Jump to content


Official Matchmaking Discussion Thread


  • Please log in to reply
1515 replies to this topic

_Crusader6_ #1461 Posted 23 September 2020 - 06:30 PM

    Vote Slime Mold

  • Players
  • 70713 battles
  • 15,699
  • [BDGER]
  • Member since:
    12-08-2014

View Postthenerfdude, on 23 September 2020 - 07:09 AM, said:

You really should not be able to have a silver league player in your team when you are platinum league, i was grinding ratings and this guy in an e100 pulled me down 3 games in row because he had no clue what he was doing :"u

 

i dont care if i need to wait longer for a match, i play ratings so i can play with people in my same.... rating. why should i play ratings if i have the same quality teams as randoms? The teams only get good when you hit +4.8k and its hard to grind up to that level when playing with 13 year old e100 drivers on your team.

 

pls fix wg

 

 

 


Teams don't get good in ratings period.

 

1) Rated MM tries to find a similar rating for both teams -- you can have a 5k player, with 6 2k's and WG can try to balance it by having 2 4k players, 2 3k, and 3 1k's.

2) Rated does not factor in how you got the rating -- some players play their rating in 7-8, and then play terribly in X to try to get cams there.

3) Rated Season is short - so many players stopped caring about rated.

4) Player Pool for Rated in small.

 

Lastly who really cares about rated?
 


 

 
Tank Hoarder: 424 tanks in Garage:  435/435 Played Tanks Aced
 
This space for rent until next terrible mechanic ;) 
 
    Wallet Warrior: Loyal Original M60 owner
 
 

 


Hye_yeon_Cho #1462 Posted 25 September 2020 - 04:32 PM

    Sergeant

  • Players
  • 954 battles
  • 440
  • [4_EVA]
  • Member since:
    09-22-2019
I love it when tier 10s see tier 8s

THERebrth #1463 Posted 26 September 2020 - 12:07 AM

    Sergeant

  • Players
  • 9960 battles
  • 427
  • [CLEAR]
  • Member since:
    12-20-2013

View Postbob_000, on 14 September 2020 - 06:34 PM, said:

 

The moment you stop blaming your team for your own performance, and the moment you understand the importance of your individual contribution to the game, the better you'll become as a player. Eventually, you'll understand that matchmaking isn't rigged after all once you recognize that you can get good teams and you can get bad teams. The only thing that is consistent in match making is you. If you can pull your weight and beyond in games, you'll become a better player and win more. 

Too bad this game is very team oriented, You can have very good games and still lose often. And MM really comes down to dumb luck. It especially sucks when you can see losses a mile away just from looking at the team comp. Castilla especially sucks when your team is down a med, so they decide to camp spawn with no spots. But at least I can say with confidence that 90% of my games that i lose can be attributed to my team for not playing well enough. Its especially common to do well and still lose in high tiers because your team decides to do as much damage as you should do in low tiers


Raging Dumbo


_Crusader6_ #1464 Posted 26 September 2020 - 11:30 PM

    Vote Slime Mold

  • Players
  • 70713 battles
  • 15,699
  • [BDGER]
  • Member since:
    12-08-2014

View PostTHERebrth, on 25 September 2020 - 07:07 PM, said:

Too bad this game is very team oriented, You can have very good games and still lose often. And MM really comes down to dumb luck. It especially sucks when you can see losses a mile away just from looking at the team comp. Castilla especially sucks when your team is down a med, so they decide to camp spawn with no spots. But at least I can say with confidence that 90% of my games that i lose can be attributed to my team for not playing well enough. Its especially common to do well and still lose in high tiers because your team decides to do as much damage as you should do in low tiers


Re-read what Bob wrote.  
    Being down a med on Castilla is not an issue if you play knowing that. 
Everyone gets the same gene pool to pull ‘team mates’. It’s what you can do with that team that defines how often you win or lose. 
 

 

 


 

 
Tank Hoarder: 424 tanks in Garage:  435/435 Played Tanks Aced
 
This space for rent until next terrible mechanic ;) 
 
    Wallet Warrior: Loyal Original M60 owner
 
 

 


wrecker1968 #1465 Posted 27 September 2020 - 01:35 AM

    Head of Security

  • Players
  • 56025 battles
  • 1,976
  • [HOUSE]
  • Member since:
    04-13-2016
:D

THERebrth #1466 Posted 27 September 2020 - 08:17 AM

    Sergeant

  • Players
  • 9960 battles
  • 427
  • [CLEAR]
  • Member since:
    12-20-2013

View Post_Crusader6_, on 26 September 2020 - 05:30 PM, said:


Re-read what Bob wrote.  
    Being down a med on Castilla is not an issue if you play knowing that. 
Everyone gets the same gene pool to pull ‘team mates’. It’s what you can do with that team that defines how often you win or lose. 
 

