Jump to content

Psr And Group Queue


60 replies to this topic

#41 Jettrik Ryflix

    Member

  • PipPipPipPipPip
  • The Star
  • The Star
  • 183 posts
  • LocationCanada, eh?

Posted 26 September 2015 - 06:50 AM

View PostYUyahoo, on 21 August 2015 - 10:28 PM, said:


I do get the analogy, really, and yes bad teamwork does factor into losses just as much as good teamwork factors into wins. this is just my opinion from the observations of many "bad beats" in the old ELO system and the current PSR system. I do not by any means believe I am assigned to the "all star" team in your analogy, quite the opposite I am pretty certain I am on the "second string" team almost every time (as I am basically a "second string" skill level player and I do not think my skill level is any higher than the other 11 people I randomly get assigned to every game). The problem lies in that the new PSR system (like the old ELO system) seems to think that the "second string" team I land on would be a fair fight for the "all star" team I tend to be matched up against when in reality it is not a fair fight far more often than it is. Perhaps the bigger issue is something like that I am on the cusp between tiers, so whenever I do get a good team/good game I get "bumped up" a tier and then have bad match after bad match until I land back in the lower tier and have another good team/good game that lands me back in the higher tier where I go back to getting stomped game after game. I don't expect that I would or should win most of the time (specially in solo/pug games) but if these match making systems are working "properly" my win/win loss ratio should be very close to 50/50 (say in the 45-55% range) for the 4000ish games I played in the old ELO system and just as close to 50% in all the games I have played in the PSR system (if for no other reason than 50% of the time I should be on the "all star" if the ELO/PSR system really does consider the "all star" teams to be a fair fight for me) but in reality my winning/good game/good team percentage seems to be closer to 30% than 50%. This creates a lot of frustration as games are less fun when most end 12-4 or less rather than 12-9/10/11 (and again I don't mind if I am on the loosing end of close games or if I have an occasional "bad beat" game) and it cascades into more time needed to level mechs (because exp rewards are 90-300 rather than 1000-4000) and lower cbill earnings/more time needed to purchase mechs, modules and other equipment/upgrades. It is hard to improve if most of the time the outcome of a game is pre-determined before the game even starts because "second stringers" keep going up against "all stars".

It may sound like it but this isn't a "whine/cry" post or a "PGI sucks" post but that is not my position at all...despite how this may read to most of you I actually do enjoy MWO and think PGI has made vast improvements over the past few months (and I have invested $100s in the game as I am a fan/supporter)...but there are still some issues and some things that (at least from my experiences) still aren't quite working as well as people/PGI may believe.


The cause of "stomps" in MWO is usually linked to the scoreboard, and how players react when they know they're ahead or behind.

In my experience, playing aggressively always beats playing defensively, or "hiding" to save your own 'mech. So, when team A gets the first 2 kills, suddenly everyone on Team A "tastes blood" and start looking for more kills. Players on team B, on the other hand, feel weak and outnumbered, causing a timid (and losing) play-style.

This, given the right variables, is what usually leads to stomps. There is nothing any matchmaker could do to remove this fact. Stomps are inherent to MWO 12 v 12.

#42 Ens

    Member

  • PipPipPipPipPipPipPipPip
  • Mercenary
  • Mercenary
  • 1,088 posts
  • LocationGermany

Posted 28 September 2015 - 12:08 AM

Dunno if it´s due to the new PSR or Tier system but:

Abnormal wait times up to 10 minutes, which enable you to just do like 3 games in one hour, plus seeing 90% of the same players match after match ( or the same 8 to 10 man groups ) is poisoning our experience :mellow:

where´s my "push me to tier 4" instant-button? :ph34r:

#43 Anachronda

    Member

  • PipPipPipPipPipPip
  • Bridesmaid
  • 293 posts

Posted 28 September 2015 - 12:50 AM

This has been up awhile, but you guys may want to check out Russ Bullock's post here. Basically he points to the mathematical fact that the more restrictions you place on the queues the longer wait times will be. There's another thread by Paul Inouye, linked there, which is for discussion of his suggestions. To be honest, I don't think I much like the possibilities he gives for change and that also seems to be the consensus in the thread. I would think that the expansion of our player base is going to be our best bet for decreasing wait times.

Edited by Anachronda, 28 September 2015 - 12:51 AM.


