players and doesnt have to put skilled players in lower skill games.
Spoiler



And what information and numbers are you basing that off, exactly? I don't know why I'm bothering to explain with stats as I know how that ends, but here we go...

Lets go with what we know for now from the data scrapes we have access too.
  • ~57k players with 10 matches per season in the last 3 months. 28k was last months actual.
  • Median for Average Match Score (AMS) is 210-220 AMS.
  • 220AMS means you are, roughly, around player 16,000 out of the 57,000.
  • 190AMS / WLR 0.9 / KDR 0.5 / 10k played games is enough to get into Tier 1, even though this puts you well below the median.
With 200AMS and above it should put you in the low end of T3 if the middle if T3 is for 220AMS. That means between Tier 1 / 2 / 3 you have around 23,000 people in the pool. Between 200AMS and 190AMS there is only around 5,000 players missing and some of those won't even be in Tier 1/2/3 anyway as they simply have not played enough matches to progress. So it is, more than likely, lower than 5k. We are now talking around 17% of players.


Now it would of course have to be watched closely and actively tuned in the inital months if things were changed. It might mean 'expanding' the match maker to include Tier 4 after say 5-6mins which would allow people with 170-190AMS back into the pool. Before you (or anyone else) uses that to try dismantle my argument - they are in Tier 1-3 games right now so there is not a big difference there to add them in, now and again as opposed to always.

So there you have it - Some stats and numbers. Really keen to see where your numbers/stats and then reasoning is drawn from.

* NOTE: This is a little bit of educated guess work, of course. There will no doubt be minor errors in the numbers as only PGI knows how many people are truly in what Tier and their AMS etc etc. However I feel that I am close enough and there is certainly a potential for a workable solution.

Edited by justcallme A S H, 12 August 2018 - 01:27 AM.


#463 Y E O N N E

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • The Nimble
  • The Nimble
  • 16,810 posts

Posted 12 August 2018 - 03:02 AM

So, it was stated in the OP of this thread that there is a gulf between Clan and IS laser vomit. However, that gulf was not elaborated on, and because it seems like solutions are being proposed without analyzing the problem, I have made an initial attempt at that analysis.

You can find it in the PTS subforum at this link.

There seems to be a prevailing thought that limiting the heat capacity will reduce this Clan gulf. I submit that it will not, because Clans run too cold and have the option for more raw DPS as well as sustained DPS, as you will see if you take the time to read through the analysis.

#464 Korz

    Member

  • PipPipPipPipPip
  • The Hawk
  • The Hawk
  • 172 posts

Posted 12 August 2018 - 06:35 AM

View Postjustcallme A S H, on 12 August 2018 - 01:21 AM, said:


And what information and numbers are you basing that off, exactly? I don't know why I'm bothering to explain with stats as I know how that ends, but here we go...

Lets go with what we know for now from the data scrapes we have access too.
  • ~57k players with 10 matches per season in the last 3 months. 28k was last months actual.
  • Median for Average Match Score (AMS) is 210-220 AMS.
  • 220AMS means you are, roughly, around player 16,000 out of the 57,000.
  • 190AMS / WLR 0.9 / KDR 0.5 / 10k played games is enough to get into Tier 1, even though this puts you well below the median.
With 200AMS and above it should put you in the low end of T3 if the middle if T3 is for 220AMS. That means between Tier 1 / 2 / 3 you have around 23,000 people in the pool. Between 200AMS and 190AMS there is only around 5,000 players missing and some of those won't even be in Tier 1/2/3 anyway as they simply have not played enough matches to progress. So it is, more than likely, lower than 5k. We are now talking around 17% of players.



Now it would of course have to be watched closely and actively tuned in the inital months if things were changed. It might mean 'expanding' the match maker to include Tier 4 after say 5-6mins which would allow people with 170-190AMS back into the pool. Before you (or anyone else) uses that to try dismantle my argument - they are in Tier 1-3 games right now so there is not a big difference there to add them in, now and again as opposed to always.

So there you have it - Some stats and numbers. Really keen to see where your numbers/stats and then reasoning is drawn from.

* NOTE: This is a little bit of educated guess work, of course. There will no doubt be minor errors in the numbers as only PGI knows how many people are truly in what Tier and their AMS etc etc. However I feel that I am close enough and there is certainly a potential for a workable solution.



I consider myself a solid tier 3-4 player and I am ok with that. I want to get better and with time I might. So me being locked to that tier with either resets or just based off match performance with no increase in number of battles is a very good idea to me.

Now here is were we have the issue. Snowflakes that think they are better then tier X and should be tier X. So the real way to do this is to get rid of displaying the tier. Fix the metrics by taking out the number of battles fought. Maybe do a reset every 3 months.