 

 

not as simple as you think, its always difficult in castilla if ur team ends up stuck in one part of the map or the other, castilla is a small map so once the enemy team gets map control, you are most likely going to lose. And Ofc I do my best in calling out directions, but its still annoying when you lose in rows even after doing your best


Raging Dumbo


BillyNair #1467 Posted 27 September 2020 - 08:53 AM

    Junior Sergeant

  • Players
  • 10596 battles
  • 116
  • [MOD3]
  • Member since:
    04-15-2020

My team killed 2 or less tanks in 23 of my last 30 games. I am not sure how the matchmaking works for players and their WR, but, I would think that after a while, I would accidentally be put on a team that can shoot a tank. We won 5 of 30, but, there are stretches of 9-12 losses in between those wins... Up until today, I have NEVER lost more than 7 in a row.

 

Some people say MM is working as intended...

 

My question is, do they take WR into account at all or is it purely based off the tanks with no regards to who is driving it?

 

If they dont base it off the players, why dont they?

 

If there is at least 1 line of code looking at players (vs tanks) what is it?

 

As I said, I have never lost more than 7 in a row ever, so, i was under the impression that they had SOME code checking that (loose too many in a row and players start to quit, put those players on decent teams, let them win a few now and then, you will keep them playing... more a matter of player retention than being fair, better for the share-holders)

 

Do they have code to look at WRs or loss-streaks?


Edited by BillyNair, 27 September 2020 - 08:54 AM.


BillyNair #1468 Posted 27 September 2020 - 08:54 AM

    Junior Sergeant

  • Players
  • 10596 battles
  • 116
  • [MOD3]
  • Member since:
    04-15-2020

DELETE THIS WAS A DUPLICATE


Edited by BillyNair, 27 September 2020 - 08:55 AM.


THERebrth #1469 Posted 27 September 2020 - 08:59 AM

    Sergeant

  • Players
  • 9960 battles
  • 427
  • [CLEAR]
  • Member since:
    12-20-2013

View PostBillyNair, on 27 September 2020 - 02:53 AM, said:

My team killed 2 or less tanks in 23 of my last 30 games. I am not sure how the matchmaking works for players and their WR, but, I would think that after a while, I would accidentally be put on a team that can shoot a tank. We won 5 of 30, but, there are stretches of 9-12 losses in between those wins... Up until today, I have NEVER lost more than 7 in a row.

 

Some people say MM is working as intended...

 

My question is, do they take WR into account at all or is it purely based off the tanks with no regards to who is driving it?

 

If they dont base it off the players, why dont they?

 

If there is at least 1 line of code looking at players (vs tanks) what is it?

 

As I said, I have never lost more than 7 in a row ever, so, i was under the impression that they had SOME code checking that (loose too many in a row and players start to quit, put those players on decent teams, let them win a few now and then, you will keep them playing... more a matter of player retention than being fair, better for the share-holders)

 

Do they have code to look at WRs or loss-streaks?

you are incredibly lucky never being able to lose more than 7in a row


Raging Dumbo


BillyNair #1470 Posted 27 September 2020 - 09:00 AM

    Junior Sergeant

  • Players
  • 10596 battles
  • 116
  • [MOD3]
  • Member since:
    04-15-2020

View PostTHERebrth, on 27 September 2020 - 08:59 AM, said:

you are incredibly lucky never being able to lose more than 7in a row

I have lost more in Platoons, but, never solo... Today was a first for [EDIT: losing more than 7 as] solo


Edited by BillyNair, 27 September 2020 - 09:01 AM.


_Crusader6_ #1471 Posted 27 September 2020 - 01:00 PM

    Vote Slime Mold

  • Players
  • 70713 battles
  • 15,699
  • [BDGER]
  • Member since:
    12-08-2014

View PostBillyNair, on 27 September 2020 - 03:53 AM, said:

My team killed 2 or less tanks in 23 of my last 30 games. I am not sure how the matchmaking works for players and their WR, but, I would think that after a while, I would accidentally be put on a team that can shoot a tank. We won 5 of 30, but, there are stretches of 9-12 losses in between those wins... Up until today, I have NEVER lost more than 7 in a row.

 

Some people say MM is working as intended...

 

My question is, do they take WR into account at all or is it purely based off the tanks with no regards to who is driving it?

 

If they dont base it off the players, why dont they?

 

If there is at least 1 line of code looking at players (vs tanks) what is it?

