Jump to content


Official Matchmaking Discussion Thread


  • Please log in to reply
1052 replies to this topic

TheFluffiestSeal #61 Posted 20 October 2017 - 01:41 PM

    Legendary Guardian of Sealkind

  • Players
  • 14870 battles
  • 1,569
  • [-RGN-]
  • Member since:
    08-09-2015
Playing in identical tanks is boring..

View PostDoc_01_, on 20 October 2017 - 01:30 PM, said:

It is a game so ALL HAVE FUN YES FUN TANKING 

 

Stop making excuses for your 40% winrate. You have a better chance at convincing me matchmaker is rigged personally against you than your crap is good. If it's in the game, mentioned countless times as a good thing you should use by the developers, it's meant to be used. 

New to tier X? Here’s the meta, with the all new tier x tier list.

http://forum.wotblitz.com/index.php?/topic/92877-the-ultimate-tier-x-tier-list/

#1 Tiger I Player NA, #4 Global

Seal Clubbing (v.)- The act of playing any strong or overpowered tank at tiers 1-6 for the purpose of dominating newer players.

Co-founder of the Training Room Police Department (TRPD)


SpartacusDiablo #62 Posted 20 October 2017 - 01:43 PM

    Highly Illogical

  • Players
  • 24243 battles
  • 7,883
  • [III-R]
  • Member since:
    02-17-2014
This belongs here.

How does matchmaking work?

Close your eyes and imagine a bucket full of s*** with just a few diamonds sprinkled in. Now reach into the bucket and grab two  handfuls​ of s*** with the occasional diamond.

You hope for a diamond but chances are you're just going to have a handful of s***.

That is basically what the Matchmaker does when it makes our teams. Two random handfuls of s*** with the occasional diamond thrown in.

Look, these events just make it even worse. Now instead of a  bucket full of s*** you have a 50 gallon drum of s***. Unfortunately the number of diamonds has stayed about the same. That just means that every time you reach into the bucket you're more than likely going to get nothing more than a handful of s***.

The only way to guarantee that you get a diamond on your team is if you are the diamond.  

Find me on Discord, [SpartacusDiablo#7879]

Rest in Peace CJ.  You will be missed.


Revenge_of_Peppa_the_pig #63 Posted 20 October 2017 - 03:31 PM

    Junior Sergeant

  • Players
  • 26088 battles
  • 127
  • [DMONS]
  • Member since:
    05-01-2016

View PostSpartacusDiablo, on 20 October 2017 - 01:43 PM, said:

This belongs here.

How does matchmaking work?

Close your eyes and imagine a bucket full of s*** with just a few diamonds sprinkled in. Now reach into the bucket and grab two handfuls​ of s*** with the occasional diamond.

You hope for a diamond but chances are you're just going to have a handful of s***.

That is basically what the Matchmaker does when it makes our teams. Two random handfuls of s*** with the occasional diamond thrown in.

Look, these events just make it even worse. Now instead of a bucket full of s*** you have a 50 gallon drum of s***. Unfortunately the number of diamonds has stayed about the same. That just means that every time you reach into the bucket you're more than likely going to get nothing more than a handful of s***.

The only way to guarantee that you get a diamond on your team is if you are the diamond.

its easy work . pretty much like pc version .  you have internal rating .that number represent your weight inside team. more weight you have weaker teammates will be. you won +1 rating . you lost -1 rating . if you keep push .and win your internal rating will increase . and you will end in fail strike. since you will be throwed in utter trash .you will loose . till your rating restore to value where you will start play  fine .

 

that internal rating does not affect only player rating . since you cant predict human stupidity . but you can  predict server side things . like

1. chance to hit . you shoot outside circle ? that random ofc .  you did not penned amx12t in rear from ap ? that random ofc .

2. chance of module take damage

3. chance of crew take damage .


Edited by Fred_Adams, 20 October 2017 - 03:37 PM.


SpartacusDiablo #64 Posted 20 October 2017 - 04:40 PM

    Highly Illogical

  • Players
  • 24243 battles
  • 7,883
  • [III-R]
  • Member since:
    02-17-2014

View PostFred_Adams, on 20 October 2017 - 10:31 AM, said:

its easy work . pretty much like pc version .  you have internal rating .that number represent your weight inside team. more weight you have weaker teammates will be. you won +1 rating . you lost -1 rating . if you keep push .and win your internal rating will increase . and you will end in fail strike. since you will be throwed in utter trash .you will loose . till your rating restore to value where you will start play  fine .

