Jump to content

Combining Group And Solo Queues - 4 Week Test


1579 replies to this topic

#1341 Excessive Paranoia

    Member

  • PipPipPipPipPipPip
  • 216 posts

Posted 16 May 2020 - 02:15 PM

View PostBlackhorse11Cav, on 16 May 2020 - 02:01 PM, said:

Yeah, Really? The likelihood of what you describe happening on a "regular" basis is nowhere near any of my experience. My experiences in solo vs groups is more like what you see here, than what you describe.

Posted Image


I can't tell if you're trying to be sarcastic or just misunderstanding what Chaotic_Harmony said, but the image you've posted is literally what they described... 4 people on one team doing more damage than the entirety of the other team, with a couple of people only able to get 1 shot off. If you're trying to claim that the opposing group was useless, that may be true, but on the other hand, you cannot claim that the group on your team was not the unbalancing factor unless you can say for certain that the 4 top damage dealers in the match were not in a group. This, btw, is part of the problem with this whole argument... Thanks to the fact that we don't have a clear indication which players are in a group, the only way to tell is if you've got multiple people with the same unit tag.

Edited by Excessive Paranoia, 16 May 2020 - 02:18 PM.


#1342 Blackhorse11Cav

    Member

  • PipPip
  • Philanthropist
  • Philanthropist
  • 24 posts

Posted 16 May 2020 - 02:57 PM

Here are at least two groups losing to a team that appears to be a bunch of solo players. And since there is no way of knowing who is in a group other than by seeing clan tags, your argument that groups are dominating is shaky to begin with.


Posted Image

Edited by Blackhorse11Cav, 16 May 2020 - 03:07 PM.


#1343 Excessive Paranoia

    Member

  • PipPipPipPipPipPip
  • 216 posts

Posted 16 May 2020 - 03:33 PM

View PostBlackhorse11Cav, on 16 May 2020 - 02:57 PM, said:

Here are at least two groups losing to a team that appears to be a bunch of solo players. And since there is no way of knowing who is in a group other than by seeing clan tags, your argument that groups are dominating is shaky to begin with.


Posted Image


Just as shaky as your argument that they are not... Again, without an ability to see if any of the "solo" players are in a group, it is difficult to present this as evidence one way or another. That said, there is a greater likelihood that I'm still correct here as there will most likely always be at least one group on both sides of a match if there are any groups at all in a match. This assumption is based on the information given to us by PGI where they stated that the MM would attempt to get as close to 4 group members and 8 solo players on each side as possible, even allowing for more than 4 total group members (i.e. if there is a 4-man, a 2-man, and a 3-man available to split between the teams). If we're to accept that statement as fact, then the probability that any match will only have a group on one side is quite low, as that would have to mean that there was only one group in the entire queue for the MM to work with. Unfortunately, that's still just an educated guess based on the information we've been given and there is no way to be sure based on the information given by these images.

On a side note, I'll point out that the group I used to drop with in 2016-2017 were all in different units, so if we were to drop together now, we would not outwardly look like a group, despite the fact that we very certainly were. I'd also go so far as to say this was not an uncommon thing to see, so I can't imagine that situation has changed all that much in the intervening years.

Edited by Excessive Paranoia, 16 May 2020 - 03:38 PM.


#1344 Constalation

    Member

  • Pip
  • Bridesmaid
  • Bridesmaid
  • 17 posts

Posted 16 May 2020 - 03:37 PM

View PostBlackhorse11Cav, on 16 May 2020 - 02:57 PM, said:

Here are at least two groups losing to a team that appears to be a bunch of solo players. And since there is no way of knowing who is in a group other than by seeing clan tags, your argument that groups are dominating is shaky to begin with.



If we go by some basic understanding of team games, we can assume that groups would generally find higher WR% anyway. I can take a phrase my clanmates back in WoT\WoWs: "Having less unpredictable players on your team is a direct advantage." And it is true! Being in a group of 4, means you have 3 players who you can guarantee are going to work together in some form or another. That is a total 4 players on your team that have a 0% chance of being, for lack of better word; complete nobheads.