Now anytime someone has to wait more then 1 min for a match setup they can get priority in the que. So the longer you have to wait either cause of your tier or mech the more priority that match maker will give you. And when making matches if you mix tiers of players mix them evenly. 5 tier 1 players then one gets set to another match setup and 2 get placed in each team in the current match build. Give the guy moved out a priority token like someone that waited. Each priority token lowers the optimal till you get a match.

Now this might mean on occasions you would have a tier 1 all alone in a tier 2 or 3 battle or even on off times a 4 or 5. But no one would have to wait more then 5 min or so for a match in the QP que this way.

#465 SilentScreamer

    Member

  • PipPipPipPipPipPipPip
  • FP Veteran - Beta 1
  • FP Veteran - Beta 1
  • 556 posts

Posted 12 August 2018 - 07:31 AM

View Postjustcallme A S H, on 12 August 2018 - 01:21 AM, said:



And what information and numbers are you basing that off, exactly? I don't know why I'm bothering to explain with stats as I know how that ends, but here we go...

Lets go with what we know for now from the data scrapes we have access too.
  • ~57k players with 10 matches per season in the last 3 months. 28k was last months actual.
  • Median for Average Match Score (AMS) is 210-220 AMS.
  • 220AMS means you are, roughly, around player 16,000 out of the 57,000.
  • 190AMS / WLR 0.9 / KDR 0.5 / 10k played games is enough to get into Tier 1, even though this puts you well below the median.
With 200AMS and above it should put you in the low end of T3 if the middle if T3 is for 220AMS. That means between Tier 1 / 2 / 3 you have around 23,000 people in the pool. Between 200AMS and 190AMS there is only around 5,000 players missing and some of those won't even be in Tier 1/2/3 anyway as they simply have not played enough matches to progress. So it is, more than likely, lower than 5k. We are now talking around 17% of players.



I really hate when players insist that a Tier 1 player should have a certain KDR, or W/L Ratio. hat insist you need to do 500 damage per match, those numbers can't be sustained on a 12 man team, there isn't enough enemy armor to go around. The same is true for kills and wins. If you populated a match with players of equal skill and identical mechs:

- number theory says at least half would die with no kills, as mutual kills are rare in MWO
- assuming the match was played down to the last mech (end score 12-11), same theory says the remaining 12 players in match would split the remaining 11 kills. So 12 of 24 players split 23 kils.
- That's an average of 1.91 kills for those 12 pilots, all but one of those pilots will have a death.
- exactly half the players get a win, exactly half get a loss. W/L Ratio 1.00

If the bottom pilots in a match switch between elimination at zero kills and 1 kill each match and alternate being on the winning team that is a KDR of 0.5 and W/L of 1.00. Yet you insist,

View Postjustcallme A S H, on 12 August 2018 - 01:21 AM, said:

*190AMS / WLR 0.9 / KDR 0.5 / 10k played games is enough to get into Tier 1, even though this puts you well below the median.
I say it is AT the median, not below it, and perfectly acceptable if you are Tier 1, but not "top" Tier1. The pilots that get a kill in 1/3 or 1/4 matches or drop to W/L of 0.8 are the ones that mighr be better suited to Tier 2 or Tier 3.

Also leaderboard stats are garbage, as a Wins, Kills, and Matchscore do not take into account what Tier a pilot plays at. I would give credit to leaderboard stats if each Tier had seperate stat tracking based on what Tier the pilot was in when he played the match. Due to how PSR works, pilots who are Tier 4 or Tier 5 for years can put up good numbers and stay in Tier 4 or 5 perpetually.

Edited by SilentScreamer, 12 August 2018 - 09:07 AM.


#466 Pricklyhedgehog

    Rookie

  • The Predator
  • The Predator
  • 3 posts

Posted 12 August 2018 - 09:32 AM

I watched the podcast, and I don't envy the task you have, Chris and Paul. One of the more useful solutions I see are the discussions centered on heat disspation and cooling - deleting cool shots or reducing their cool down time as one effort to quell that element in the game. It does seem, in general, that the Clan 'mechs enjoy the ability to not only cluster large number of laser weapons for incredible focused damage, and strike more repeatedly for sustained DPS than the IS counterparts. I think one example in the podcast was the seldom seen Grasshopper v. Ebon Jaguar, and the GH not being able to out laze its Clan equivalents.

I recognize that while Clan 'tech enjoys lighter tonnage weapons, and less slots, is it feasible to increase the number of slots for Clan double heat-sinks from 2 to 3, to match IS? Increasing DBHS slots will reduce the number of slots available for weapons, hindering to some extent the capacity for cooling, and boating the larger laser weapons. I'm still intrigued by the heat disspation thread that someone else posted earlier, oh Jman5's post, and I think he's got an interesting point about the way in which heat dissipation works, which is hard to grasp.

