Jump to content


My Dear MM, please let me know what is wrong with the following picture :)

match maker

  • Please log in to reply
14 replies to this topic

Trippledot #1 Posted Jan 08 2019 - 01:18

    Corporal

  • Players
  • 18927 battles
  • 29
  • [GOONZ] GOONZ
  • Member since:
    02-21-2014

Greetings, 

you most probably got this request several times ... but just to be sure you are considering, in your TOP next fixes.

 look on the picture below, and try to guess what was the outcome?

 

we all WISH and HOPE, that MM (Matchmaker) will start considering balancing the PR, between the team (we all for a fair game here, right ? ;)

just to answer = the game ended 15-0 (it took 3min and 20s.) for the left team.

 

 

 



PapioTitan #2 Posted Jan 08 2019 - 01:35

    Staff sergeant

  • -Players-
  • 42780 battles
  • 413
  • Member since:
    12-21-2013

Seems legit. So you think PR is the key to balanced teams? I think there are a lot of different opinions on what represents true skill (WN8 and many other acronyms). What about numbers of battles? Oh wait, then there's rerolls....Oh, and platoons....and xxx. It's not a perfect system and never will be.

 

I did like the tank balance tho...until the last pairing. 432 vs Skorp G?



Ratnikk #3 Posted Jan 08 2019 - 01:37

    Sergeant

  • -Players-
  • 10267 battles
  • 128
  • Member since:
    07-20-2017
Honestly i dont think the MM is ''random'' as they state. I think they programmed the MM to favor one team over the other that way the matches end quicker this is useful since the population in the NA server is so low it makes the MM queue go faster. Its not normal or ''random'' that the majority of matches end in complete slaughter for one side.

Edited by Ratnikk, Jan 08 2019 - 01:37.


GeorgePreddy #4 Posted Jan 08 2019 - 01:48

    Major

  • Players
  • 14345 battles
  • 10,840
  • Member since:
    04-11-2013

View PostTrippledot, on Jan 07 2019 - 21:18, said:

we all WISH and HOPE, that MM (Matchmaker) will start considering balancing the PR, between the team...

 

Actually... most players do NOT want SBMM for Random Battles.

 

 

 

 

 

 



GeorgePreddy #5 Posted Jan 08 2019 - 01:49

    Major

  • Players
  • 14345 battles
  • 10,840
  • Member since:
    04-11-2013

View PostRatnikk, on Jan 07 2019 - 21:37, said:

Honestly i dont think the MM is ''random'' as they state. I think they programmed the MM to favor one team over the other that way the matches end quicker this is useful since the population in the NA server is so low it makes the MM queue go faster. Its not normal or ''random'' that the majority of matches end in complete slaughter for one side.

 

 

 

 

 

 

 

 



Trippledot #6 Posted Jan 08 2019 - 01:56

    Corporal

  • Players
  • 18927 battles
  • 29
  • [GOONZ] GOONZ
  • Member since:
    02-21-2014

View PostPapioTitan, on Jan 08 2019 - 01:35, said:

Seems legit. So you think PR is the key to balanced teams? I think there are a lot of different opinions on what represents true skill (WN8 and many other acronyms). What about numbers of battles? Oh wait, then there's rerolls....Oh, and platoons....and xxx. It's not a perfect system and never will be.

 

I did like the tank balance tho...until the last pairing. 432 vs Skorp G?

 

PR is not ideal (true) but its a very good starting point and by far better than nothing.

keep in mind that PR includes in it many aspects (win rate, damage, survivable rate ,rate,any more) yes not ideal but a great starting point.

 

i think WG need to publish the number of battles which ending in less than 5min. any battle which ends in less than 5min its a direct failure of MM.


Edited by Trippledot, Jan 08 2019 - 01:56.


Fat_Bob_1 #7 Posted Jan 08 2019 - 01:59

    Sergeant

  • -Players-
  • 19934 battles
  • 230
  • [INSNE] INSNE
  • Member since:
    12-06-2016

View PostTrippledot, on Jan 07 2019 - 17:18, said:

...

please let me know what is wrong...

 

 

That's easy Dot. You tied your shoelaces wrong today. :great:

 

f.y.i. The MM blows chunks...



NeatoMan #8 Posted Jan 08 2019 - 02:09

    Major

  • Players
  • 26979 battles
  • 18,445
  • Member since:
    06-28-2011

Balancing those two teams is a terrible solution.

 

A better solution would be to not place those two teams in the same battle together, and wait for other teams with more suitable PR ratings to come along, and put them into battle together instead.  You'd end up with a higher skilled battle and a lower skilled battle.  This way you don't need an entirely new team building process in the MM.  Just use already existing mechanisms, and keep teams random.

 

Most teams are suitable matches already so this wouldn't affect a majority of battles anyway, just the extreme mismatches like this one.

 

 



Trippledot #9 Posted Jan 08 2019 - 15:00

    Corporal

  • Players
  • 18927 battles
  • 29
  • [GOONZ] GOONZ
  • Member since:
    02-21-2014

View PostNeatoMan, on Jan 08 2019 - 02:09, said:

Balancing those two teams is a terrible solution.

 

A better solution would be to not place those two teams in the same battle together, and wait for other teams with more suitable PR ratings to come along, and put them into battle together instead.  You'd end up with a higher skilled battle and a lower skilled battle.  This way you don't need an entirely new team building process in the MM.  Just use already existing mechanisms, and keep teams random.

 

Most teams are suitable matches already so this wouldn't affect a majority of battles anyway, just the extreme mismatches like this one.

 

 

 

not sure if its a better solution, but a valid one for sure.

