Jump to content
You need to play a total of 20 battles to post in this section.
mechengrmike

Lopsided Matchmaking

45 comments in this topic

Recommended Posts

7
[DALE]
Members
8 posts
8,957 battles

I know this comes up pretty regularly, but lopsided roffle-stomps seem to happen more and more often.  Matchmaking must include inexperienced players with veteran players, I have no problems with that.  However, loading up one side of the match with inexperience and the other side with veteran players is unfair.  

Matchmaking Monitor is a tool for viewing player stats while in game.  Player Rating is controversial and may not be the best option for basing matchmaking.  However, player Karma is solely based on player feedback.  If you look at player's Karma rating compared to their stats, you will see a direct correlation.  

I don't like to complain without making a suggestion, so here is mine.....I propose WG equalize matchmaking by organizing players not only by ship type, but also by Karma.  Below is a screenshot of a lopsided roffle-stomp.  The opposing team won in record fashion and had a cumulative Karma of 524.  The team team that lost had a Karma of 85.  

Karma equalizing would have made a huge difference.  

Thoughts?

 

BadMatchmaking_11-1-20.jpg

  • Cool 2
  • Meh 4

Share this post


Link to post
Share on other sites
1,924
[PVE]
Members
6,920 posts
22,666 battles
16 minutes ago, mechengrmike said:

I know this comes up pretty regularly, but lopsided roffle-stomps seem to happen more and more often.  Matchmaking must include inexperienced players with veteran players, I have no problems with that.  However, loading up one side of the match with inexperience and the other side with veteran players is unfair.  

Matchmaking Monitor is a tool for viewing player stats while in game.  Player Rating is controversial and may not be the best option for basing matchmaking.  However, player Karma is solely based on player feedback.  If you look at player's Karma rating compared to their stats, you will see a direct correlation.  

I don't like to complain without making a suggestion, so here is mine.....I propose WG equalize matchmaking by organizing players not only by ship type, but also by Karma.  Below is a screenshot of a lopsided roffle-stomp.  The opposing team won in record fashion and had a cumulative Karma of 524.  The team team that lost had a Karma of 85.  

Karma equalizing would have made a huge difference.  

Thoughts?

 

BadMatchmaking_11-1-20.jpg

Maybe before the rework but many have zero karma based on nothing more than running rework CVs.

(Sorry...not trying to turn this into yet another CV thread but it is a significant factor in the equation unfortunately).

Also...just 1 of those players has 350 of that karma so how would you balance him in that example of players?

Putting all of the zero karma players on the team w/just him will still not balance out the karma w/all of the positive karma players on both teams combined...but it is interesting that not even 1 of the 4 zero karma players is on his team.

Edited by IfYouSeeKhaos

Share this post


Link to post
Share on other sites
6,270
[WORX]
Members
11,407 posts
19,183 battles
18 minutes ago, mechengrmike said:

Thoughts?

IMO, Stop using the "potato" mod and just play the match as best you can...

WOWS is a win one side, need to be a lost on the other side type game. No game are the same because its random. Failure to adapt to the situation only indicates your fallacy as a human being.

On the math side, I post it in a recent post (without pointing out the applied math and laws pertaning to WOWS MM mech). Here

 

  • Boring 1

Share this post


Link to post
Share on other sites
1,924
[PVE]
Members
6,920 posts
22,666 battles
6 minutes ago, black_hull4 said:

MM in Random Battles is random. 

WHY IS THAT SO HARD TO UNDERSTAND?! :etc_red_button:

What makes you think OP (or anybody else that starts a thread about the MM subject) doesn't understand that?

Seems pretty obvious they do & are requesting that changes.

Not saying I agree w/them but your post is definitely not anything new added to the subject...just stating the obvious that everybody obviously understands already.

  • Thanks 2

Share this post


Link to post
Share on other sites
8,218
[GWG]
[GWG]
Supertester
27,128 posts
14,728 battles
30 minutes ago, mechengrmike said:

I know this comes up pretty regularly,

This is a day ending in y subject but you did manage to pick on a new stat to complain about. However, karma is the most meaningless stat because it is totally anonymous so people down vote people for the ship they play or for actually doing well.

  • Cool 1
  • Boring 1

Share this post


Link to post
Share on other sites
66
[HARU]
Beta Testers
117 posts
6,445 battles

Sorry but this screen shot is a terrible  example for OPs argument. Karma does not have any real correlation in your picture, yes one purple has great karma the other three on his team have a TOTAL of 8(5,2,1). Their is a red on the other side that has karma of 10. No correlation with karma at all. I am sure you could have picked a better example so using this is just sad.

Face it, skill based might help things a little but we would still have blow outs because of leverage. Skill based still could not compensate for a purple in a DD verses a purple in a BB and compensating for divisions gets very difficult.