Part of the problem are the variables we have in-game also, which I think affects player perception, and I've touched on this in vid's I've done when dealing with tactical coordination - trying to rally and herd cats to common objectives because of competing experience levels and build types. But I wonder often in terms of data analysis how this influences information gathering It keeps coming up in the discussion here in terms of player skill levels etc. My own ob's here I'm referring to quick play:

We have a wide variety of player skill levels, even within the various tiers themselves at times. Within the 24 players we have all kinds of builds ranging from the oft despised LRM boats, to the laser vomit clans such as the Hellbringer, and across variable maps that favor some builds over others. Even this can change during the course of the match as it evolves or devolves mid to late game. Aside from perhaps the skill issue, which may be perceived or real - it's sometimes hard to tell - this variety is usually good, but it can, and often does mean for lopsided and unpredictable games in QP, especially during events where people haul out their LRM boats to quickly amass the damage/kills they need, and then have time to pack up, kick back to Knickleback, and have a cup of tea. Those of us with pride and a conscience, are still grinding away...

If out of all this variability the common thread is the clan laser vomit, again perhaps the answer centers on the heat viability. As Mr. Lang said in the podcast, he does not think lasers are a skilled weapon, and he has a point. It begs questions about their effectiveness in pin-point component destruction for quick kills - the thing you're wanting to tone down.

I wish you guys luck, and thumbs up to the smarty-pants spreadsheet mathmatical-mathlete types for posting all these thought provoking concepts, reinforced with numbers...I'm not but a simple history/English minor type.

Salute - Pricklyhedgehog

#467 Jonathan8883

    Member

  • PipPipPipPipPipPipPip
  • Shredder
  • Shredder
  • 708 posts

Posted 12 August 2018 - 09:54 AM

View PostPricklyhedgehog, on 12 August 2018 - 09:32 AM, said:

I recognize that while Clan 'tech enjoys lighter tonnage weapons, and less slots, is it feasible to increase the number of slots for Clan double heat-sinks from 2 to 3, to match IS? Increasing DBHS slots will reduce the number of slots available for weapons, hindering to some extent the capacity for cooling, and boating the larger laser weapons. I'm still intrigued by the heat disspation thread that someone else posted earlier, oh Jman5's post, and I think he's got an interesting point about the way in which heat dissipation works, which is hard to grasp.

This applies only to mechs with sufficient tonnage and crit space (typically, but not always, Battlemechs). It's very difficult to build a Warhawk, Executioner, or Gargoyle that can keep up in ANY stat other than raw HP with a Hellbringer, Mad Dog, or even some of the Hunchback IIC variants. Summoners aren't much better. There are a lot of non-problematic Omnimechs that are walking nerf bats due to limited hardpoints (Warhawk), terrible size/hitboxes, lack of crit slots (Warhawk), or lack of suitable tonnage (Gargoyle, Summoner).

Any global weapon nerf/debuff is going to be unfair to the bad Omnimechs unless they are compensated with sufficient quirks.

#468 Horseman

    Member

  • PipPipPipPipPipPipPipPipPip
  • The Mercenary
  • The Mercenary
  • 4,737 posts
  • LocationPoland

Posted 12 August 2018 - 10:25 AM

View PostPricklyhedgehog, on 12 August 2018 - 09:32 AM, said:

I recognize that while Clan 'tech enjoys lighter tonnage weapons, and less slots, is it feasible to increase the number of slots for Clan double heat-sinks from 2 to 3, to match IS? Increasing DBHS slots will reduce the number of slots available for weapons, hindering to some extent the capacity for cooling, and boating the larger laser weapons. I'm still intrigued by the heat disspation thread that someone else posted earlier, oh Jman5's post, and I think he's got an interesting point about the way in which heat dissipation works, which is hard to grasp.
That's a dead negative. Increasing weapon (or equipment) size/tonnage messes up stock variants.

Edited by Horseman, 12 August 2018 - 12:21 PM.


#469 SilentFenris

    Member

  • PipPipPipPipPip
  • Bridesmaid
  • Bridesmaid
  • 163 posts
  • LocationCalifornia

Posted 12 August 2018 - 10:37 AM

View PostJonathan8883, on 12 August 2018 - 09:54 AM, said:


This applies only to mechs with sufficient tonnage and crit space (typically, but not always, Battlemechs). It's very difficult to build a Warhawk, Executioner, or Gargoyle that can keep up in ANY stat other than raw HP with a Hellbringer, Mad Dog, or even some of the Hunchback IIC variants. Summoners aren't much better. There are a lot of non-problematic Omnimechs that are walking nerf bats due to limited hardpoints (Warhawk), terrible size/hitboxes, lack of crit slots (Warhawk), or lack of suitable tonnage (Gargoyle, Summoner).