#44 White Bear 84

    Member

  • PipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 3,857 posts

Posted 28 September 2015 - 01:08 AM

View PostJettrik Ryflix, on 26 September 2015 - 06:50 AM, said:

The cause of "stomps" in MWO is usually linked to the scoreboard, and how players react when they know they're ahead or behind.

In my experience, playing aggressively always beats playing defensively, or "hiding" to save your own 'mech. So, when team A gets the first 2 kills, suddenly everyone on Team A "tastes blood" and start looking for more kills. Players on team B, on the other hand, feel weak and outnumbered, causing a timid (and losing) play-style.

This, given the right variables, is what usually leads to stomps. There is nothing any matchmaker could do to remove this fact. Stomps are inherent to MWO 12 v 12.


You know what also leads to stomps.. ..when the whole team bundles up on one spot and allows themselves to get surrounded by enemy mechs.

People never learn..

#45 50 50

    Member

  • PipPipPipPipPipPipPipPipPip
  • 2,145 posts
  • LocationTo Nova or not to Nova. That is the question.

Posted 28 September 2015 - 03:48 AM

Well of course more restrictions on the queues will increase wait times.
If we look at all the variables currently in place we have:
Solo or Group
Server selection
Game mode selection
3/3/3/3 matching (as best it can but often this is not what you see anyway)
And PSR.... which also does not strictly match.
Maybe more.

Removing some of these factors and changing some of the options you would think would not only improve the match making time but hopefully also improve average quality of games for players.
Personally I would suggest:
1. Combine solo and group queue. As it is, a group who doesn't communicate is no better off than solos. But this should help fill the gaps.
2. Remove the option to select game mode. If we had a preparation screen to select a different mech once the match had been made, like we do for CW, that would be nice. Otherwise, it is just an unnecessary complication.
3. Change from the 3/3/3/3 limit to a team/lance tonnage minimum/maximum. This would bring the standard queue more in line with CW drop decks but would also be very close to private match limits.... don't understand why we aren't using the existing feature.
4. I think we need to keep server selection for the reason that disconnects and lag where you have a bad ping does not make for a good experience.
5. Tighten up PSR matching and add in a selection option for tier.
This is an interesting one. If you could select to play in your tier or higher (not lower) it puts the option back on the player about how difficult they want to make their match. A greater tonnage limit, more XP and c-bills could be used as an incentive players to select a higher tier to battle in and hopefully keep things rolling along. Showing tie populations would likely be needed.

That's my thoughts on it.

#46 Dran Dragore

    Member

  • PipPipPipPipPip
  • Force Commander
  • Force Commander
  • 151 posts

Posted 28 September 2015 - 10:23 AM

View PostYUyahoo, on 21 August 2015 - 10:28 PM, said:


I do get the analogy, really, and yes bad teamwork does factor into losses just as much as good teamwork factors into wins. this is just my opinion from the observations of many "bad beats" in the old ELO system and the current PSR system. I do not by any means believe I am assigned to the "all star" team in your analogy, quite the opposite I am pretty certain I am on the "second string" team almost every time (as I am basically a "second string" skill level player and I do not think my skill level is any higher than the other 11 people I randomly get assigned to every game). The problem lies in that the new PSR system (like the old ELO system) seems to think that the "second string" team I land on would be a fair fight for the "all star" team I tend to be matched up against when in reality it is not a fair fight far more often than it is. Perhaps the bigger issue is something like that I am on the cusp between tiers, so whenever I do get a good team/good game I get "bumped up" a tier and then have bad match after bad match until I land back in the lower tier and have another good team/good game that lands me back in the higher tier where I go back to getting stomped game after game. I don't expect that I would or should win most of the time (specially in solo/pug games) but if these match making systems are working "properly" my win/win loss ratio should be very close to 50/50 (say in the 45-55% range) for the 4000ish games I played in the old ELO system and just as close to 50% in all the games I have played in the PSR system (if for no other reason than 50% of the time I should be on the "all star" if the ELO/PSR system really does consider the "all star" teams to be a fair fight for me) but in reality my winning/good game/good team percentage seems to be closer to 30% than 50%. This creates a lot of frustration as games are less fun when most end 12-4 or less rather than 12-9/10/11 (and again I don't mind if I am on the loosing end of close games or if I have an occasional "bad beat" game) and it cascades into more time needed to level mechs (because exp rewards are 90-300 rather than 1000-4000) and lower cbill earnings/more time needed to purchase mechs, modules and other equipment/upgrades. It is hard to improve if most of the time the outcome of a game is pre-determined before the game even starts because "second stringers" keep going up against "all stars".

