Edited by suffocater, 29 August 2017 - 03:02 PM.


Is The Matchmaker Getting Worse?
#1
Posted 29 August 2017 - 10:23 AM
#2
Posted 29 August 2017 - 11:25 AM
I saw solo drops in twitch, where MM had the brilliant idea to put Proton and Boswer in the same team for 3 (three) consecutive matches.
guess how the matches ended....
I assume MM was sure there were 2-3 ultra competitive players in the other team..... but sadly they got all stomped. (pure misfortune, right?)
#3
Posted 29 August 2017 - 11:29 AM
#4
Posted 29 August 2017 - 11:46 AM
It simply puts you into a game with the sole purpose of feeding the other team cbills and XP.
Over the last week and a half no matter what I play or how hard I try it's stomp after stomp.
I doubt even 8v8 would even fix this, when the MM already does a horrible job.
#5
Posted 29 August 2017 - 11:48 AM
it´s perfect.
no lopsided matches, perfect combinations of skilled players and matching tonnage.
how dare you say the MM is getting worse.
pgi works on it´s flaws nonstop.
do not critizise what you do not understand.
in an alternate universe...
#6
Posted 29 August 2017 - 12:56 PM


#7
Posted 29 August 2017 - 02:13 PM
As soon as mechs start dying on one side, every mech has an extra mech shooting at it, and things just snowball from there.
The matchmaker may be fubar, but not every stomp is immediately because one team had all of the good pilots.
Edited by LT. HARDCASE, 29 August 2017 - 02:14 PM.
#8
Posted 29 August 2017 - 02:18 PM
#9
Posted 29 August 2017 - 02:22 PM
LT. HARDCASE, on 29 August 2017 - 02:13 PM, said:
As soon as mechs start dying on one side, every mech has an extra mech shooting at it, and things just snowball from there.
The matchmaker may be fubar, but not every stomp is immediately because one team had all of the good pilots.
At the same time, we notice the stomp because they werent so prevalent before. Losing a mech now is just that worse than losing one 2 years ago. Perfect stomp 12-0 in quick play means something went wrong. With all the firepower on the field now and with so much tonnage to go around, If your team happen to have the worse chassis and the worse loadout and etc etc. It goes downhill fast.
#10
Posted 29 August 2017 - 03:30 PM

#11
Posted 30 August 2017 - 03:56 AM
Else how is it that MM takes so long to give me a crap match after crap match in which MM somehow puts all the good players on one side constantly more then not.
I betting if we can see the tiers, we will not be shocked to see T1 and T5 players playing together.
Either that or the logical progress of their XP bar has arrived.
In which regardless of skill all are in T1. Some took 2 weeks, while others took 1 year. But all are now T1 regardless of any skill or sense.
Edited by Gaden Phoenix, 30 August 2017 - 04:03 AM.
#12
Posted 30 August 2017 - 04:09 AM
With PSR only dividing those that don't play often and those that play more the usability of this "MM"criteria is questionable. (don't forget that you can(Could?) boost your PSR by just mounting AMS.... so the difference between a good and average player are the mount of 1.5-3tons.
I don't even want to start why the idea of a MM in QP is flawed right from the beginning. A MM for what?
Come two Tier 1 player in the bar both are LRM Bobs, on the other side of the bar are already 2 GaussPPC Bullies.... hey same tier level - or even if not - you don't know anything about your team when dropping?
And afaik the MM don't give a damn about rolls.... in extreme you end in a game with 12 LRM Boats in one team and 12 scouts in the other
so please if the MM is still increasing the wait times shut the damn thing off, it can't get worse
Edited by Karl Streiger, 30 August 2017 - 04:10 AM.
#13
Posted 30 August 2017 - 04:15 AM
I agree with the above, over the past few weeks there have been fewer and fewer complete teams dropping.... Now, from what I've seen, and this isn't scientific in any way, 6 or 7 team members and the rest are random teams players and the number of non-team associated players seems to be growing with an average over 30 + drops of 3....... Didn't see more than 1 for months....
????? I've cut playing WAY back. The people I know that use to be playing all day aren't there anymore. In fact many, haven't been logged on since the CW update........ ????? So, I guess I'm as confused as everybody else....
#14
Posted 30 August 2017 - 04:16 AM
Gaden Phoenix, on 30 August 2017 - 03:56 AM, said:
See my post above.
No reason at all to put Proton and Boswer in the same team in 3 consecutive solo drops, even though playerbase is short.
Edited by Stefka Kerensky, 30 August 2017 - 04:16 AM.
#15
Posted 30 August 2017 - 04:16 AM
suffocater, on 29 August 2017 - 10:23 AM, said:
Why shut it down, when they can just go 8v8?
With smaller teams the MM will be able to build teams more effectively with the declining population. Then PGI doesn't have to bother with pesky algorithms or programming to fix the thing, shut it down and admit there is a problem, or bother making the game better or advertising so that maybe more folks would be drawn to it and give the MM something to work with.
See? No reason to shut it down. Going 8v8 will make everything all better. For a while.
Edited by Bud Crue, 30 August 2017 - 04:17 AM.
#16
Posted 30 August 2017 - 04:26 AM
Bud Crue, on 30 August 2017 - 04:16 AM, said:
With smaller teams the MM will be able to build teams more effectively with the declining population. Then PGI doesn't have to bother with pesky algorithms or programming to fix the thing, shut it down and admit there is a problem, or bother making the game better or advertising so that maybe more folks would be drawn to it and give the MM something to work with.
See? No reason to shut it down. Going 8v8 will make everything all better. For a while.
8v8 doesn't change the fact that the concept is flawed to begin with.
I can accept a MM as kind of whelp protection - but for anything else its bad
bad because those whelps need to walk through the ranks and learn bad habits (by watching other bad habits)
and after some weeks or months they finally get into T3 just to realize that it is a complete different game?
Heck when MM was intact and (thanks to the missing intention to play this old **** again and again ) i was T4 I could YOLO-Atlas drop and usually ended with 3 SoloKills - and then the MM was dead and suddenly i get killed and killed and killed.... before I was able to remember things I had forgotten -(time before PSR)
So PSR is a bad metric, but its not the PSR at all - also the reason to have player skill based MM is bad. Maybe I'm to stupid but I don't understand the incentive behind a player skill based MM in a TeamGame
#17
Posted 30 August 2017 - 04:31 AM
Really, I just had a team on Turmaline, that was stomped because, emm...it was team of laser vomit Clanners, who were simply overheating after every single shot. Who they are? Tier 5?
Edited by MrMadguy, 30 August 2017 - 04:36 AM.
#18
Posted 30 August 2017 - 05:10 AM
Exilyth, on 29 August 2017 - 12:56 PM, said:
LOL. That is actually, provably, true...
- In the past 10 days of PUGing, my win:loss dropped from 1.2 to 1.01.
- By reverse calculation, that means I was placed on the losing team 75% of the time.
- I achieved a top 3 score, and top 3 kills on my team, in almost all of those matches. I wasn't leeching with LRMs, sniping from the back, or any of the other selfish douche playstyles designed to pad stats - I was trying to win.
- My k/d ratio climbed from 1.6 to 1.7 during the period, despite all the losses.
Bad luck, I guess...? Unless Paul has enabled some kind of secret retribution kill switch due to the way I've been more candidly calling out PGI development BS lately


