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

Rejoice Skill Based MM is Here!!!

64 comments in this topic

Recommended Posts

2,639
[SALVO]
Members
4,533 posts
20,849 battles

Seen a lot of talk this last year and some before that the match making needs to be based on skill levels and balance between the teams.   Of course if they did this it wouldn't really be random but hey who doesn't like the deck stacked?

 Constant blowouts, steamrolls,  and whatever conspiracy theory to explain why games are bad anymore.

 Anyways beginning this year I installed a MM monitor and been looking at the games a little closer.   As funny as it may sound I've noticed most of my games are balanced.  

Wut?

 Yes the PR and WR of each team I've noticed is very closely balanced. 

2020-02-17_5_LI.thumb.jpg.972aee93cc10020ef157b68bcd5456e0.jpg

2020-02-17_4_LI.thumb.jpg.1e242edd0874b6d0ee0e89c76f14c947.jpg

2020-02-17_3_LI.thumb.jpg.dfd63a2331cc798ab67787e4f0ea7bc5.jpg

 Granted it's a small sample size just the games I played yesterday but again I've been watching this for almost 2 months now and noticed most games are very close to same skill levels. 

 So for the players at believe WG needs to do a skilled MM well I think they've been playing with the MM quite a bit lately but being skill based isn't the problem. 

   These were all primetime numbers So the matchmaker had plenty of players to fill and I'd be happy to add more screenshots as days go on.

 But rejoice your SBMM is already here!  So you can quit blaming the MM for your problems and look elsewhere for the real issue why games stink.

 

 

 

  • Cool 9
  • Boring 5

Share this post


Link to post
Share on other sites
911
[ANKER]
Members
1,417 posts
11,457 battles

Speak for yourself.  I've have been doing the exact same thing and of my 10 games yesterday 7 had avg WR rates of 3% or greater.  All of the 4% ones were total stomps.

WR doesn't mean everything you say?  Well, I agree, but when you see big avg damage and avg xp disparities then that does mean something.... kinda like this game....  I had a cleveland and bismark supporting me taking a cap while I smoked up.  I actually texted them to turn hydro but neither did until too late so now it like 5 on 1.... was over inside of 4 minutes.

image.thumb.png.e699ec120c8f77232fe86a8e37c70d0f.png

Edited by YouSatInGum
  • Cool 1

Share this post


Link to post
Share on other sites
911
[ANKER]
Members
1,417 posts
11,457 battles
26 minutes ago, NoLoveForPhatShips said:

 Granted it's a small sample size just the games I played yesterday .....

 

 

 

Even if it only happens 1/3 of the time (which probably isn't all that far off) that is still far too often for a game to be basically predetermined.

Share this post


Link to post
Share on other sites
2,639
[SALVO]
Members
4,533 posts
20,849 battles
14 minutes ago, YouSatInGum said:

Even if it only happens 1/3 of the time (which probably isn't all that far off) that is still far too often for a game to be basically predetermined.

 Well MM can only do so much with limited time and players available

 

But my understanding is this used to be skewed by 10% or more prior think they've done alot to bring it closer lately

Share this post


Link to post
Share on other sites
911
[ANKER]
Members
1,417 posts
11,457 battles
3 minutes ago, NoLoveForPhatShips said:

 Well MM can only do so much with limited time and players available

 

But my understanding is this used to be skewed by 10% or more prior think they've done alot to bring it closer lately

I doubt that.  We are talking about averages here.  10% would probably be like top 5% ALL being on one team.  The odds of that are probably 1 in 500 or greater, or in other words very uncommon. 

What I am saying is that at least 1/3 of the time, the outcome of the match can be predicted within a certain confidence interval (probably 95% or greater).

 

The thing is, it is easy to fix.  Usually all it takes is swapping 2 or 3 players from one side to the other and things are competitive again.

  • Cool 4

Share this post


Link to post
Share on other sites
Banned
259 posts
5,736 battles
1 hour ago, NoLoveForPhatShips said:

But rejoice your SBMM is already here!

 

You must be mistaken cause dozens and dozens of forum users here and on EU explained how that is completely impossible to achieve, and even to attempt it would completely blow up the queueing times.

Share this post


Link to post
Share on other sites
381
[BB35]
[BB35]
Members
506 posts
14,354 battles

MM still dealing out bad teams. Somedays I cant win a fight to save my life. And other days I cant lose a fight.

I gave up looking at W/L rate because you cant carry a team if all your teammates run, go afk, hide in the bottom corner of the map, etc...

  • Cool 1

Share this post


Link to post
Share on other sites
2,639
[SALVO]
Members
4,533 posts
20,849 battles

Granted there are a few outliers but for the most part the MM is trying to put together the most competitive teams

sure it's not perfect I don't think it could be with Random humans but the games are definitely not random 

 And I really don't think they're trying to skew it just because of any reasons

 

Edited by NoLoveForPhatShips

Share this post


Link to post
Share on other sites
911
[ANKER]
Members
1,417 posts
11,457 battles
18 minutes ago, FixCVs said:

You must be mistaken cause dozens and dozens of forum users here and on EU explained how that is completely impossible to achieve, and even to attempt it would completely blow up the queueing times.