It may sound like it but this isn't a "whine/cry" post or a "PGI sucks" post but that is not my position at all...despite how this may read to most of you I actually do enjoy MWO and think PGI has made vast improvements over the past few months (and I have invested $100s in the game as I am a fan/supporter)...but there are still some issues and some things that (at least from my experiences) still aren't quite working as well as people/PGI may believe.


thats right. PSR realy kills in my opinion the fun in the game. Its just frustrating, sry to say that. Where is the "turn off PSR" Button in the Options? :-D

#47 V O L T R O N

    Member

  • PipPipPipPipPipPip
  • The People's Hero
  • The People
  • 318 posts
  • LocationThe Flat and Motionless Earth

Posted 03 October 2015 - 12:05 PM

View PostOnimusha shin, on 22 August 2015 - 06:17 PM, said:

honestly, i think group queue doesn't have the prerequisite population to make for faster wait times. don't forget that in addition to PSR, it also has the 3/3/3/3 valve to manage.

I think taking away 3/3/3/3 and making it weight will balance things out more

#48 Enda Wyld

    Rookie

  • Survivor
  • 1 posts

Posted 03 October 2015 - 01:40 PM

View Post50 50, on 28 September 2015 - 03:48 AM, said:

3. Change from the 3/3/3/3 limit to a team/lance tonnage minimum/maximum. This would bring the standard queue more in line with CW drop decks but would also be very close to private match limits.... don't understand why we aren't using the existing feature.

I'd have to agree. Far to many times I've entered matches whereby one team is skewed toward lighter mechs while the other is skewed toward the assaults. Around 90% of the time you can tell which side will win just by the list of mechs during drop preparation. (The Heavier Team will almost always slaughter the Lighter Team.)

View Post50 50, on 28 September 2015 - 03:48 AM, said:

5. Tighten up PSR matching and add in a selection option for tier.
This is an interesting one. If you could select to play in your tier or higher (not lower) it puts the option back on the player about how difficult they want to make their match. A greater tonnage limit, more XP and c-bills could be used as an incentive players to select a higher tier to battle in and hopefully keep things rolling along. Showing tie populations would likely be needed.

I created an alternate account to test this. My main account is so low, I only have one or two hash marks in Tier 5, and there is little to nothing I can do to raise it more than a hash mark or two before getting knocked back down. The alternate account STARTED with about 80% of the Tier 4 bar filled. My main account takes forever to get into a match, but the alternate account gets matches within seconds. The annoying part is that the mechs on my main account are far better than the one on the alternate account, but my alternate account wins more often.

#49 c0de4014

    Rookie

  • The God
  • The God
  • 6 posts

Posted 29 February 2016 - 12:41 AM

Pilot skill rating is not an effective stat on judging an individuals skill. i have been doing a ton of research on how the skill actually increases and it is ONLY based on if your TEAM wins.... how is this a good judge of an individual skill? I have to rely on my entire team (usually full of meat heads) to increase My skill level? that is absurd... here is a link to a picture of a spreadsheet that shows in detail the ridiculousness of this stat...

https://gyazo.com/19...912ed91107d05c0

#50 c0de4014

    Rookie

  • The God
  • The God
  • 6 posts

Posted 29 February 2016 - 12:45 AM

If i do terrible but my team wins my stat doesn't go down... if i do great and my team loss my skill goes down or doesnt move.... how does this make sense on an individual skill rating?!?!

#51 Shirow

    Member

  • PipPip
  • Knight Errant
  • Knight Errant
  • 45 posts
  • LocationTexas :)

Posted 29 February 2016 - 04:08 PM

While we are waiting for game to start would love a couple of mins to change my mech

#52 Extra Guac

    Member

  • PipPipPipPipPipPip
  • Star Colonel
  • Star Colonel
  • 202 posts

Posted 28 January 2017 - 09:14 AM

Hello Developers,

I have been having an issue which is that I frequently get dropped from the game at the start of a match (while it is launching).

When this happens, I can re-join it via the quickplay button. And once I get back in, everything is fine. There are no problems with my connection at all.