Edited by Appogee, 30 August 2017 - 05:20 AM.
#19
Posted 30 August 2017 - 05:17 AM
Stefka Kerensky, on 30 August 2017 - 04:16 AM, said:
Yeah, it's just sloppy coding. It would be so easy to add a couple of lines to at least distribute unit tags equally across both teams. A competent developer would actually distribute players based on their individual score or ranking across the two teams. Hell, sorting is one of the first things they teach you in programming.
Incidentally, I too had two solo queue matches two weeks ago where Proton and Bowser were on the same team. They also had a Kcom player. And yet there were only two people on my side who had a unit tag at all.
PGI seems to have no pride of workmanship.
#20
Posted 30 August 2017 - 05:51 AM
DAYLEET, on 29 August 2017 - 02:22 PM, said:
My guess is that stomps in solo QP could have increased since voip was introduced. We all now that coordination is key to winning. Maybe the team which has the luck to get a competent leader has such a big advantage over the team without comm activity that it also has a quite high chance for stomping them.
In my experience / memory most stomps where my team has won had a leader who called targets and movement directions. If such a team engages a passive one, a stomp is probably almost the only possible outcome. This observation is further supported by watching guys like Baradul who very often encourages his team to push early. And it's very common to see that those matches end in a stomp.
So maybe all it takes to stomp in solo queue is an aggressive team vs. a passive team, and voip made it much easier to induce this aggressiveness into a pug team, therefore the perceived increase in stomps. It could be that MM has not much influence here.
Stefka Kerensky, on 30 August 2017 - 04:16 AM, said:
Appogee, on 30 August 2017 - 05:17 AM, said:
Problem is that MM algorithm just looks at PSR Tier. And it just sees that Proton and Boswer are T1 like many others in the queue. It can't differentiate between a very good and an average T1 player, so MM only tries to match the number of T1 players on each side. The fact that both end up on the same team repeatedly is just coincidence, like it's possible to flip a coin and get heads 3 or even much more times in a row...
A better MM which takes more factors into account or differentiates within a tier level however would probably mean longer queue times because playerbase is not big enough for a more detailed MM. The small playerbase is probably also the reason why i often play with guys who behave like T4 or T5. But as frustrating as it can be from time to time, it's hard to improve this situation without getting lots of fresh blood into the game...
Edited by Daggett, 30 August 2017 - 05:53 AM.
1 user(s) are reading this topic
0 members, 1 guests, 0 anonymous users