Share this post


Link to post
Share on other sites
SuperTest Coordinator, Beta Testers
6,459 posts
11,560 battles

I don't know how karma relates to skill, or how it can be normalized. I doubt the person with 350 Karma could handle 5 people with 70 karma alone. He's basically an outlier.

Share this post


Link to post
Share on other sites
1,706
[RKLES]
[RKLES]
Beta Testers, In AlfaTesters
6,245 posts
21,670 battles
1 hour ago, black_hull4 said:

MM in Random Battles is random. 

WHY IS THAT SO HARD TO UNDERSTAND?! :etc_red_button:

Because, when I lose...

The other team has better stats and basically wins before the match starts...

If I win, it's not because I was carried.....

......

Seriously, 

If you take out the effect divisions have, MM is basically even when you compare 100 battles...

The thing MM cannot know is:

Unicom getting detonated. 

That 500 wtr shimmy driver actually landing his first salvos...

 

Share this post


Link to post
Share on other sites
Members
3,429 posts
6,898 battles

I read the damages and amount of battles played for each player and plugged them into the spreadsheet.   I could be wrong but I think the numbers cn speak for themselves as far as if MM would use those numbers to try to balance out the team roster.

Left left side has an average of DAMAGES of 36800.83 as opposed to the right team of 55862.92.   Obviously right team can output quite a bit more damage using the current numbers

Left team amount of battles played is on average 46.67 as opposed to right side 84.67.  The right team as played almost twice the amount of battles as you can see...

Now I want to plug in more numbers to make a comparison and get to "See" in data form what we have here. I am thinking that MM could sort some field that would be the criteria to Balance out the teams so that it is as even as possible.....Sorting multiple fields would be better but I don't know if that were so it would slow down the MM setups.. What does anyone think about this?   Plug in your numbers and come to your conclusions. I'm not saying im right and even could be totally wrong but hey..its worth looking into.....

Share this post


Link to post
Share on other sites
8,218
[GWG]
[GWG]
Supertester
27,128 posts
14,728 battles
58 minutes ago, Highblooded said:

Sorry but this screen shot is a terrible  example for OPs argument. Karma does not have any real correlation in your picture, yes one purple has great karma the other three on his team have a TOTAL of 8(5,2,1). Their is a red on the other side that has karma of 10. No correlation with karma at all. I am sure you could have picked a better example so using this is just sad.

Face it, skill based might help things a little but we would still have blow outs because of leverage. Skill based still could not compensate for a purple in a DD verses a purple in a BB and compensating for divisions gets very difficult.

Skill based doesn't work but skill balancing could but a stat still needs to be chosen and there are many factors that could inflate, make someone look better, or deflate, make someone look worse, for any stat chosen.

Share this post


Link to post
Share on other sites
843
[HINON]
[HINON]
Members
2,295 posts
22,163 battles

Karma it's not a great indication of much.   I have 34x.  It goes up at a slightly higher rate then it down.

That said,  I have my good ships and bad ships,  good rng days and bad rng days. 

It all balances out. 

Share this post


Link to post
Share on other sites
4,110
[TARK]
Members
7,331 posts
2,754 battles
2 hours ago, black_hull4 said:

MM in Random Battles is random. 

WHY IS THAT SO HARD TO UNDERSTAND?! :etc_red_button:

...because it's NOT random.

It follows a collection of highly specific rules to ensure a somewhat 'balanced' match...in that balance is defined as a certain number of classes of ships...plus tiers...plus divisions.

That is actually a LARGE reduction in the ability of the MM to ACTUALLY be random.

Plus, we don't know (but suspect) that WG would use some of the ideas outlined in their patent...which means MM is part of the monetization strategy in some way.

THAT is why it is 'hard to understand that random battles are random'...because they are NOT random.

  • Cool 1
  • Boring 1
  • Meh 1

Share this post


Link to post
Share on other sites
4,536
[WOLF9]
Privateers
14,824 posts
4,733 battles
3 hours ago, mechengrmike said:

I know this comes up pretty regularly,

230554783_thisthreadagain.jpg.b21957734474cc4e0d3c10f1656bfb0b.jpg

  • Cool 1

Share this post


Link to post
Share on other sites
9,735
[SALVO]
Members
25,457 posts
27,411 battles
3 hours ago, black_hull4 said:

MM in Random Battles is random. 

WHY IS THAT SO HARD TO UNDERSTAND?! :etc_red_button:

A lot of people don't like it.

WHY IS THAT SO HARD TO UNDERSTAND?!

  • Haha 1

Share this post


Link to post
Share on other sites
9,735
[SALVO]
Members
25,457 posts
27,411 battles
1 hour ago, Warped_1 said:

Karma it's not a great indication of much.   I have 34x.  It goes up at a slightly higher rate then it down.