The problem is that if I am in a 12-man faction play group, I have to leave the group in order to rejoin the match, because the quickplay button disappears when you're in a group.

Could you please add a new way to re-join a match, without having to leave the group? How about a "re-join match" button, which is separate from the quickplay button?

#53 BTGbullseye

    Member

  • PipPipPipPipPipPipPipPip
  • The Solitary
  • The Solitary
  • 1,540 posts
  • LocationI'm still pissed about ATMs having a minimum range.

Posted 24 July 2017 - 05:04 AM

View Postc0de4014, on 29 February 2016 - 12:45 AM, said:

If i do terrible but my team wins my stat doesn't go down... if i do great and my team loss my skill goes down or doesnt move.... how does this make sense on an individual skill rating?!?!

More than a year, and it's still this horribly un-individualized... I get that teamwork is extremely important, but having to get 250 match points on a loss just to break even is a little much. (especially when a win only requires 50)

Edited by BTGbullseye, 24 July 2017 - 05:06 AM.


#54 MW Waldorf Statler

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 9,459 posts
  • LocationGermany/Berlin

Posted 26 July 2017 - 05:34 PM

View PostDeepfryer, on 28 January 2017 - 09:14 AM, said:

Hello Developers,

I have been having an issue which is that I frequently get dropped from the game at the start of a match (while it is launching).

When this happens, I can re-join it via the quickplay button. And once I get back in, everything is fine. There are no problems with my connection at all.

The problem is that if I am in a 12-man faction play group, I have to leave the group in order to rejoin the match, because the quickplay button disappears when you're in a group.

Could you please add a new way to re-join a match, without having to leave the group? How about a "re-join match" button, which is separate from the quickplay button?

once we have it ...is gone like many features

#55 Imperialus

    Member

  • PipPip
  • Knight Errant
  • Knight Errant
  • 37 posts

Posted 02 November 2017 - 12:29 PM

View PostSilentScreamer, on 21 August 2015 - 04:28 PM, said:

Not to step on Paul's toes to answer you YUyahoo, but matching Tier pilots might work well on a team or they could end up getting stomped 0-12 because they refuse to communicate and coordinate. My apologies for throwing a sports analogy in but....Think of it like a All-Star Basketball team. You have all the big name NBA Stars on your team. If the TEAM is not willing to work together, they will loose to "second string" players that are capable of working as a team. Generally, the more sucessful a player is, the less they are willing to sacrifice or compromise for the team. A common term is "ego-case." I am just guessing, but I would say Bad teamwork results in more stomps than the matchmaker skill ratings.


I notice that quite a bit too and I think the analogy holds. It's kinda like when they first opened up the Olympics to NHL players in 98. Canada got stomped and didn't take home a medal, even though our lineup was a fantasy hockey fan's wet dream with Gretzky, Yzerman, Lindros, Nieuwendyk ect.

I've ended up with a lot of games where I'll see unit tags and faction loyalty badges splattered all over one side of the match, and oftentimes that's the team that gets worked over by the team with 2 or 3 unit tags.

#56 pZiuz

    Rookie

  • Little Helper
  • 8 posts
  • LocationFunland of Finland

Posted 15 August 2019 - 11:33 AM

Hey!
Fix it!

#57 Tarl Cabot

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Tai-sho
  • Tai-sho
  • 7,776 posts
  • LocationImperial City, Luthien - Draconis Combine

Posted 15 August 2019 - 05:48 PM

Once PSR has opened up, it really is not in play. It is more about the group sizes and population of the actual group queue at the time a team queues up for it.

View PostTarl Cabot, on 15 August 2019 - 05:06 PM, said:

See below, include time of day also, and how long they are willing to wait for a drop. And once players have been bitten by the not-getting-a-groupdrop-today bug, fewer players group up. And obviously there aren't 12x 2-man attempting to drop at the same time. Nor is there anything to indicate if anyone is actually queue'd up in the group queue.

Population and tons of different and separate queues. And group queue does not have any solo pugs to "fill" in the spots.

Group size requirements and this is for BOTH sides..... And PSR is out the door after 30secs.