 

As I said, I have never lost more than 7 in a row ever, so, i was under the impression that they had SOME code checking that (loose too many in a row and players start to quit, put those players on decent teams, let them win a few now and then, you will keep them playing... more a matter of player retention than being fair, better for the share-holders)

 

Do they have code to look at WRs or loss-streaks?


WoT used to use a dynamic MM method that incorporated recent WR etc for your tank. 
  The idea was that it make for more balanced results and made it tougher on better players and easier on less efficient players. 
 

WG said they removed that from WoT and never introduced it into Blitz.   
  From what WG says, all MM does is:

1) Find tanks in the battle tier (being either +1,0, or -1, 0) 

2) Matches Platoons (keep on mind MM only looks at the highest tier tank in the toon) 

3) Matches team by tank type (no more than +/-1 of each tank type being different in each team) 

4) Speed of Battle forming 

5) Attempts to give new tanks -1,0 MM for first 8-12 game ‘honeymoon’ period. See note 1

6) Attempts to reduce repeatedly being queues on same maps. 

 

WG has stated MM does NOT look at the current WR, overall WR etc.  

 


 

 
Tank Hoarder: 424 tanks in Garage:  435/435 Played Tanks Aced
 
This space for rent until next terrible mechanic ;) 
 
    Wallet Warrior: Loyal Original M60 owner
 
 

 


Rou_Garou #1472 Posted 27 September 2020 - 02:05 PM

    Carriers needed - apply inside

  • Players
  • 34393 battles
  • 1,031
  • [OP-4]
  • Member since:
    05-04-2013

I saw in a stream that they are actively looking at ways to improve MM.  One thing mentioned was using the strength of tanks when comparing to similar type tanks on other team. So you would have OP heavy vs OP heavy in same tier and not OP versus weakest heavy in the tier. I could see some positives if implemented.

 

So in essence, there are discussions and research in WG world.

 

 


Edited by Rou_Garou, 27 September 2020 - 02:05 PM.

 

 

 


_Crusader6_ #1473 Posted 27 September 2020 - 02:25 PM

    Vote Slime Mold

  • Players
  • 70713 battles
  • 15,699
  • [BDGER]
  • Member since:
    12-08-2014

View PostRou_Garou, on 27 September 2020 - 09:05 AM, said:

I saw in a stream that they are actively looking at ways to improve MM.  One thing mentioned was using the strength of tanks when comparing to similar type tanks on other team. So you would have OP heavy vs OP heavy in same tier and not OP versus weakest heavy in the tier. I could see some positives if implemented.

 

So in essence, there are discussions and research in WG world.

 

 


The problem with that is we are getting more and more cookie cutter. 

  My solution would be to balance the tanks not balance by offsetting.  I don’t like the current MM with the tank type balancing as it pretty much forces a singular META to each map.  That just caters to lowest common denominators. 
 

 I would much rather have a 7 v 7 random tank type with one caveat no more than 3 of any one tank type per side. 

 


 

 
Tank Hoarder: 424 tanks in Garage:  435/435 Played Tanks Aced
 
This space for rent until next terrible mechanic ;) 
 
    Wallet Warrior: Loyal Original M60 owner
 
 

 


Hye_yeon_Cho #1474 Posted 27 September 2020 - 02:36 PM

    Sergeant

  • Players
  • 954 battles
  • 440
  • [4_EVA]
  • Member since:
    09-22-2019
Its kind of unfair when tanks two tiers lower see tanks two tiers higher. For example, a kv1s can one shot an m8a1, or sometimes a deathstar can one shot tier 8s.

BAR_001 #1475 Posted 27 September 2020 - 04:23 PM

    Mechanized com poster

  • Players
  • 33040 battles
  • 2,527
  • [HAHA]
  • Member since:
    02-18-2017

View Post_Crusader6_, on 27 September 2020 - 06:00 AM, said:


WoT used to use a dynamic MM method that incorporated recent WR etc for your tank. 
  The idea was that it make for more balanced results and made it tougher on better players and easier on less efficient players. 
 

WG said they removed that from WoT and never introduced it into Blitz.   
  From what WG says, all MM does is:

1) Find tanks in the battle tier (being either +1,0, or -1, 0) 

2) Matches Platoons (keep on mind MM only looks at the highest tier tank in the toon) 

3) Matches team by tank type (no more than +/-1 of each tank type being different in each team) 

4) Speed of Battle forming 

5) Attempts to give new tanks -1,0 MM for first 8-12 game ‘honeymoon’ period. See note 1

6) Attempts to reduce repeatedly being queues on same maps. 

 

WG has stated MM does NOT look at the current WR, overall WR etc.  