To give an example from one of my matches, I was playing the PiranhaCI and my friend the FleaFA. I didn't notice at first, but he got ambushed by a Spider with ECM. Now, if this was any other player and he didn't use voice comms in-game (my friend forgets to use voice comms in-game) he would be dead. But since I was nearby and we were acting as wingmen I came in and blasted the spider back to the mechbay. That would rarely if ever happen in a solo-game.

Edit: Here are the only two screenshots I have taken of end-game screens. Naturally, these are cherry picked and I only screenshot memorable\really good score but here you go. (I'll try to grab other screenshots next time I play)

Posted Image

Posted Image

Edited by Constalation, 16 May 2020 - 03:46 PM.


#1345 DAEDALOS513

    Member

  • PipPipPipPipPipPipPipPipPip
  • The Flame
  • The Flame
  • 2,633 posts
  • LocationArea 52

Posted 16 May 2020 - 04:08 PM

View PostExcessive Paranoia, on 16 May 2020 - 03:33 PM, said:


Just as shaky as your argument that they are not... Again, without an ability to see if any of the "solo" players are in a group, it is difficult to present this as evidence one way or another. That said, there is a greater likelihood that I'm still correct here as there will most likely always be at least one group on both sides of a match if there are any groups at all in a match. This assumption is based on the information given to us by PGI where they stated that the MM would attempt to get as close to 4 group members and 8 solo players on each side as possible, even allowing for more than 4 total group members (i.e. if there is a 4-man, a 2-man, and a 3-man available to split between the teams). If we're to accept that statement as fact, then the probability that any match will only have a group on one side is quite low, as that would have to mean that there was only one group in the entire queue for the MM to work with. Unfortunately, that's still just an educated guess based on the information we've been given and there is no way to be sure based on the information given by these images.

Cmon.. that's the point he was making when he posted the ss.. it's all speculation..

#1346 Excessive Paranoia

    Member

  • PipPipPipPipPipPip
  • 216 posts

Posted 16 May 2020 - 04:20 PM

View PostDAEDALOS513, on 16 May 2020 - 04:08 PM, said:

Cmon.. that's the point he was making when he posted the ss.. it's all speculation..


Except its not... PGI themselves have said what I posted there, I'm not inferring anything beyond what they have said in this thread:

View PostPaul Inouye, on 27 April 2020 - 10:23 AM, said:

How the combined public queue will work:
- All matches will remain 12v12.
- Max group size will be restricted to 4 players. All group sizes of 2,3 and 4 are valid group sizes.
- When the Match Maker builds a 12-player team, it will ensure that there is no more than one 4 player group or as close as possible (2 player + 2 player for example) per team and all remaining slots will be filled with solo players.


The matchmaker sets 1-2 groups up on each side of the match and then fills in the rest with solo players. Again, this means that the likelihood that any match will have only one group is low, as groups are given first priority in the MM and solos are considered "filler".

#1347 Blackhorse11Cav

    Member

  • PipPip
  • Philanthropist
  • Philanthropist
  • 24 posts

Posted 16 May 2020 - 04:35 PM

View PostConstalation, on 16 May 2020 - 03:37 PM, said:


If we go by some basic understanding of team games, we can assume that groups would generally find higher WR% anyway. I can take a phrase my clanmates back in WoT\WoWs: "Having less unpredictable players on your team is a direct advantage." And it is true! Being in a group of 4, means you have 3 players who you can guarantee are going to work together in some form or another. That is a total 4 players on your team that have a 0% chance of being, for lack of better word; complete nobheads.

To give an example from one of my matches, I was playing the PiranhaCI and my friend the FleaFA. I didn't notice at first, but he got ambushed by a Spider with ECM. Now, if this was any other player and he didn't use voice comms in-game (my friend forgets to use voice comms in-game) he would be dead. But since I was nearby and we were acting as wingmen I came in and blasted the spider back to the mechbay. That would rarely if ever happen in a solo-game.

Edit: Here are the only two screenshots I have taken of end-game screens. Naturally, these are cherry picked and I only screenshot memorable\really good score but here you go. (I'll try to grab other screenshots next time I play)



i don't know much about Jarl's site, but I have learned it tracks stats. Yours, are *interesting*

Posted Image

#1348 Excessive Paranoia

    Member

  • PipPipPipPipPipPip
  • 216 posts

Posted 16 May 2020 - 04:41 PM

View PostBlackhorse11Cav, on 16 May 2020 - 04:35 PM, said:


i don't know much about Jarl's site, but I have learned it tracks stats. Yours, are *interesting*

Posted Image


Not sure what this is supposed to show unless you're going for some kind of ad hominem attack on the quantity of games they've played?

