Jump to content
You need to play a total of 5 battles to post in this section.
Plaatduutsch

MM Has Been Improving

23 comments in this topic

Recommended Posts

2,764
Members
9,864 posts

Yes, all those divisions on one team are bound to keep things balanced.....:Smile_teethhappy:

  • Cool 1

Share this post


Link to post
Share on other sites
417
[CAST]
Members
1,398 posts
5,815 battles

Let's see....

  • 1 carrier per side.
  • 4 DDs per side.
  • 3 radar vs. 2 radar.
  • tiers are perfectly mirrored.

Where's the issue?

Share this post


Link to post
Share on other sites
44
[SI-YC]
Beta Testers
377 posts
4,290 battles

Back in the day, there were all sorts of MM based hijinx that people had to deal with.  Uptiered carriers, DD vs No DD, lots of BB imbalance. 

 

Sure they could tweak the player skill thing a bit but unicum divisions are always going to be an issue if you allow divisions.  I think we've reached peak MM.

  • Cool 1

Share this post


Link to post
Share on other sites
1,824
[ARRGG]
[ARRGG]
Members
5,770 posts

Hummmmmmmmmmmmm 5 players who have ranked out top 5 on the winning team. Maybe if the other 7 players were added to the red team it still wouldn't be "Balanced".

Share this post


Link to post
Share on other sites
413
[VW]
Members
1,833 posts
12,288 battles
4 hours ago, Harv72b said:

Let's see....

  • 1 carrier per side.
  • 4 DDs per side.
  • 3 radar vs. 2 radar.
  • tiers are perfectly mirrored.

Where's the issue?

Someone didnt learn to count in kindergarten lol!

do you see the ranks (little numbers next to play names?)

Share this post


Link to post
Share on other sites
413
[VW]
Members
1,833 posts
12,288 battles
4 hours ago, Plaatduutsch said:

Strong and balanced

shot-18.06.22_22.25.41-0665.jpg

Defenders of this kind of mm will come here and tell you statistically this is normal and you dont have a large enough sample size to conclude etc etc... 

Edited by monpetitloup
  • Cool 1

Share this post


Link to post
Share on other sites
57
[LOD_2]
Members
263 posts
4 hours ago, Plaatduutsch said:

Strong and balanced

shot-18.06.22_22.25.41-0665.jpg

You realize MM cant predict the future of a match? The two teams seem to be balanced well, get over it.

Share this post


Link to post
Share on other sites
2,409
[PNG]
Supertester
5,657 posts
6,455 battles

It's almost like matchmaking doesn't take skill or ranked performance into account, and good players from good clans will division together OwO.

Share this post


Link to post
Share on other sites
417
[CAST]
Members
1,398 posts
5,815 battles
3 hours ago, monpetitloup said:

Someone didnt learn to count in kindergarten lol!

do you see the ranks (little numbers next to play names?)

Have you ever played ranked?  And if so, how could you possibly think rank has anything to do with individual skill?

  • Funny 2

Share this post


Link to post
Share on other sites
3,634
[INTEL]
Members
8,437 posts
25,312 battles
16 hours ago, Harv72b said:

Have you ever played ranked?  And if so, how could you possibly think rank has anything to do with individual skill?

Because I collected many games of data on the MM, and teams which had at least one player who had achieved Rank 1-5 won 65% of the matches. On the whole, players who can Rank out are both skilled and dedicated, qualities that lead to victory. 

  • Cool 2

Share this post


Link to post
Share on other sites
417
[CAST]
Members
1,398 posts
5,815 battles
7 hours ago, Taichunger said:

Because I collected many games of data on the MM, and teams which had at least one player who had achieved Rank 1-5 won 65% of the matches. On the whole, players who can Rank out are both skilled and dedicated, qualities that lead to victory. 

So the fix would be to ban rank 1-5 players from random battles?  Or to ban rank 1-5 players from joining divisions together?  Or to force such players to wait in the queue until such time as they can be "balanced" between teams?  All so you could look at "many games of data" and see a 50/50 split?

Share this post


Link to post
Share on other sites
3,634
[INTEL]
Members
8,437 posts
25,312 battles
18 hours ago, Harv72b said:

So the fix would be to ban rank 1-5 players from random battles?  Or to ban rank 1-5 players from joining divisions together?  Or to force such players to wait in the queue until such time as they can be "balanced" between teams?  All so you could look at "many games of data" and see a 50/50 split?

Arg miswritten.

At least ONE MORE player Rank 1-5

No, the fix would be to evenly distribute skilled players. 

Share this post


Link to post
Share on other sites
37
[FFLY]
Members
39 posts
5,255 battles
7 minutes ago, Taichunger said:

Arg miswritten.

At least ONE MORE player Rank 1-5

No, the fix would be to evenly distribute skilled players. 

Look at you, trying to use common sense here :)

  • Cool 1

Share this post


Link to post
Share on other sites
1,571
[BIAS]
Members
3,047 posts
9,016 battles
10 hours ago, Harv72b said:

Have you ever played ranked?  And if so, how could you possibly think rank has anything to do with individual skill?

When I first started playing the game I thought that anyone who could get into single digits in ranked was super good. Now that I regularly play in the 5-2 bracket, I realized how wrong I was....