That said,  I have my good ships and bad ships,  good rng days and bad rng days. 

It all balances out. 

I agree.  I'll give the OP points for making a new suggestion (using karma), even if I think that it's a bad one.

The problem is that there are bound to be many very good players who are also salty as all eff and end up with terrible karma ratings.  If Karma was used as a substitute for a skill metric using the assumption that a high karma indicated good players and a low karma indicated weaker players, some less ethical players might intentionally tank their karma rating just so that they could look like they were bad players when in fact they were quite good.

 

Share this post


Link to post
Share on other sites
8,801
[PSP]
[PSP]
Members
13,673 posts
5 hours ago, BrushWolf said:

karma is the most meaningless stat because it is totally anonymous so people down vote people for the ship they play or for actually doing well.

It's hilarious to read in chat why people think that another player should be "reported." One of my favorite conversations went like this.

:Smile-angry:Report :fish_viking: for being a bot.

 :fish_viking:I'm not a bot.

:fish_book:Well, you are still at the spawn point and just sailing around in circles.

:fish_viking:My "Q" key got stuck!

:fish_book: But you just typed "Q" and you still are going in circles.

:fish_viking:I'm eating lunch now.

 

  • Funny 4
  • Haha 1

Share this post


Link to post
Share on other sites
37
[SIT]
Members
74 posts
5,422 battles
Quote

Left team amount of battles played is on average 46.67 as opposed to right side 84.67.  The right team as played almost twice the amount of battles as you can see...

number of battles played = amount of experience a player has. If it were going to try and balance the teams in some additional way, that would be the logical stat to use.

Tonight i was on a serious losing streak and started checking enemy teams profiles out of curiosity.  So many players with more games in a single ship than i have played in total. One example of a player i checked had almost 15,000 games played. 

Edited by Ariochaotik

Share this post


Link to post
Share on other sites
SuperTest Coordinator, Beta Testers
6,459 posts
11,560 battles
1 hour ago, Ariochaotik said:

number of battles played = amount of experience a player has. If it were going to try and balance the teams in some additional way, that would be the logical stat to use.

Tonight i was on a serious losing streak and started checking enemy teams profiles out of curiosity.  So many players with more games in a single ship than i have played in total. One example of a player i checked had almost 15,000 games played. 

Do you believe a player with these stats has more experience than 99% of the player base? He has played more games than me by at least 3-fold.

8EB87D0E-03BA-42A0-B560-3F88B1AD0CC6.thumb.jpeg.aefd1ed6289b54dcbf6b63a6f3b2683d.jpeg
 

If you sort by number of battles played you will find that a fair number of players with more than 10K games have the demonstrable experience of pressing battle, but not combat effectiveness.

Share this post


Link to post
Share on other sites
4,110
[TARK]
Members
7,331 posts
2,754 battles
4 hours ago, Compassghost said:

Do you believe a player with these stats has more experience than 99% of the player base? He has played more games than me by at least 3-fold.

8EB87D0E-03BA-42A0-B560-3F88B1AD0CC6.thumb.jpeg.aefd1ed6289b54dcbf6b63a6f3b2683d.jpeg
 

If you sort by number of battles played you will find that a fair number of players with more than 10K games have the demonstrable experience of pressing battle, but not combat effectiveness.

Most of those are bot accounts...

...something WG really should do more about, but dont.

Share this post


Link to post
Share on other sites
73
[RCNW6]
Members
155 posts
3,505 battles

Correct me if I'm wrong but isn't "karma" completely subjective and easily manipulated; as in, get together with some friends and give them "karma" while they do the same for you. ? Not to mention, I suspect that like me, a lot of players have no idea how to give "karma" to anybody else nor care whether they receive it themselves. 

Given the above how is the OP's proposed system going to improve things with match making? Wouldn't using actual game stats be a better way to even out teams? 

Share this post


Link to post
Share on other sites
2,319
[SALVO]
Members
2,513 posts
6,672 battles
25 minutes ago, Daniel_Allan_Clark said:

Most of those are bot accounts...

...something WG really should do more about, but dont.

There are quite a few people here on the forums with numbers that are either similar or getting there yet they also have thousands of posts on the forums.  Will you be accusing them of being bots too ? 

 

Point being  tens of thousands of matches  with a WR in the low 40s doesn't make them a bot. 

Edited by eviltane

Share this post


Link to post
Share on other sites
Members
857 posts
1,261 battles
58 minutes ago, WaywardVariable said:

as in, get together with some friends and give them "karma" while they do the same for you. 

This is at least something made impossible by game mechanics. Well, not impossible, but you would need to get that 1 out of 10 game where your friends are in the match but not in a division.

  • Thanks 1

Share this post


Link to post
Share on other sites

  • Recently Browsing   0 members

    No registered users viewing this page.

×