NeatoMan #10 Posted Jan 08 2019 - 15:12

    Major

  • Players
  • 26979 battles
  • 18,445
  • Member since:
    06-28-2011

View PostTrippledot, on Jan 08 2019 - 09:00, said:

not sure if its a better solution, but a valid one for sure.

It doesn't saddle good players with more tomatoes, like swapping players would.  You get to keep your random teams.   Good players don't get penalized nearly as much as they would with skill swaps. 

 

If you end up on a tomato team you simply won't have to play the "A" teams.  You'd get matched against an average team or similar scrub team.  If you are on an "A" team, you won't get to beat up on scrub teams.

 

If you are a good player, rather than dragging down the level of your team by adding more tomatoes, it would raise the level of your competition.  Your team mates would be no worse than the ones you currently get.

 

From a coding perspective it would be much easier



Ich_bin_Hass #11 Posted Jan 08 2019 - 15:16

    Captain

  • Players
  • 78760 battles
  • 1,025
  • [1VETS] 1VETS
  • Member since:
    04-30-2011

View PostRatnikk, on Jan 07 2019 - 18:37, said:

Honestly i dont think the MM is ''random'' as they state. I think they programmed the MM to favor one team over the other that way the matches end quicker this is useful since the population in the NA server is so low it makes the MM queue go faster. Its not normal or ''random'' that the majority of matches end in complete slaughter for one side.

 

Correct ! 

Ich_bin_Hass #12 Posted Jan 08 2019 - 15:18

    Captain

  • Players
  • 78760 battles
  • 1,025
  • [1VETS] 1VETS
  • Member since:
    04-30-2011

View PostGeorgePreddy, on Jan 07 2019 - 18:49, said:

 

 

 

 

 

 

 

 

 

Here is George carrying WG's water against. As he always does. Keep carrying that water Gunga Din !

Ich_bin_Hass #13 Posted Jan 08 2019 - 15:36

    Captain

  • Players
  • 78760 battles
  • 1,025
  • [1VETS] 1VETS
  • Member since:
    04-30-2011

View PostRatnikk, on Jan 07 2019 - 18:37, said:

Honestly i dont think the MM is ''random'' as they state. I think they programmed the MM to favor one team over the other that way the matches end quicker this is useful since the population in the NA server is so low it makes the MM queue go faster. Its not normal or ''random'' that the majority of matches end in complete slaughter for one side.

 

But really the population is so low now. You gonna face the illegal or legal mod using unicums 95% of the time anymore. So until those at WG's realize that the 3/5/7 MM is GARBAGE. Replace it with the old system. of + or - 2. Reduce the TD's to three per team, like they did SPG's. Stop the useless screwing with premium ammo to plz said unicums that whine all the time about it and the maps they hate. Those maps that supported every vehicle type in the game. That they whined about and successfully had removed. Where as now 90% of the maps are heavy tank brawling maps. Where you have to fight nose to nose with the Russian OP'd or under tier'd heavies and super heavies. As far as being random it is. But with such a large group being made up of the so called GUUD unicums. The only thing that will stop the slaughters of 15 to 0 or 1. Is to make sure that the teams are evenly matched. Which means making sure you have the same amount of unicums on each team. But these very same unicums HATE facing each other. They would much rather seal club at tiers 2 through 5 and then play 9's and 10's to get tier 8's served up on a silver plate by WG's. 

 

   Now my definition of a unicum. Is anyone who is Blue or Purple as far as XVM is concerned. 



Manimal__ #14 Posted Jan 08 2019 - 16:00

    Major

  • Players
  • 40024 battles
  • 3,558
  • Member since:
    08-20-2013

View PostPapioTitan, on Jan 07 2019 - 16:35, said:

Seems legit. So you think PR is the key to balanced teams?

 

In general, having 6 trippledots instead of 6 of you will increase the chances of a team winning. I'm not an advocate for SBMM because I don't believe it will improve the game drastically because we have seen in past events where it is based off of skill and blow outs still happen.

 

However I do believe that if you've played over 40k games...



Zepath #15 Posted Jan 09 2019 - 03:13

    Corporal

  • -Players-
  • 5465 battles
  • 68
  • Member since:
    05-28-2017

I use the mod that that compares the aggregate Player Win8 rating of one team vs. the other.   And there are often HUGE disparities in the two teams based off that.

I am not big on all these stats, there are too many factors that effect them and too many ways to exploit having high ones ... but there is a reality to, "The team with the most Win8 typically wins" ... not always, but typically.

Now, since these stats come from WG, and since they have many ways to monitor and test indicators for team wins (including how each player does with the given tank), they could make a better match-maker by just grabbing the same group of people and shuffling them within that group until they have a better balanced team.  They don't have to keep shuffling people in and out of the group ... just shuffle the people they have.

Admittedly it becomes more of a challenge when your have a platoon of 4+ that must be put on one side together.  But that is why there are people to get paid the big bucks to work out the statistics and write algorithms.

And there is the white turd in the room ... they just don't want to do it.   They keep bringing in new programmers, that aren't trained in statistics, and as unqualified as they are, throwing them at the match maker.  Then they throw us some red meat by telling us in an announcement they are working on.

The fact is, when WG wants a better MM, we will get one.

And before some of you pinheads go running off to check my stats ... this is my third account, and after losing my first account, I stopped caring about stats.  I will commonly suicide into a person who is holding up a group of my team mates.  That makes for crappy stats, low damage per game, few games survived, and a lot of missed shots because I'm moving.   I don't care ... its the most fun I've had in this game in a long time. 

I also enjoy playing SPGs which are the only challenge I find left in this game ... and no, I don't care about your opinion of SPG players.







Also tagged with match maker

1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users