 

that internal rating does not affect only player rating . since you cant predict human stupidity . but you can  predict server side things . like

1. chance to hit . you shoot outside circle ? that random ofc .  you did not penned amx12t in rear from ap ? that random ofc .

2. chance of module take damage

3. chance of crew take damage .

Are you implying that a system like this is in place in Blitz?  If so, I just don't see it.  If a system like this is in use why are there still players at extreme ends of the spectrum. 30%ers and 70%ers alike.  If a system like you described was in use we wouldn't see a spread like this.  We wouldn't see a player base where 3 out of 5 players are unable to pass the 50% winrate mark.  Sorry but I just don't see it.


Find me on Discord, [SpartacusDiablo#7879]

Rest in Peace CJ.  You will be missed.


_Mendokusaii #65 Posted 20 October 2017 - 04:56 PM

    Senior Sergeant

  • Players
  • 32067 battles
  • 704
  • [-RGN-]
  • Member since:
    06-12-2016
Am I the only one who doesn’t care about mm here? I mean it’s a flaw for awhile and the solutions presented aren’t going to be used by Wargaming. I’m sure they’re aware of the issue, but the solutions are a bit more complicated than we all give credit for.

Grayghost2000 #66 Posted 20 October 2017 - 09:44 PM

    First Sergeant

  • Players
  • 17081 battles
  • 2,071
  • [MBTS]
  • Member since:
    09-27-2015
Can someone actually provide me with the algorithm used for the match making? Everything being stated here is hearsay.

airstiker #67 Posted 20 October 2017 - 11:13 PM

    Sergeant

  • Players
  • 15467 battles
  • 338
  • Member since:
    07-06-2014
This thread is for people who think WG actually gives a single [edited]. They’re just trying to clean up the forums from MM posts

Revenge_of_Peppa_the_pig #68 Posted 20 October 2017 - 11:15 PM

    Junior Sergeant

  • Players
  • 26088 battles
  • 127
  • [DMONS]
  • Member since:
    05-01-2016

View PostSpartacusDiablo, on 20 October 2017 - 04:40 PM, said:

Are you implying that a system like this is in place in Blitz?  If so, I just don't see it.  If a system like this is in use why are there still players at extreme ends of the spectrum. 30%ers and 70%ers alike.  If a system like you described was in use we wouldn't see a spread like this.  We wouldn't see a player base where 3 out of 5 players are unable to pass the 50% winrate mark.  Sorry but I just don't see it.

 

that an old story . surprising you still don't know . how it works .

main thing . even if you have all server might and best possible ( for that moment players . ) . its not 100% that you will win . its only add variativity ( atleast in paper ) for the win . coz you must do something which will contribute to victory .like proper moving and shooting things . without that all server might is useless . why 20-40% still in game ? that's simple . portions of auditory is BAAAAAD players for variety of reasons .

1. outdated device or LOL connection - well . tbh blitz isssss tough for many phones. even on lowest possible settings . also require decent connection . what if your device is old like you and connection is LOL . but you wanna play ? what will hold you from that ?

2. schoolers and people who just bad at games . just 2 stories from my life .

 

when i was 8 year old kid . mom buyed me first pc ( for studies ofc )  . and like all teenagers those years .i installed counterstrike 1.5  and GTA ( coz all cool kids play it . you don't wanna be cool ? . after install i joined internet game . well my connection was - MODEM. cool sweet US robotics sportster 56 kilobit/sec .in sad end . i was a teenager who are noob at CS . also my connection was awesome . 700 ping . yes i was kicked many times . and so on . but you cant imagine amount of joy .  when i first time fragged someone . like all cool kids .but could i contribute something to the team ? nope .

 

colleague . saw  me playing blitz .  installed on his phone . barely passed  basic training . and started to play . after 2 week i saw something really awful .

30% winrate on kv1 . he just standing open .and throw ap in siluette . that's all .