Share this post


Link to post
Share on other sites
1,571
[BIAS]
Members
3,047 posts
9,016 battles
18 hours ago, Plaatduutsch said:

Strong and balanced

wshot-18.06.22_22.25.41-0665.jpg

Wow, thats awful MM, The enemy team has more players that progressed to at least rank 15. 

Share this post


Link to post
Share on other sites
170
[OSG]
Beta Testers
1,349 posts
14,768 battles
24 minutes ago, Taichunger said:

...................

No, the fix would be to evenly distribute skilled players. 

WG has already stated they will not be using a skilled based MM......nor imo should they.  Why would anyone try to improve in a game when no matter what you do you'll eventually end up at a 50% win rate which is what would happen to everyone if a skilled based MM was perfectly done?

  • Cool 1

Share this post


Link to post
Share on other sites
413
[VW]
Members
1,833 posts
12,288 battles
12 hours ago, Harv72b said:

Have you ever played ranked?  And if so, how could you possibly think rank has anything to do with individual skill?

Because everyone who ranked out was carried and all those who dont participate in ranked are the best players? 

Share this post


Link to post
Share on other sites
413
[VW]
Members
1,833 posts
12,288 battles
4 hours ago, Harv72b said:

So the fix would be to ban rank 1-5 players from random battles?  Or to ban rank 1-5 players from joining divisions together?  Or to force such players to wait in the queue until such time as they can be "balanced" between teams?  All so you could look at "many games of data" and see a 50/50 split?

You clearly see that enough rank one players were present. It costs no time for mm to assign them equally to both teams. Ranked players could easily be evenly distributed and nonranked players as well. This would greatly improve match quality and reduce roflstomps.

  • Cool 1

Share this post


Link to post
Share on other sites
417
[CAST]
Members
1,398 posts
5,815 battles
6 hours ago, Taichunger said:

No, the fix would be to evenly distribute skilled players. 

But even ignoring the point made by @Merc85, what's your definition of skill?  Rank?  A lot of good players don't participate in ranked for a multitude of reasons, and a lot of pretty poor players manage to rank out through sheer perseverance.  Do you only consider rank 5 or better, or do you "balance" all the way up through rank 23?  Do you treat a player who ranked out in 80 battles the same as one who did it in 1100 (true story)?  What about a division made up of players with different ranks?  Or with some unranked players?

Before you can advocate for skill-based matchmaking, you need to have a realistic method of determining skill.  And even if they change it so that most recent ranks are perfectly mirrored on each team you still end up with dozens of threads complaining that their team got all the 45% WR rank 1s and the other team had all 60%+.  Or if you base it around win rate, do you weight somebody with a 45% WR over 300 battles the same as someone at 45% over 15k?  Do you go by win rate in that specific ship, or that class of ship, or overall?  Or, for that matter, what do you do about a player who ranked out exclusively playing cruisers when (s)he queues up in a carrier? 

The matchmaking system currently being used is as good as Warships has ever had, and while you are going to end up with "stacked" teams from time to time, it's going to even out in the long run.  Trying to figure out an accurate way to assess skill in this game, and then trying to implement a matchmaking system that would take that into account, would eat up a ton of time and resources which could be much better spent fixing more pressing balance issues or developing a port UI which runs smoother than a Model T or adding new content or or or.

  • Cool 1

Share this post


Link to post
Share on other sites
3,634
[INTEL]
Members
8,437 posts
25,312 battles
55 minutes ago, Harv72b said:

But even ignoring the point made by @Merc85, what's your definition of skill?  Rank?  A lot of good players don't participate in ranked for a multitude of reasons, and a lot of pretty poor players manage to rank out through sheer perseverance.  Do you only consider rank 5 or better, or do you "balance" all the way up through rank 23?  Do you treat a player who ranked out in 80 battles the same as one who did it in 1100 (true story)?  What about a division made up of players with different ranks?  Or with some unranked players?

Before you can advocate for skill-based matchmaking, you need to have a realistic method of determining skill.  And even if they change it so that most recent ranks are perfectly mirrored on each team you still end up with dozens of threads complaining that their team got all the 45% WR rank 1s and the other team had all 60%+.  Or if you base it around win rate, do you weight somebody with a 45% WR over 300 battles the same as someone at 45% over 15k?  Do you go by win rate in that specific ship, or that class of ship, or overall?  Or, for that matter, what do you do about a player who ranked out exclusively playing cruisers when (s)he queues up in a carrier? 

The matchmaking system currently being used is as good as Warships has ever had, and while you are going to end up with "stacked" teams from time to time, it's going to even out in the long run.  Trying to figure out an accurate way to assess skill in this game, and then trying to implement a matchmaking system that would take that into account, would eat up a ton of time and resources which could be much better spent fixing more pressing balance issues or developing a port UI which runs smoother than a Model T or adding new content or or or.

Someone asked what the fix would be. I answered. I didnt say whether that should happen.

Share this post


Link to post
Share on other sites
232
[SHOOT]
Beta Testers
1,492 posts
8,036 battles
On 6/22/2018 at 11:03 PM, awiggin said:

Yes, all those divisions on one team are bound to keep things balanced.....:Smile_teethhappy:

I think the point he's making is the presence of only 2 radars on the enemy team with 3 on his.

Share this post


Link to post
Share on other sites

  • Recently Browsing   0 members

    No registered users viewing this page.

×