Trying to kit out a Summoner for laser vomit like a EbonJag or Hellbringer is not the way to go. Not every mech can excel at trading laser vomit. In addition to crit space and free tonnage there is also profile (narrow with high hardpoints prefered) and hitboxes (do not want a huge CT if you are spreading incoming damage).

The Summoner performs well at short range when loaded with SRMs, or long range with double PPCs. Variety is good, if you like laser vomit, don't use the Summoner. Try it if you want something different.

View PostPricklyhedgehog, on 12 August 2018 - 09:32 AM, said:


I recognize that while Clan 'tech enjoys lighter tonnage weapons, and less slots, is it feasible to increase the number of slots for Clan double heat-sinks from 2 to 3, to match IS?


This would create a huge number of issues with omnimechs carrying sinks locked in a fixed location, particulary the CT and legs.

Edited by SilentFenris, 12 August 2018 - 10:44 AM.


#470 Pricklyhedgehog

    Rookie

  • The Predator
  • The Predator
  • 3 posts

Posted 12 August 2018 - 11:09 AM

View PostJonathan8883, on 12 August 2018 - 09:54 AM, said:

This applies only to mechs with sufficient tonnage and crit space (typically, but not always, Battlemechs). It's very difficult to build a Warhawk, Executioner, or Gargoyle that can keep up in ANY stat other than raw HP with a Hellbringer, Mad Dog, or even some of the Hunchback IIC variants. Summoners aren't much better. There are a lot of non-problematic Omnimechs that are walking nerf bats due to limited hardpoints (Warhawk), terrible size/hitboxes, lack of crit slots (Warhawk), or lack of suitable tonnage (Gargoyle, Summoner).

Any global weapon nerf/debuff is going to be unfair to the bad Omnimechs unless they are compensated with sufficient quirks.


Good point! I forgot about these 'mechs, and indeed the Summoner is one of my favorite heavy 'mechs. I was able to pull 700 damage the other day in the Pride, with only two PPC's, and a couple of machine guns, not bad considering the PPC's have been utterly nerfed to splash damage. The Madcat II has eclipsed the Warhawk, and both the WH and DWF are so big and boxy they are difficult to bring to bear in the fight.

It's a point I forgot to make too, in that the evolution of the game post the timeline advance, meant these new 'mechs, like the Mad Cat II, present yet another variable that have overshadowed and outshined many of the IS mechs in equivalent weight classes, often of course punching above their weight. I love the Gargoyle and even the Executioner, the latter I spent hundreds of hours in climbing through the tiers, and still holds my all time personal record of 8 kills in a match...sadly those thrilling days are long gone, like the old Frozen City Map, where said record was acquired. :(

#471 Kageru Ikazuchi

    Member

  • PipPipPipPipPipPipPipPip
  • The Determined
  • The Determined
  • 1,190 posts

Posted 12 August 2018 - 12:26 PM

View Postjustcallme A S H, on 12 August 2018 - 01:21 AM, said:

  • ~57k players with 10 matches per season in the last 3 months. 28k was last months actual.
  • Median for Average Match Score (AMS) is 210-220 AMS.
  • 220AMS means you are, roughly, around player 16,000 out of the 57,000.
  • 190AMS / WLR 0.9 / KDR 0.5 / 10k played games is enough to get into Tier 1, even though this puts you well below the median.



View PostSilentScreamer, on 12 August 2018 - 07:31 AM, said:

If the bottom pilots in a match switch between elimination at zero kills and 1 kill each match and alternate being on the winning team that is a KDR of 0.5 and W/L of 1.00. Yet you insist,
I say it is AT the median, not below it, and perfectly acceptable if you are Tier 1, but not "top" Tier1. The pilots that get a kill in 1/3 or 1/4 matches or drop to W/L of 0.8 are the ones that mighr be better suited to Tier 2 or Tier 3.

Median players should be in the middle tier.

Quote

Also leaderboard stats are garbage, as a Wins, Kills, and Matchscore do not take into account what Tier a pilot plays at. I would give credit to leaderboard stats if each Tier had seperate stat tracking based on what Tier the pilot was in when he played the match. Due to how PSR works, pilots who are Tier 4 or Tier 5 for years can put up good numbers and stay in Tier 4 or 5 perpetually.

Leaderboard stats are all we have to go with. And they are an aggregate of pilot performance over time, over tens, hundreds, and thousands of matches.