My sarcasm detector is still broken so can't tell....

But it would not add any queue time except for a few milliseconds for additional server calculations.  Most unbalanced teams can be evened out by swapping 2 or 3 players within their respective ship classes.

  • Cool 1

Share this post


Link to post
Share on other sites
10,391
[INTEL]
Members
13,459 posts
37,747 battles
1 hour ago, NoLoveForPhatShips said:

Seen a lot of talk this last year and some before that the match making needs to be based on skill levels and balance between the teams.   Of course if they did this it wouldn't really be random but hey who doesn't like the deck stacked?

 Constant blowouts, steamrolls,  and whatever conspiracy theory to explain why games are bad anymore.

 Anyways beginning this year I installed a MM monitor and been looking at the games a little closer.   As funny as it may sound I've noticed most of my games are balanced.  

Wut?

 Yes the PR and WR of each team I've noticed is very closely balanced. 

 Granted it's a small sample size just the games I played yesterday but again I've been watching this for almost 2 months now and noticed most games are very close to same skill levels. 

 So for the players at believe WG needs to do a skilled MM well I think they've been playing with the MM quite a bit lately but being skill based isn't the problem. 

   These were all primetime numbers So the matchmaker had plenty of players to fill and I'd be happy to add more screenshots as days go on.

 But rejoice your SBMM is already here!  So you can quit blaming the MM for your problems and look elsewhere for the real issue why games stink.

Haha. Three games. Most games remained stacked on one side. MM even pits two divisions against one, and I haven't been in a game in ages where the purples were split. 

Even worse is that the devs have made no attempt to match CVs, so four games tonight the CVs we were given had win rates of 34%, 34%, 45% and 49%. No need to explain what the outcome was, we were able to overcome that handicap only once. The 34% moron had 7500 games and had to know he sucked.  

I have no idea how people can be that selfish....

Share this post


Link to post
Share on other sites
Members
6,900 posts
21,408 battles

Those games show me that at a minimum at least 33% of all players have PR's in the 600's or lower at high tier play. I had a thread shut down because although the moderator admitted to me that it was funny, it was also citing unrest. I was challenged to run my Yamato in circles using only auto pilot to make the circles. I went 4-0.

 

Assuming all games are as close (MM WR) as the ones you show, it simply allows us to determine the steamroll. Which terrible players make mistakes the earliest. One of the dd's is averaging less than 1000 damage in 4 games. that means he has yet to land a torpedo and has not landed more than 4 shells per game. 

And I wonder why I am always chat banned…...

Share this post


Link to post
Share on other sites
237
[TIGRB]
Members
235 posts
2,640 battles
1 hour ago, NoLoveForPhatShips said:

Seen a lot of talk this last year and some before that the match making needs to be based on skill levels and balance between the teams.   Of course if they did this it wouldn't really be random but hey who doesn't like the deck stacked?

 Constant blowouts, steamrolls,  and whatever conspiracy theory to explain why games are bad anymore.

 Anyways beginning this year I installed a MM monitor and been looking at the games a little closer.   As funny as it may sound I've noticed most of my games are balanced.  

Wut?

 Yes the PR and WR of each team I've noticed is very closely balanced. 

2020-02-17_5_LI.thumb.jpg.972aee93cc10020ef157b68bcd5456e0.jpg

2020-02-17_4_LI.thumb.jpg.1e242edd0874b6d0ee0e89c76f14c947.jpg

2020-02-17_3_LI.thumb.jpg.dfd63a2331cc798ab67787e4f0ea7bc5.jpg

 Granted it's a small sample size just the games I played yesterday but again I've been watching this for almost 2 months now and noticed most games are very close to same skill levels. 

 So for the players at believe WG needs to do a skilled MM well I think they've been playing with the MM quite a bit lately but being skill based isn't the problem. 

   These were all primetime numbers So the matchmaker had plenty of players to fill and I'd be happy to add more screenshots as days go on.

 But rejoice your SBMM is already here!  So you can quit blaming the MM for your problems and look elsewhere for the real issue why games stink.

 

 

 

Obviously a White Knight who thinks if he defends  WG he will get treated better in game.

Edited by SkullDuggery05
  • Cool 1

Share this post


Link to post
Share on other sites
1,375
[R-F]
Members
1,787 posts
10,858 battles

I'm curious... how did those three battles go?  Were they tight contests, or were they blowouts in spite of the even teams?

Share this post


Link to post
Share on other sites
2,639
[SALVO]
Members
4,533 posts
20,849 battles
2 minutes ago, DrHolmes52 said:

I literally spit coffee when I read it.