 

moral ? that's one of the many examples.  game mechanic itself DOES NOT PROMOTE GOOD PLAYING . for now it based on - play as you can . will it force ppl to learn atleast game basics

3. funny but free accounts as well . if you play without prem . creds shortage . will often force you go battle wo equipment and so on. it will affect your performance .

  



SpartacusDiablo #69 Posted 20 October 2017 - 11:29 PM

    Highly Illogical

  • Players
  • 24243 battles
  • 7,883
  • [III-R]
  • Member since:
    02-17-2014

View PostFred_Adams, on 20 October 2017 - 06:15 PM, said:

 

that an old story . surprising you still don't know . how it works .

main thing . even if you have all server might and best possible ( for that moment players . ) . its not 100% that you will win . its only add variativity ( atleast in paper ) for the win . coz you must do something which will contribute to victory .like proper moving and shooting things . without that all server might is useless . why 20-40% still in game ? that's simple . portions of auditory is BAAAAAD players for variety of reasons .

1. outdated device or LOL connection - well . tbh blitz isssss tough for many phones. even on lowest possible settings . also require decent connection . what if your device is old like you and connection is LOL . but you wanna play ? what will hold you from that ?

2. schoolers and people who just bad at games . just 2 stories from my life .

 

when i was 8 year old kid . mom buyed me first pc ( for studies ofc )  . and like all teenagers those years .i installed counterstrike 1.5  and GTA ( coz all cool kids play it . you don't wanna be cool ? . after install i joined internet game . well my connection was - MODEM. cool sweet US robotics sportster 56 kilobit/sec .in sad end . i was a teenager who are noob at CS . also my connection was awesome . 700 ping . yes i was kicked many times . and so on . but you cant imagine amount of joy .  when i first time fragged someone . like all cool kids .but could i contribute something to the team ? nope .

 

colleague . saw  me playing blitz .  installed on his phone . barely passed  basic training . and started to play . after 2 week i saw something really awful .

30% winrate on kv1 . he just standing open .and throw ap in siluette . that's all .

 

moral ? that's one of the many examples.  game mechanic itself DOES NOT PROMOTE GOOD PLAYING . for now it based on - play as you can . will it force ppl to learn atleast game basics

3. funny but free accounts as well . if you play without prem . creds shortage . will often force you go battle wo equipment and so on. it will affect your performance .

  

I'm sorry but that is virtually unreadable. If you wish to debate this issue I am going to have to ask that you do a better job of putting your thoughts into the written word.


Find me on Discord, [SpartacusDiablo#7879]

Rest in Peace CJ.  You will be missed.


CA_vampire #70 Posted 20 October 2017 - 11:37 PM

    First Sergeant

  • Players
  • 25915 battles
  • 1,103
  • [MOMB]
  • Member since:
    11-21-2016

View PostSpartacusDiablo, on 20 October 2017 - 06:43 AM, said:

This belongs here.

How does matchmaking work?

Close your eyes and imagine a bucket full of s*** with just a few diamonds sprinkled in. Now reach into the bucket and grab two handfuls​ of s*** with the occasional diamond.

You hope for a diamond but chances are you're just going to have a handful of s***.

That is basically what the Matchmaker does when it makes our teams. Two random handfuls of s*** with the occasional diamond thrown in.

Look, these events just make it even worse. Now instead of a bucket full of s*** you have a 50 gallon drum of s***. Unfortunately the number of diamonds has stayed about the same. That just means that every time you reach into the bucket you're more than likely going to get nothing more than a handful of s***.

The only way to guarantee that you get a diamond on your team is if you are the diamond.

 

I understand what you are saying, but I hope you realize these are *your* assumptions. You don't know if the programmers and managers in WG choose the software to work like this. You are making an assumption based on your feeling of how the game should work. 

 

If you think a little bit about it, you will realize it cannot be that simple. Those players are in 10 tiers, and you have to pick two tiers (or three if there is a failtoon). For example, how do you decide if the tiers are 5-6 and not 5-4?  And you also need to pick two platoons or no platoons, or if this is not possible then a single platoon. And you have to do this in a reasonable time, so that people do not wait too long, so there is a time variable as well.

