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

Didnt have much of a chance in this one. MM Help!

22 comments in this topic

Recommended Posts

7
[FAP]
Members
42 posts
4,206 battles

What do you think our chances of success were? 

I honestly have no beef with MM other than perhaps the +/- 2, especially now that I know it is a business decision.  Yet, it would be nice to split up the high ranking players a little better than this.  

op 3 players were bwray71 (14,544 battles, 1222 rating), rabidrabit (6,907 battles, 1242 rating), and _teef Dasaltyclown (21,113 battles, 161 rating) with a combined 42K+ games and average rating around 1,2000.

We were so Screwed LOL.  (insert emoji of captain bending over facing away)

In looking at this, it was an honor to lose to them with the top loser score.:cap_win:

 

image.thumb.png.9c88efd043d4bf8e55607d791730ddb9.png

Edited by Hobobeast
  • Cool 1
  • Bad 1

Share this post


Link to post
Share on other sites
720
[NEUTR]
Members
2,207 posts
6,175 battles

Rank 1 means really nothing when players can achieve it with less than 50% win rate.

 

 

Share this post


Link to post
Share on other sites
242
[SF-3]
Members
1,527 posts
8,114 battles
3 minutes ago, NeutralState said:

Rank 1 means really nothing when players can achieve it with less than 50% win rate.

 

 

Not many people get R1 with sub 50%. The issue I think the OP is getting at, there is no skill based MM. Its random mode, and that's what you get, potatoes and unicums together, some times there all on one team, other times there on the team, and some times the teams are evenly balanced.

My suggestion, play randoms to win, grind, unlock, w/e but dont expect it to be a fair fight, unless you plan to bring your on unicum div.

Share this post


Link to post
Share on other sites
Beta Testers
2,029 posts
3 minutes ago, Kevs02Accord said:

 

My suggestion, play randoms to win, grind, unlock, w/e but dont expect it to be a fair fight, unless you plan to bring your on unicum div.

Then it's still unlikely to be a fair fight :Smile-_tongue:

Share this post


Link to post
Share on other sites
242
[SF-3]
Members
1,527 posts
8,114 battles
2 minutes ago, AspiringCodger said:

Then it's still unlikely to be a fair fight :Smile-_tongue:

But in who's favor :cap_hmm:

Share this post


Link to post
Share on other sites
720
[NEUTR]
Members
2,207 posts
6,175 battles
5 minutes ago, Kevs02Accord said:

Not many people get R1 with sub 50%. The issue I think the OP is getting at, there is no skill based MM. Its random mode, and that's what you get, potatoes and unicums together, some times there all on one team, other times there on the team, and some times the teams are evenly balanced.

My suggestion, play randoms to win, grind, unlock, w/e but dont expect it to be a fair fight, unless you plan to bring your on unicum div.

People wishing for "skill based MM" really should consider what they are actually wishing for.

 

Look at your own stats, now imagine being in a team of other 11 people with your stats. Or, for high performing players, imagine if everyone plays just like them.

 

The games would be horrific.

Share this post


Link to post
Share on other sites
3,206
[GWG]
[GWG]
Alpha Tester, In AlfaTesters
15,615 posts
9,014 battles

The only way that skill could be put into the MM without negative effects would be if it was only applied after the 24 players were chosen. In other words shuffle the players back and forth between teams until the factors are as close as possible.

Share this post


Link to post
Share on other sites
717
[SKDSH]
[SKDSH]
Beta Testers, In AlfaTesters
3,981 posts
9,916 battles
1 minute ago, BrushWolf said:

The only way that skill could be put into the MM without negative effects would be if it was only applied after the 24 players were chosen. In other words shuffle the players back and forth between teams until the factors are as close as possible.

 I would agree most definitely with this .  Out of emoji's already Dagnabit !

:Smile_honoring:

Share this post


Link to post
Share on other sites
7
[FAP]
Members
42 posts
4,206 battles

Coming up with a perfect MM is an impossible task.  I really have no issues with the way it is done as it all averages out in the end.  I must say that the greatest correlation I have seen with winning teams is the most ranked players followed closely by the most premium ships.  I just thought this was an interesting battle as the winning team's top 3 players were all Rank 1.  That says something.  

 

Enjoy your weekend.

Share this post


Link to post
Share on other sites
25
[HIT]
Members
69 posts
5,964 battles
11 minutes ago, BrushWolf said:

The only way that skill could be put into the MM without negative effects would be if it was only applied after the 24 players were chosen. In other words shuffle the players back and forth between teams until the factors are as close as possible.

Good point. Of course it wouldn’t work since now the ships themselves can’t be taken into account.

Share this post


Link to post
Share on other sites
242
[SF-3]
Members
1,527 posts
8,114 battles
35 minutes ago, NeutralState said:

People wishing for "skill based MM" really should consider what they are actually wishing for.

 

Look at your own stats, now imagine being in a team of other 11 people with your stats. Or, for high performing players, imagine if everyone plays just like them.

 

The games would be horrific.

I have played numerous skill based MP games, I much prefer skill based MM. My stats are just an average. If we had skill based MM the games should be much closer, as everyone playing would be around the same skill level. Of course you would need brackets like other skill based MM use.

  • Cool 1

Share this post


Link to post
Share on other sites
1,741
[YORHA]
Members
3,325 posts
5,912 battles
27 minutes ago, HQ_21 said:

Good point. Of course it wouldn’t work since now the ships themselves can’t be taken into account.

Sure it could.  WG has volumes of data on each individual and the ships they play.  It would be relatively easy to create a metric out of that data for each player.