  Lol Me Too! That got me to genuinely bust out laughing

    Might have to change my name

Edited by NoLoveForPhatShips

Share this post


Link to post
Share on other sites
911
[ANKER]
Members
1,417 posts
11,457 battles

One more point I want to make... even when you do have a close game, if one side has a significantly better CV that will usually make a big difference in the outcome.  (Let me say this is not such much a MM problem)

In the game I just had, it was somewhat even in skill, and the red team had a cap advantage, but only a 1 ship advantage (5 vs. 6)  I was the sole DD and they had 3 dd.  Their CV was smart and kept spotting me and making me go evasive, keeping me from from doing a lot of damage or trying to cap.  Our CV, was doing bombing attacks on a couple of ships on the edge of the map, so instead of killing the 2 dd's that only had 1/3 left (or at least making them dodge stuff to give me freedom of movement). 

So when I see a big difference CV stats, it is a complete "here we go again" moment.

Edited by YouSatInGum

Share this post


Link to post
Share on other sites
2,854
[PVE]
Members
4,828 posts
21,205 battles
2 hours ago, NoLoveForPhatShips said:

But rejoice your SBMM is already here!  So you can quit blaming the MM for your problems and look elsewhere for the real issue why games stink.

OK, what is skill?    Wins?  Damage?   No, this game can't even calculate a metric of skill because it's really "Combat Effectiveness (CE)" that you need to run a SBMM.... 

And, since we can't even capture all of the game modes into a single stat sheet, how would you calculate CE????  Randoms only?  Clan Battles only?   And, ask yourself: to what end????  Why even have a SBMM for a small map, time compressed young adult FFA shooter???   Have you ever seen a "Leader Board" that runs 24/7/365 for the entire game?   This isn't a game based on being a meritocracy for heavens sake!  You can 'buy in' at the highest levels with zero skills and a good credit card.......

Sorry, a SBMM just isn't needed.....

  • Boring 1

Share this post


Link to post
Share on other sites
159
[KSC]
Volunteer Moderator
210 posts
22,692 battles

From a few years' worth of experience running Matchmaking Monitor, I wouldn't be surprised if there has been some tweaking done around win rate between teams in the MM.  Months/years ago it wasn't uncommon to have one team with a sub 46% WR and the other with a 55-60% WR.  Lately it seems like the averages are less than 4-5% spread, and the ship win rates are very close.

 

I've attached the match I just came out of.

 

The hilarious part was that someone was actually ranting in chat about how WG unfairly stacks teams with the good players on one side and the bad players on the other side.  I didn't really know how to break it to him.......351093383_2020-02-18(2).thumb.png.05c1a28114138c8c6d99892a22405793.png

Share this post


Link to post
Share on other sites
Members
224 posts
8,704 battles

If MM/team balance were somehow perfect then everyone's win rates would go to 50% - a coin toss - correct? If that is true then every 100-200 games you are likely to have an 8 game losing streak.

Share this post


Link to post
Share on other sites
2,639
[SALVO]
Members
4,533 posts
20,849 battles
2 hours ago, Brhinosaurus said:

I'm curious... how did those three battles go?  Were they tight contests, or were they blowouts in spite of the even teams?

 They were all really close

Share this post


Link to post
Share on other sites
2,639
[SALVO]
Members
4,533 posts
20,849 battles
1 hour ago, Asym_KS said:

OK, what is skill?    Wins?  Damage?   No, this game can't even calculate a metric of skill because it's really "Combat Effectiveness (CE)" that you need to run a SBMM.... 

And, since we can't even capture all of the game modes into a single stat sheet, how would you calculate CE????  Randoms only?  Clan Battles only?   And, ask yourself: to what end????  Why even have a SBMM for a small map, time compressed young adult FFA shooter???   Have you ever seen a "Leader Board" that runs 24/7/365 for the entire game?   This isn't a game based on being a meritocracy for heavens sake!  You can 'buy in' at the highest levels with zero skills and a good credit card.......

Sorry, a SBMM just isn't needed.....

 Well I'm not going down the dark hole of how they rate skill but think it's obvious that they are using some sort of SBMM

  • Boring 1

Share this post


Link to post
Share on other sites
2,639
[SALVO]
Members
4,533 posts
20,849 battles
1 hour ago, Borla78 said:

From a few years' worth of experience running Matchmaking Monitor, I wouldn't be surprised if there has been some tweaking done around win rate between teams in the MM.  Months/years ago it wasn't uncommon to have one team with a sub 46% WR and the other with a 55-60% WR.  Lately it seems like the averages are less than 4-5% spread, and the ship win rates are very close.

 

I've attached the match I just came out of.

 

The hilarious part was that someone was actually ranting in chat about how WG unfairly stacks teams with the good players on one side and the bad players on the other side.  I didn't really know how to break it to him.......351093383_2020-02-18(2).thumb.png.05c1a28114138c8c6d99892a22405793.png

 Need more of the bottom page were you in a  division?

Share this post


Link to post
Share on other sites
159
[KSC]
Volunteer Moderator
210 posts
22,692 battles
1 minute ago, NoLoveForPhatShips said:

 Need more of the bottom page were you in a  division?

Already ditched the original picture.  But no, I was solo.  There were no divs in that particular game.  

  • Cool 1

Share this post


Link to post
Share on other sites
Sign in to follow this  

  • Recently Browsing   0 members

    No registered users viewing this page.

×