 

Already, I gave you a lot of parameters. Those are the obvious parameters. We don't know if those are the only parameters. And that is only for picking 14 players. The fun part follows: how do you split 14 players into two teams? No, it does not happen by magic or chance. A programmer wrote an algorithm to do this. We do not know what is included in that algorithm. 

 

 



__Crusader6__ #71 Posted 21 October 2017 - 01:30 AM

    Annoying Mod

  • Players
  • 58763 battles
  • 11,490
  • [-C6-]
  • Member since:
    12-08-2014

WG has stated repeatedly that the WOT MM is NOT USED IN BLITZ. 

  While my faith in WG is low, they make no more or less $ from the player base by hiding the MM methods 

 

WG has repeatedly said for Blitz that MM simply picks 7 tanks for red and 7 tanks for green. 

   Based on the fact you get 7-0 games and a lot of imbalanced tank types - based on 42,000 plus games and logging well over 2,500 battles by team/individual tank and WR  I can reasonably deduce that no MM weighting exists like WOT.  

 

Furthermore I am a stats junkie here.   I track pretty much everything I can. 

My purchasing stuff from WG hasn’t seem to have any influences in my teams. 

   One can buy $100 worth of gold etc and then have absolutely terrible teams - and one can have great teams with a few months of non spending. 

 

One thing I have noted from all of my wasted time tracking this - you the individual player can have the largest impact on whether you win or lose.  

  1) Don’t take early Damage

  2) Get early spots (see #1 though yolo is out) 

  3) Herd your cats (direction/plan and calling targets). 

 

 

 

 

 


 
Tank Hoarder: 365 tanks in Garage:  366/368 aced (AMX 30B, Mk1 and T49A repo),    wallet warrior.  Loyal M60 owner
I think 5.5 was good for the game - But I also want my Kenny OP nerfed. 
Sorry I haven’t aced the Smasher yet - So I’m still ruining tier 6-8 

 


MarBearCat #72 Posted 21 October 2017 - 04:18 PM

    Pennsylvania's Favorite Cat!

  • Players
  • 22386 battles
  • 11,610
  • [SDWA]
  • Member since:
    11-13-2015

MM is especially bad when you have your "honeymoon" and when your tank's anniversary comes up.

 

I've noticed this. I can't ever remember getting good players on my teams when I'm involved with one of

the two above.


 

 

Heavy Is The Cat That's Carried!


SanguinaryDan #73 Posted 21 October 2017 - 08:05 PM

    First Sergeant

  • Players
  • 17068 battles
  • 1,237
  • [III-P]
  • Member since:
    07-06-2014
I just wish it would put a little more effort into mixing the types of tanks on each side.  Particularly at the upper tiers where running into a team with three heavies and four 150mm plus TDs while driving a medium is no damned fun at all.
The point of the game is to win.  The purpose of the game is to have fun.  They are not mutually exclusive.

TheFluffiestSeal #74 Posted 22 October 2017 - 12:06 PM

    Legendary Guardian of Sealkind

  • Players
  • 14870 battles
  • 1,569
  • [-RGN-]
  • Member since:
    08-09-2015
I've always done better solo than in a platoon. Is it because platoon matchmaker is different? Another reason could be that our play styles don't mix. I have a strange, selfish playstyle that does wonders solo.

New to tier X? Here’s the meta, with the all new tier x tier list.

http://forum.wotblitz.com/index.php?/topic/92877-the-ultimate-tier-x-tier-list/

#1 Tiger I Player NA, #4 Global

Seal Clubbing (v.)- The act of playing any strong or overpowered tank at tiers 1-6 for the purpose of dominating newer players.

Co-founder of the Training Room Police Department (TRPD)


SpartacusDiablo #75 Posted 22 October 2017 - 06:08 PM

    Highly Illogical

  • Players
  • 24243 battles
  • 7,883
  • [III-R]
  • Member since:
    02-17-2014
My personal thoughts backed up by my in game experiences are that Matchmaking is random.  While yes there are adjustments made at times (honeymoon period) they are applied to all players equally. There is no rigging of one player over another.

That being said there are some things I'd change.

1. Limit tank type per team.  2 TD, 2 L, 3 M, and 3 H.
2. Match tiers between platoons. No more having your tier 6 platoon matched against a tier 7.
3.  Every player is given a point rating depending on their winrate at the tier they are currently playing.  A winrate that is padded playing tier 3 is irrelevant at tier 7. These points would be used when assigning teams, similar to ranked battles.