Also, if a Tier 4 player is "putting up good numbers", his PSR increases, and he will get advanced to Tier 3.

Edited by Kageru Ikazuchi, 12 August 2018 - 12:27 PM.


#472 SilentScreamer

    Member

  • PipPipPipPipPipPipPip
  • FP Veteran - Beta 1
  • FP Veteran - Beta 1
  • 556 posts

Posted 12 August 2018 - 12:37 PM

View PostKin3ticX, on 10 August 2018 - 03:12 PM, said:


To me, the people wanting more TTK are the ones that read Sarna where it says mech duels are supposed last for hours which is cool and all but it doesn't help MWO one bit.


View PostChris Lowrey, on 10 August 2018 - 03:50 PM, said:


I never understood that myself since even in TT, each turn is supposed to be a rough equivalent of a 10 second time period, and you can usually focus down a 'Mech in 4-5 turns depending on the 'Mechs involved with "Average" Pilots. Even less once Advanced / Clan tech or elite / Ace pilots start to get involved.


Duels starting face-to-face and mech-to-mech in an open arena which is smaller than effective weapon range would not take long. Like Victorian-era pistol duels, or two knights or samurai with swords, they would be quick and public. Clan duels, as well as famous mechwarriors and leaders, Kell, Kurrita, Wolf setup these kinds of duels.

Consider instead a duel between Mechwariors who are paid soldiers. Veteran combatants are survivors. Instead of an open arena they pick a secluded mountain range, jungle valley, desert or canyon. Instead of starting face-to-face they get dropped off seperately the night before, with the match to begin at a pre-aranged time. There is no time limit, there are no spectators, neither knows the location of their opponent.

Now you high stakes game of hide-and-seek. The match could very well last hours or even days without one sighting the other's mech. If you are caught off-guard, try to get to cover/disengage to begin the cat-and-mouse. Hopefully you will spot them first on your terms thus making you the cat, not the mouse. Think of the most gritty Raven 3L pilots you know playing Skirmish mode with no time limit.

#473 SilentScreamer

    Member

  • PipPipPipPipPipPipPip
  • FP Veteran - Beta 1
  • FP Veteran - Beta 1
  • 556 posts

Posted 12 August 2018 - 12:47 PM

View PostKageru Ikazuchi, on 12 August 2018 - 12:26 PM, said:


Also, if a Tier 4 player is "putting up good numbers", his PSR increases, and he will get advanced to Tier 3.


I like your thought, but honestly that is not how PSR increases. PSR increases with VICTORY, faster if you do well. If your team gets the LOSS, you have to have to an AMAZING score, typically top on your team for your PSR to go up. A 300 match score on a LOSS usually nets me a "=", or "no change" in my PSR.

View PostKageru Ikazuchi, on 12 August 2018 - 12:26 PM, said:


Median players should be in the middle tier.



Is "Median Tier" Tier 3? Do you realize that KDR and WIN/LOSS Ratio stats are calculated independent of a players actual PSR/Tier?

There may be a fundamental disconnect here. Try to follow:
First, PSR isn't working right. But let's pretend it is and that all Tier 1 pilots are THE BEST.
Second, let's pretend that the matchmaker can and does put only Tier 1 pilots in a match together.
- Half of your Tier 1 players will have below median performance in any number of given matches, so do you plan on instantly demoting them two Tiers to Tier 3?

That's like taking an "average" professional/olympic athlete and sending them back to play Highschool competition. Such an athlete would crush opponents at that level. Just because they aren't #1 in the race/game/score doesn't mean they are the top 1% in Skill/Talent. It just means tge competition is harder. Expecting every Tier 1 pilot to maintain top match score is impossible because they are playing against the other top pilots.

Edited by SilentScreamer, 12 August 2018 - 01:08 PM.


#474 Kageru Ikazuchi

    Member

  • PipPipPipPipPipPipPipPip
  • The Determined
  • The Determined
  • 1,190 posts

Posted 12 August 2018 - 01:34 PM

View PostSilentScreamer, on 12 August 2018 - 12:47 PM, said:

I like your thought, but honestly that is not how PSR increases. PSR increases with VICTORY, faster if you do well. If your team gets the LOSS, you have to have to an AMAZING score, typically top on your team for your PSR to go up. A 300 match score on a LOSS usually nets me a "=", or "no change" in my PSR.

Yes, but on average in the long term, with "good numbers" your PSR will increase.

Quote

Is "Median Tier" Tier 3? Do you realize that KDR and WIN/LOSS Ratio stats are calculated independent of a players actual PSR/Tier?