EDIT: Unless you're trying to insinuate that they've not been playing during the test in which case, I can factually refute that as they have 107 matches played in the current leaderboard this month.

Edited by Excessive Paranoia, 16 May 2020 - 05:07 PM.


#1349 DAEDALOS513

    Member

  • PipPipPipPipPipPipPipPipPip
  • The Flame
  • The Flame
  • 2,633 posts
  • LocationArea 52

Posted 16 May 2020 - 05:08 PM

View PostExcessive Paranoia, on 16 May 2020 - 04:20 PM, said:

The matchmaker sets 1-2 groups up on each side of the match and then fills in the rest with solo players. Again, this means that the likelihood that any match will have only one group is low, as groups are given first priority in the MM and solos are considered "filler".

PGI's post doesn't state the matchmaker will only launch when there are groups queued up.. only that it will attempt to sort them if they are there. I've seen many matches where it was impossible to tell if there were groups due to lack of clan tags. His point stands..

#1350 Excessive Paranoia

    Member

  • PipPipPipPipPipPip
  • 216 posts

Posted 16 May 2020 - 05:18 PM

View PostDAEDALOS513, on 16 May 2020 - 05:08 PM, said:

PGI's post doesn't state the matchmaker will only launch when there are groups queued up.. only that it will attempt to sort them if they are there. I've seen many matches where it was impossible to tell if there were groups due to lack of clan tags. His point stands..


An inability to identify a group does not prove its absence, my point still stands as well. I still stand by my assertion that the likelihood of any match having a group on one side and not the other is incredibly low. Its not impossible, but that would mean that out of all of the people playing, there is only one available group to place into any given match. That said, it is entirely possible that there are matches without any groups as there are still probably many more solo players than group players at any given time of the day. Ultimately though, if you see one group, you're likely seeing two and just don't know it.

#1351 Constalation

    Member

  • Pip
  • Bridesmaid
  • Bridesmaid
  • 17 posts

Posted 16 May 2020 - 05:23 PM

View PostBlackhorse11Cav, on 16 May 2020 - 04:35 PM, said:


i don't know much about Jarl's site, but I have learned it tracks stats. Yours, are *interesting*



If you read the bottom of the page. It says Database last modified: 01 May 2020 meaning the page does not include my most recent games. I assume you can read dates and how the latest data on that graph shows August 2019?

#1352 Blackhorse11Cav

    Member

  • PipPip
  • Philanthropist
  • Philanthropist
  • 24 posts

Posted 16 May 2020 - 05:39 PM

View PostExcessive Paranoia, on 16 May 2020 - 04:41 PM, said:


Not sure what this is supposed to show unless you're going for some kind of ad hominem attack on the quantity of games they've played?

EDIT: Unless you're trying to insinuate that they've not been playing during the test in which case, I can factually refute that as they have 107 matches played in the current leaderboard this month.


perhaps....What it does prove is that Jarls's website ain't all its cracked up to be. if his stats are as FUBAR as that, maybe everyone else's are as well and maybe we can't use that particular website to prove/disprove ANYTHING.

#1353 Blackhorse11Cav

    Member

  • PipPip
  • Philanthropist
  • Philanthropist
  • 24 posts

Posted 16 May 2020 - 05:45 PM

View PostConstalation, on 16 May 2020 - 05:23 PM, said:

If you read the bottom of the page. It says Database last modified: 01 May 2020 meaning the page does not include my most recent games. I assume you can read dates and how the latest data on that graph shows August 2019?


Are you new to this discussion?

#1354 Constalation

    Member

  • Pip
  • Bridesmaid
  • Bridesmaid
  • 17 posts

Posted 16 May 2020 - 05:51 PM

View PostBlackhorse11Cav, on 16 May 2020 - 05:45 PM, said:


Are you new to this discussion?


What? The event started Tuesday, April 28/2020. I didn't play in April. Page updated on May01 and hasn't since. Therefore it won't show data on my performance.