While this system may lead to longer wait times I feel it would lead to fairly balanced teams as well as a more dynamic game.

Find me on Discord, [SpartacusDiablo#7879]

Rest in Peace CJ.  You will be missed.


SpartacusDiablo #76 Posted 22 October 2017 - 10:17 PM

    Highly Illogical

  • Players
  • 24243 battles
  • 7,883
  • [III-R]
  • Member since:
    02-17-2014

Lol.  A 48%er whining about Matchmaking on Facebook and the answer from Wargaming.


Find me on Discord, [SpartacusDiablo#7879]

Rest in Peace CJ.  You will be missed.


Hbreezy #77 Posted 23 October 2017 - 05:10 AM

    Lance-corporal

  • Players
  • 23242 battles
  • 51
  • [TSLEG]
  • Member since:
    05-16-2017
Anybody read their mm algo patent application? Discusses curbing win streaks and losing streaks by placing the particular player on more or less favorable teams.Win 8 in a row, be put on team full of inexperienced and or low wr rate players. Could be random could be the algo but dont think for a second all the variables cant be computed in a reasonable amount of time to "rig" the team imbalances

TheFluffiestSeal #78 Posted 23 October 2017 - 02:41 PM

    Legendary Guardian of Sealkind

  • Players
  • 14870 battles
  • 1,569
  • [-RGN-]
  • Member since:
    08-09-2015

How to rig matchmaker to your own needs in 1 easy step today!
 

Step 1: Uninstall. If you're below 50, then the average player base skill will go up. If you're above 50, the opposite happens.


New to tier X? Here’s the meta, with the all new tier x tier list.

http://forum.wotblitz.com/index.php?/topic/92877-the-ultimate-tier-x-tier-list/

#1 Tiger I Player NA, #4 Global

Seal Clubbing (v.)- The act of playing any strong or overpowered tank at tiers 1-6 for the purpose of dominating newer players.

Co-founder of the Training Room Police Department (TRPD)


krietenstein34 #79 Posted 23 October 2017 - 03:22 PM

    First Sergeant

  • Players
  • 29273 battles
  • 1,972
  • [HAHA]
  • Member since:
    01-03-2015
Totally from the cuff observation since coming back after 6 months...............   I've only been playing Tiers 3, 4 and a little 5.     I've actually taken the time to look at my mates, and foes WR's and number of battles.............    In a nutshell.......................  they rarely have any experience or a WR worth a hoot.   Very common to see less than 100 battles.     I can go into a Tier 3 fairly confident ODDS wise, I'm the best player on the field.   And If I get anyone with a clue, we'll probably win.  It's very rare to see ANYONE with any skill or a WR over 50....................   I don't remember it being that bad before.    Yes, they were 45%s in past, but they had 3k battles.    I was astounded at the lack of battles people had over my last week.   No math, just observation.
:girl:

SpartacusDiablo #80 Posted 23 October 2017 - 03:36 PM

    Highly Illogical

  • Players
  • 24243 battles
  • 7,883
  • [III-R]
  • Member since:
    02-17-2014

View Postkrietenstein34, on 23 October 2017 - 10:22 AM, said:

Totally from the cuff observation since coming back after 6 months...............   I've only been playing Tiers 3, 4 and a little 5.     I've actually taken the time to look at my mates, and foes WR's and number of battles.............    In a nutshell.......................  they rarely have any experience or a WR worth a hoot.   Very common to see less than 100 battles.     I can go into a Tier 3 fairly confident ODDS wise, I'm the best player on the field.   And If I get anyone with a clue, we'll probably win.  It's very rare to see ANYONE with any skill or a WR over 50....................   I don't remember it being that bad before.    Yes, they were 45%s in past, but they had 3k battles.    I was astounded at the lack of battles people had over my last week.   No math, just observation.

You're playing in tiers 3,4, and 5... Of course the majority of players there will be inexperienced and have a low battle count.


Find me on Discord, [SpartacusDiablo#7879]

Rest in Peace CJ.  You will be missed.





3 user(s) are reading this topic

0 members, 3 guests, 0 anonymous users