12-man makes up one side, then only need the other 12-man to be matched up, or a combination of the following....
10-man requires a 2-man
9-man requires 3-man
8-man requires 4-man or 2x2-man
7-man requires 5-man or 3+2-man
6-man requires 6-man, 4+2-man or 3x2-man or 2x3-man
5-man requires 7-man, 5+2-man, 3-man+2x2-man
4-man requires 8-man, or 6+2-man, or 5+3-man or 2x4-man or 4x2-man
3-man requires 9-man, or 7+2-man, or 6+3-man, or 5+4-man, or 5+2x2-man, or 3+3x2-man
2-man...etc...

And the above has to be done for BOTH sides, which is the reason many suggest dropping in a larger group, to be less of a gap filler. Of course, it would also mean that there are not 12 2-man dropping at the same time, there is no actual incentive, and once it is attempted a few times one who is less enthusiasm about MWO then stops playing. A vicious cycle.

Would allowing solo pugs (max 3 per side) and/or 2-man to opt into the other queue without opt out of their default queue help? Go with the solo pugs into group first, may allow for fillers but until there is something else in place to actually encourage players forming up teams for other than FP, I do not see it changing too much. Maybe reduce the max group size for the teams down to 8-man.

/shrugs....


#58 Aivazovsky

    Member

  • PipPipPipPipPipPipPip
  • The Dragoon
  • The Dragoon
  • 801 posts

Posted 21 August 2019 - 05:42 AM

View PostShirow, on 29 February 2016 - 04:08 PM, said:

While we are waiting for game to start would love a couple of mins to change my mech

this really beautiful decision was voiced, as it seems to me, more than once. Three years have passed since then ... but alas ... how much time is lost, how much money is lost?

#59 Dark Fenrir the Fluffy

    Rookie

  • The Territorial
  • The Territorial
  • 8 posts

Posted 24 September 2024 - 01:16 PM

The Elephant in the room is the ability to retain a player base with a broken Tier system. Tier one and two players shouldn't be anywhere near quick dropping with tiers 3-5. Tier 4 an 3 people are literally just chum for top players and you developers know it. You make money selling hope that the next new mech might give lacking or casual players a chance in hell to win. Meanwhile, the lifelong top tier dudes are paying you far less because they can win/buy most anything they need. People quit this game all the time after hours and hours of bad games where they had zero chance to even get a decent score. Ive brought three new players in over the years and all of them love the universe love the game in general, but feel between wait time and matches they are rolled over constantly, the game just isn't fun. I spend a lot on this game and enjoy building mechs and reading lore and watching experienced players play,..WAY more than actually playing. Top players will bash this because they DON'T WANT IT TO CHANGE because they have their unit and have the opportunity to beat on the new guys still learning and earn a tone of money and prizes to boot. Games with players at or around my play level are fun, educational, challenging and just cool in general, win or lose. 12-0 stomps because a Unit times their drops and is on private coms is too big an advantage.
solution: smaller teams, no multiple unit members in a single quick drop allowed, a specific type of quick drop for teams only (not faction play the number of unit members it takes to make it happen is difficult) 8v8 for teams with no more than four from the same unit. Most of all, a better tier system that has tiers playing ONLY their equals in skill. You cant say you have plenty of players and then say you don't have enough to do that. Why is that not possible?

#60 Dark Fenrir the Fluffy

    Rookie

  • The Territorial
  • The Territorial
  • 8 posts

Posted 24 September 2024 - 01:27 PM

View PostImperialus, on 02 November 2017 - 12:29 PM, said:


I notice that quite a bit too and I think the analogy holds. It's kinda like when they first opened up the Olympics to NHL players in 98. Canada got stomped and didn't take home a medal, even though our lineup was a fantasy hockey fan's wet dream with Gretzky, Yzerman, Lindros, Nieuwendyk ect.

I've ended up with a lot of games where I'll see unit tags and faction loyalty badges splattered all over one side of the match, and oftentimes that's the team that gets worked over by the team with 2 or 3 unit tags.


TRUE but there is no fun way to teach new players how to work as a team or how much benefit there is to do so. Nothing in training grounds for that. Making new players go through a tutorial with other AI team members holding their hand would help although I realized programming that might be a nightmare. Or perhaps specific pilot rating analysis and a code numbers to see training tutorial videos that match those numbers. (I.E. watch Code 1 and Code 5 videos for bonus cbills from that match.) Code 1 is about firing lines and five is about overheating and build vs play style and mech design. I know that seems complicated but lets figure it out.





1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users