There may be a fundamental disconnect here. Try to follow:
First, PSR isn't working right. But let's pretend it is and that all Tier 1 pilots are THE BEST.
Second, let's pretend that the matchmaker can and does put only Tier 1 pilots in a match together.
- Half of your Tier 1 players will have below median performance in any number of given matches, so do you plan on instantly demoting them two Tiers to Tier 3?

First, an argument made on bad assumptions ... that all Tier 1 pilots are "the best" (I'm not. I should probably be in Tier 2) ... and that the match maker only puts Tier 1 players against other Tier 1 players (we all know it doesn't) ... is impossible to attack (or defend).

PSR should be re-sorted, made zero-sum, adjusted based on recent playing history (maybe the past six months), and set up to sort active players into more-or-less even "buckets".

At a minimum, the match score threshold for maintaining your PSR should be higher for Tier 1, so that those who can't keep up get downgraded to Tier 2. (And the scores required for Tier 2 should be higher than Tier 3, etc.)

Quote

That's like taking an "average" professional/olympic athlete and sending them back to play Highschool competition. Such an athlete would crush opponents at that level. Just because they aren't #1 in the race/game/score doesn't mean they are the top 1% in Skill/Talent. It just means tge competition is harder. Expecting every Tier 1 pilot to maintain top match score is impossible because they are playing against the other top pilots.

The real "top tier" players get consistently better scores than the median, win or lose. For a small sample, watch a top player in solo queue for ten matches. He will probably win more than he loses, and score better than average most of the time. Even his bad matches will probably be above the median, because the great players are also good at correcting their mistakes (most of the time ... see "12 damage EXE" for an outlying "bad game" by a great player).

Over the long term, if PSR and match maker ever get fixed, the players that are consistently better than others will rise to the top, and everyone else will fall to the appropriate level of mediocrity.

Edited by Kageru Ikazuchi, 12 August 2018 - 01:45 PM.


#475 Kin3ticX

    Member

  • PipPipPipPipPipPipPipPipPip
  • The People's Hero
  • The People
  • 2,926 posts
  • LocationSalt Mines of Puglandia

Posted 12 August 2018 - 03:38 PM

View Postjustcallme A S H, on 12 August 2018 - 01:21 AM, said:


And what information and numbers are you basing that off, exactly? I don't know why I'm bothering to explain with stats as I know how that ends, but here we go...

Lets go with what we know for now from the data scrapes we have access too.
  • ~57k players with 10 matches per season in the last 3 months. 28k was last months actual.
  • Median for Average Match Score (AMS) is 210-220 AMS.
  • 220AMS means you are, roughly, around player 16,000 out of the 57,000.
  • 190AMS / WLR 0.9 / KDR 0.5 / 10k played games is enough to get into Tier 1, even though this puts you well below the median.
With 200AMS and above it should put you in the low end of T3 if the middle if T3 is for 220AMS. That means between Tier 1 / 2 / 3 you have around 23,000 people in the pool. Between 200AMS and 190AMS there is only around 5,000 players missing and some of those won't even be in Tier 1/2/3 anyway as they simply have not played enough matches to progress. So it is, more than likely, lower than 5k. We are now talking around 17% of players.




Now it would of course have to be watched closely and actively tuned in the inital months if things were changed. It might mean 'expanding' the match maker to include Tier 4 after say 5-6mins which would allow people with 170-190AMS back into the pool. Before you (or anyone else) uses that to try dismantle my argument - they are in Tier 1-3 games right now so there is not a big difference there to add them in, now and again as opposed to always.

So there you have it - Some stats and numbers. Really keen to see where your numbers/stats and then reasoning is drawn from.

* NOTE: This is a little bit of educated guess work, of course. There will no doubt be minor errors in the numbers as only PGI knows how many people are truly in what Tier and their AMS etc etc. However I feel that I am close enough and there is certainly a potential for a workable solution.


I agree with this but I think it could be simplified down to this.

There are average and below average players making it to tier 1 that don't belong there. This means that people are getting tier 1 status, yet half of everyone in the entire game is better than them.

People forget that half of everyone scores below or above average, thats how averages work.

Maybe we can't improve matchmaking but we can correctly tier the players, that PGI could actually do.

Edited by Kin3ticX, 12 August 2018 - 03:40 PM.


#476 Callsigntal0n

    Member

  • PipPip
  • The Howl
  • The Howl
  • 28 posts

Posted 12 August 2018 - 08:42 PM

I very rarely post in forums,(main reason being there is definately a cool kids club in forums but thats beside the point) so here gors my reply to this.