 

 

 Interesting. That explains the lop sided games. Maybe you are, but I am not talented enough to carry a 3 potato or 4 potato team vs reds with maybe one lame duck contributor and 5 or 6 decent to VG players. Most players are NOT that good. I am on record as stating that I dont think MM is rigged, on the contrary I dont think it does enough to provide a balance. Your post bears that out.

 I completely agree with THERebrth. I experience considerable WR swings like I know lots of other players do, it really does come down to "dumb luck". Most of the time I feel like I have zero control over my own WR. 

 I still think the fix for this issue could be simple. Instill into the game WR achievement requirements for advancing up the tiers. In other words, stop allowing people who really belong in tier 4 or 5 into tier 9 - 10 games! They really shouldnt be there in the first place. But WG will probably never do that. They would rather let us complain then upset potential $$ contributors by putting them in their rightful place. 


Edited by BAR_001, 27 September 2020 - 04:38 PM.

RNG doesn't bother hackers.

_Crusader6_ #1476 Posted 27 September 2020 - 04:43 PM

    Vote Slime Mold

  • Players
  • 70713 battles
  • 15,699
  • [BDGER]
  • Member since:
    12-08-2014

View PostHye_yeon_Cho, on 27 September 2020 - 09:36 AM, said:

Its kind of unfair when tanks two tiers lower see tanks two tiers higher. For example, a kv1s can one shot an m8a1, or sometimes a deathstar can one shot tier 8s.


While we no longer have +/-2, so now the only way anyone sees a +2 tank is they failtoon into a higher tier game, but in +/-2 never had tier 4 see tier 6.   
  Tier 5 was the first tier that saw +2, and so it was a +2/-1 tier, it and tier 6 where tiers that required a lot of learning as they bother had +2/-1.  Tier 7-9 where the only true +/-2 tiers, as 9 was +1/-2 and X was 0/-2 

 

The beauty in +/-2 was there was way more diversity in tanks - but it was definitely not fun as a stock -2 in a tier 8-10 game. 
 

 

 

 


 

 
Tank Hoarder: 424 tanks in Garage:  435/435 Played Tanks Aced
 
This space for rent until next terrible mechanic ;) 
 
    Wallet Warrior: Loyal Original M60 owner
 
 

 


VonTink #1477 Posted 29 September 2020 - 11:37 PM

    Senior Sergeant

  • Players
  • 26083 battles
  • 580
  • [MONTE]
  • Member since:
    12-05-2015

30 games in a Type 59- was bottom tiered 87% of the time.

30 games in a Chimera- was bottom tired 32% of the time.

 

Same tier.  

 

Yeah, MM is "random".  Only someone illiterate in basic statistics would think that, or your typical forumer.  Same thing.  


Edited by VonTink, 29 September 2020 - 11:38 PM.


_Crusader6_ #1478 Posted 30 September 2020 - 01:30 AM

    Vote Slime Mold

  • Players
  • 70713 battles
  • 15,699
  • [BDGER]
  • Member since:
    12-08-2014

View PostVonTink, on 29 September 2020 - 06:37 PM, said:

30 games in a Type 59- was bottom tiered 87% of the time.

30 games in a Chimera- was bottom tired 32% of the time.

 

Same tier.  

 

Yeah, MM is "random".  Only someone illiterate in basic statistics would think that, or your typical forumer.  Same thing.  


Do you realize the # of battles it requires in a tank to get a valid sample set -- hint its significantly more than 100


 

 
Tank Hoarder: 424 tanks in Garage:  435/435 Played Tanks Aced
 
This space for rent until next terrible mechanic ;) 
 
    Wallet Warrior: Loyal Original M60 owner
 
 

 


CommanderSean_2014 #1479 Posted 30 September 2020 - 03:23 AM

    Yukari Ma Best Waifu

  • Players
  • 13371 battles
  • 264
  • [RUDY]
  • Member since:
    12-15-2014

Just love when a stock kv3 gets matched up against a smasher.

Or a is3 against a 252u


Edited by CommanderSean_2014, 30 September 2020 - 03:23 AM.

   A   Proud        About Taiwan =>https://www.taiwan.gov.tw/about.php

 Taiwanese     Taiwan wasn't, and never would be a part of CHINA

1011921428

NON-LOS ATGMS ARE A HORRIBLE MECHANIC  

 

 

whatzup22 #1480 Posted 30 September 2020 - 07:14 AM

    Sergeant

  • Players
  • 3038 battles
  • 423
  • [FOIL]
  • Member since:
    03-25-2017

View PostCommanderSean_2014, on 30 September 2020 - 11:23 AM, said:

Just love when a stock kv3 gets matched up against a smasher.

Or a is3 against a 252u

Unless the stock KV-3 and IS-3 are on your team






1 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users