#1355 Blackhorse11Cav

    Member

  • PipPip
  • Philanthropist
  • Philanthropist
  • 24 posts

Posted 16 May 2020 - 05:58 PM

View PostConstalation, on 16 May 2020 - 05:51 PM, said:

What? The event started Tuesday, April 28/2020. I didn't play in April. Page updated on May01 and hasn't since. Therefore it won't show data on my performance.

And the data shows that you are a less than average performer, and thus your opinion is nullified, according to others who have posted here.

Edited by Blackhorse11Cav, 16 May 2020 - 05:59 PM.


#1356 Constalation

    Member

  • Pip
  • Bridesmaid
  • Bridesmaid
  • 17 posts

Posted 16 May 2020 - 06:03 PM

View PostBlackhorse11Cav, on 16 May 2020 - 05:58 PM, said:

And the data shows that you are a less than average performer, and thus your opinion is nullified, according to others who have posted here.


Except you came to that conclusion based on outdated information coalesced from 9 months ago. Since this is a discussion about SoupQueue, comparisons of stats before and after soup implementation should be an important data point. You can also use my new stats from the few Solo Soupqueue matches I have completed as another data point.

#1357 Blackhorse11Cav

    Member

  • PipPip
  • Philanthropist
  • Philanthropist
  • 24 posts

Posted 16 May 2020 - 06:06 PM

View PostConstalation, on 16 May 2020 - 05:51 PM, said:

What? The event started Tuesday, April 28/2020. I didn't play in April. Page updated on May01 and hasn't since. Therefore it won't show data on my performance.


Not only didn't you play in April, you didn't play, appreciably in Sep, Oct, Nov, Dec,Jan, Feb, or March...Not judging, just stating the facts according to Jarl.

#1358 Ballistic Panicmode

    Member

  • PipPipPip
  • The Scattershot
  • 53 posts
  • LocationMaryland, USA

Posted 16 May 2020 - 06:10 PM

Rather than trying to fix the broken PSR system, why not rework the match scoring system to reflect the most fair assessment of player performance, then track average match score per owned mech? For new mechs, simply start with the player's average for the weight class then refine from there as matches are played. It would better account for the strengths / weaknesses of different builds and the individual's affinity for different playstyles. The MM could start with the 3/3/3/3 search for weight classes then tweak from there based on scores.

Edited by Ballistic Panicmode, 16 May 2020 - 06:12 PM.


#1359 DAEDALOS513

    Member

  • PipPipPipPipPipPipPipPipPip
  • The Flame
  • The Flame
  • 2,633 posts
  • LocationArea 52

Posted 16 May 2020 - 06:20 PM

View PostExcessive Paranoia, on 16 May 2020 - 05:18 PM, said:

An inability to identify a group does not prove its absence, my point still stands as well. I still stand by my assertion that the likelihood of any match having a group on one side and not the other is incredibly low. Its not impossible, but that would mean that out of all of the people playing, there is only one available group to place into any given match. That said, it is entirely possible that there are matches without any groups as there are still probably many more solo players than group players at any given time of the day. Ultimately though, if you see one group, you're likely seeing two and just don't know it.

When you start making assumptions you ain't goin nowhere.. his point stands.

Edited by DAEDALOS513, 16 May 2020 - 06:20 PM.


#1360 Zulu211

    Member

  • PipPip
  • Bridesmaid
  • 26 posts

Posted 16 May 2020 - 06:28 PM

View PostHorseman, on 16 May 2020 - 02:21 AM, said:

If I'm reading it right, he actually went from attacking Anomalocaris for only playing 100 games last season to claiming Ano only played 100 games total:
Something that is objectively false: https://leaderboard....?u=Anomalocaris




Well, I know you keep implying you've told me something but, I haven't been reading anything you've said because I dont care what you think. You're OBVIOUSLY not comfortable or remotely acquainted with the truth.

You fancy lads stick with your list.

As of my writing this Anomalocaris has no results found on the leaderboards in game.

https://mwomercs.com...er=Anomalocaris

Again, showing quite a bit of concern about a game he doesn't play. Or is the game file not a good enough source?





4 user(s) are reading this topic

0 members, 4 guests, 0 anonymous users