Clan laser vomit... everyone is scared of the clans big scary alpha strikes... i get it. No one likes being the first kill. However what about the dual hgauss carriers? The IS lrmboats? All i have seen the last 2 to 5 patches have been nothing but nerfs against the clans and buffs for IS. In the last match i played (now over a month ago.) I got headshotted by a dual gauss fafnir. Because he MISSED his intended target and hit me at 450m away and gave me my first 0 dmg game. Many times i have been deleted by dual hgauss and deleted many myself using the weapon. Why? Because thats 50 pinpoint damage instantly hitting closer than 600m for 0 heat. Ac 20? Uac20? Clan or IS? Nope you cant dualwield those without massive heatspike. And guess what? CLANS CANT COUNTER DUAL HGAUSS. We dont have a similar weapon (though if we did it would be nerfed to obscurity) medium pulse lasers should NOT be colder than regular medium lasers (clan med pulse are colder than the regular ermeds atm) and the fact that the pulse lasers are normally the better as the ermeds are so damn hot taking more than three is suicide.

Now clans have a high damage cap. My favorite hellbringer build does 50 damage. 1 lbx20 and 3 ermeds with ecm, sounds pretty strong right? Lol you havent played clans. The Main problem is the distince way the two tech trees are handled. Clan mech are supposed to outrange and out damage IS mechs at the cost of armor. Clans favor speed and firepower. With the current build i would not can ANY clan mech, not even the IIC ones "tanky." As they all feel like they are wrapped in tissue paper even at max armor. Now youre taking away the cannon part of the term "glass cannon." Lately (and this is the reason ive stopped playing despite sinking more than 2k usd in this game on both purchases for myself and friends who no longer play either.) Is that PGI seems to hate the clans. You want people to come back? Let me put it in all caps for you. START BUFFING MECHS AND WEAPONS INSTEAD OF NERFING ALL THE M***********G TIME!!!! Seriously l, just think about it, all of the complaints leveled so far are centered around the NONSTOP NERFING. You want to stop a LASER meta? BUFF BALLISTICS. Make it Worth it to take something else. Too much has the gaming community heard from devs (in general not just pgi though pgi has the same problem) "players dont know what they want" or "players arent devs so they dont know how to balance." Lookie here. Players ARE WHAT KEEP YOUR GAME ALIVE and the community balance pass shows for a fact and beyond a doubt WE KNOW WHAT WE WANT AND IT ISNT NERFS.

In conclusion and TL:DR this is part rant and part heartfelt and very passionate pleas. STOP NERFING as a player, and long time supporter of MWO i can say beyond a doubt until i see that either A: pgi actually listens to its playerbase and starts. Buffing instead of nerfing or B: there is a pts with the community numbers ( even if its literally just to shut people up) my wallet and my game time are going elsewhere.

P.S. although i say i wont be returning i sincerely hope pgi listens to its players (not the 5%that make ups the pro circuit but the other 95%) if pgi can turn this s******w around i will gladly come back. But as of now i just cant bring myself to open that launcher

(Now the interesting part will be seeing how much gets redacted...)

#477 Y E O N N E

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • The Nimble
  • The Nimble
  • 16,810 posts

Posted 12 August 2018 - 08:55 PM

View PostYeonne Greene, on 12 August 2018 - 03:02 AM, said:

So, it was stated in the OP of this thread that there is a gulf between Clan and IS laser vomit. However, that gulf was not elaborated on, and because it seems like solutions are being proposed without analyzing the problem, I have made an initial attempt at that analysis.

You can find it in the PTS subforum at this link.

There seems to be a prevailing thought that limiting the heat capacity will reduce this Clan gulf. I submit that it will not, because Clans run too cold and have the option for more raw DPS as well as sustained DPS, as you will see if you take the time to read through the analysis.


Also, if it's not too much to ask, I would appreciate hearing some of PGI's thoughts on this.

#478 Ninjah

    Member

  • PipPipPipPipPipPip
  • Shredder
  • Shredder
  • 307 posts
  • LocationComstar Lounge

Posted 13 August 2018 - 03:01 AM

My 2 cents on the balance and mechs:

1. LRMs are not that much of an issue, Narc is. I have maxed out Narcers for both IS and Clan and I play them a lot. As someone said some mechs can take 2 narcs and have 12 enemy mechs narced in a jiffy. I suggest drastic cut on the Narc duration instead of cooldown, maybe even ammo reduction per ton.

2. Fix Clan UACs and I'll play a lot more dakka (instead of lazor vomit) - reduce base jam time and maybe buff velocity a little bit. Current jam time is a death sentance for many dakka builds. For example I love my Uac20, Uac10 Ebon with Uac20 on the roof for poking behind cover. One Uac20 jam causes so much problems cause my damage is reduced by 2 thirds and that 1 third is at risk of jamming too. When it works it's awesome but it depends too much on luck so I avoid it. Maybe make JAM happen more often on builds with 4+ UACs (they have backup). You could try to to reduce Jam time on Uac20s and 10s, Uac5 and Uac2 seem fine.

3. Buff small and small pulse lazors a bit, Cheetah is pretty useless, Jenners are collecting dust while Pirahnas are having a bloodbath party. Nerf the fish. Add heat generation for MGs, slowly and constantly building heat while firing with ghost heat for more than 6. Heat buildup should be Stealh Armor style.

4. Mobility buffs - great news! A lot of times I get cored cause torso twist speed is so low I only manage to twist away 20% of incoming damage. Defending from lights will be possible at last. Woot!

5. IS PPC family just sucks, needs damage buff, not kidding. No one I know uses IS PPCs.

6. Clan lazor damage... I'd say Clan lazors need a minimal nerf to damage. Simple as that. Only mediums and heavier thought, small lazors need a buff.

7. Default SRM spread is too big. Hit detection is horrible. I've seen 36 srm tubes do 0 (potato) damage too many times. Something similar is happening with MRMs too. Avoiding SRMs because of hitreg issues.

8. Stop jerking balance once you get out of this mess. There are so many simple fixes that will make the game a lot better... Clean up hitboxes on Omega gun for instance. Easy but means so much, nothing breaks immersion better then thin air stopping lazors. Might update Gun/Gen/Turret/Domination tower models just for fun, they're so ugly right now. Unify shaders on all mechs, my IS Jenner looks like a wet rag instead of a metallic robot. A number of maps is missing destructable objects. Setting trees on fire? You tested it before.. what happened? We need better placement of walls in FP spawn. People sometimes need a lot of time to find the exit, paint some directional arrows on the walls. Etc. Etc. Etc. Can't type anymore, will update if remember something importmant I missed.

Thanks for reading ;)

#479 Khobai

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPipPip
  • Elite Founder
  • Elite Founder
  • 23,969 posts

Posted 13 August 2018 - 06:18 AM

View PostYeonne Greene, on 12 August 2018 - 08:55 PM, said:

There seems to be a prevailing thought that limiting the heat capacity will reduce this Clan gulf. I submit that it will not, because Clans run too cold and have the option for more raw DPS as well as sustained DPS, as you will see if you take the time to read through the analysis.


the best solution IMO is to buff IS DHS. all IS DHS should be 2.0 because they take up 3 crit slots instead of 2 crit slots.

putting IS DHS at 2.0, in conjunction with many IS mechs having -10% heat gen quirks, would largely bring IS and Clan mechs into parity with eachother for heat capacity/dissipation.

the two tech bases need to be equalized. heatsinks are an obvious place to start. but also ISXL needs to survive side torso blowout. and IS ES/FF needs to be more on par with the clan versions.

Edited by Khobai, 13 August 2018 - 06:19 AM.


#480 Josh Seles

    Member

  • PipPipPipPipPip
  • 164 posts

Posted 13 August 2018 - 08:02 AM

View PostKhobai, on 13 August 2018 - 06:18 AM, said:


the best solution IMO is to buff IS DHS. all IS DHS should be 2.0 because they take up 3 crit slots instead of 2 crit slots.

putting IS DHS at 2.0, in conjunction with many IS mechs having -10% heat gen quirks, would largely bring IS and Clan mechs into parity with eachother for heat capacity/dissipation.

the two tech bases need to be equalized. heatsinks are an obvious place to start. but also ISXL needs to survive side torso blowout. and IS ES/FF needs to be more on par with the clan versions.


I mostly agree with this, except with just sticking IS DHS at 2 dissipation. I'd set IS DHS set at 1.9 and a 3/4 investment in ops skill tree gives 2 dissipation. Same for Clan DHS, except they top out at 1.85 or 1.9 because we can mount so many more C-DHS than IS ones. It's like early 2017 when IS DHS were slightly better than the Clan ones due to bulk.

As for buffing IS Ferro, Endo, and XL Engines, I'd say yes please. Reduce slot requirements for IS Ferro and Endo to 10 or 12 slots each instead of the 14 slots that they are now. If it were ever implemented, Endo-Composite structure would then be 5 or 6 slots instead of 7. As for IS XL, I'm not sure, but any buff would be welcome. I used to be against getting rid side torso loss deaths with IS XL, but alas, no longer.

I'm just sticking to IS Ferro, Endo, DHS and XL for now. I'd apply slot reductions to some IS weapons (LBX-20 and AC-10), and I have ideas for Gauss (all of them) but I'm saving that for a different post since it is relevant in this thread.





9 user(s) are reading this topic

0 members, 9 guests, 0 anonymous users