By way of ( a very simplified) example you create a metric of between 1 and 10 for the way each player plays each ship over x number of games.  You have 24 players , some at 10, some at 5 and some at 1.  Then the matchmaker juggles these until the teams are as closed to balanced as the available crop will allow.

So you have a group of 24 players in the queue.

4 have 10 ship skill points

1 has 9 ship skill points

3 have 8 ship skill points

4 have 7 ship skill points

6 have 6 ship skill points

5 have 5 ship skill points

1 has 4 ship skill points.

 

Team A                                        Team B

2 @ 10 ssp                                   2 @ 10 ssp 

1 @ 9 ssp                                     0 @ 9 ssp 

1 @ 8 ssp                                     2 @ 8 ssp 

2 @ 7 ssp                                     2 @ 7 ssp 

3 @ 6 ssp                                    3 @ 6 ssp 

2 @ 5 ssp                                      3 @ 5 ssp 

1 @ 4 ssp                                      0 @ 4 ssp 

83 Ship Skill Points               83 Ship Skill Points 

 

I knocked that out in about 10 minutes on a simple spreadsheet.  Having seen some of WG's IT infrastructure I suspect they might be able to do ever better!

Share this post


Link to post
Share on other sites
7
[FAP]
Members
42 posts
4,206 battles

Yeah, but you also have to account for ship types.  Cant have 4 CVs, 4 DDs, and 4 BBs going up against 12 BBs.

 

While you are at it, can you balance out the pink players and jerks as well?

Share this post


Link to post
Share on other sites
1,150
[ARGSY]
Beta Testers, In AlfaTesters
5,178 posts
18,895 battles
6 minutes ago, JCC45 said:

Sure it could.  WG has volumes of data on each individual and the ships they play.  It would be relatively easy to create a metric out of that data for each player.

By way of ( a very simplified) example you create a metric of between 1 and 10 for the way each player plays each ship over x number of games.  You have 24 players , some at 10, some at 5 and some at 1.  Then the matchmaker juggles these until the teams are as closed to balanced as the available crop will allow.

So you have a group of 24 players in the queue.

4 have 10 ship skill points

1 has 9 ship skill points

3 have 8 ship skill points

4 have 7 ship skill points

6 have 6 ship skill points

5 have 5 ship skill points

1 has 4 ship skill points.

 

Team A                                        Team B

2 @ 10 ssp                                   2 @ 10 ssp 

1 @ 9 ssp                                     0 @ 9 ssp 

1 @ 8 ssp                                     2 @ 8 ssp 

2 @ 7 ssp                                     2 @ 7 ssp 

3 @ 6 ssp                                    3 @ 6 ssp 

2 @ 5 ssp                                      3 @ 5 ssp 

1 @ 4 ssp                                      0 @ 4 ssp 

83 Ship Skill Points               83 Ship Skill Points 

 

I knocked that out in about 10 minutes on a simple spreadsheet.  Having seen some of WG's IT infrastructure I suspect they might be able to do ever better!

Captain skillpoints is not equal player skill. They sell premiums with a 10 point captain. If you are right skill can be bought. You also don't care about the ship types they use.

Share this post


Link to post
Share on other sites
25
[HIT]
Members
69 posts
5,964 battles
6 minutes ago, Hobobeast said:

Yeah, but you also have to account for ship types.  Cant have 4 CVs, 4 DDs, and 4 BBs going up against 12 BBs.

This is exactly what I mean. And then it’d still need to account for differences between same type ships. The pool of players in the queue at any given time simply isn’t large enough to make it work.

Share this post


Link to post
Share on other sites
1,497
Beta Testers
6,868 posts
4,189 battles
1 hour ago, NeutralState said:

Rank 1 means really nothing when players can achieve it with less than 50% win rate.

 

 

 

This is true. But look at the top three on the other team. Typhoon, Typhoon, Hurricane.

 

They're not the sad people that play a thousand games in a ranked season to get rank 1.

Share this post


Link to post
Share on other sites
1,497
Beta Testers
6,868 posts
4,189 battles
1 hour ago, BrushWolf said:

The only way that skill could be put into the MM without negative effects would be if it was only applied after the 24 players were chosen. In other words shuffle the players back and forth between teams until the factors are as close as possible.

 

Even that wouldn't have helped in that game.

There were 3 good players, they're in a division.   About 70-80% of the time they're winning that game.

That said, they should.  Absolutely no to skill based MM in any form.  They already have something to try to balance divisions, that's good enough. 

Share this post


Link to post
Share on other sites
38
[CRMSN]
Members
147 posts
12,379 battles
15 minutes ago, Erebthoron said:

Captain skillpoints is not equal player skill. They sell premiums with a 10 point captain. If you are right skill can be bought. You also don't care about the ship types they use.

I don't think he is talking about Captain skills, but how well you, as a player, perform in the ship you have chosen for the given battle.

Share this post


Link to post
Share on other sites
7
[FAP]
Members
42 posts
4,206 battles

top 3 players were bwray71 (14,544 battles, 1222 rating), rabidrabit (6,907 battles, 1242 rating), and _teef Dasaltyclown (21,113 battles, 161 rating) with a combined 42K+ games and average rating around 1,2000.

We were so Screwed LOL.  (insert emoji of captain bending over facing away)

In looking at this, it was an honor to lose to them with the top loser score.:cap_win:

 

Edited by Hobobeast

Share this post


Link to post
Share on other sites
1,741
[YORHA]
Members
3,325 posts
5,912 battles
2 minutes ago, FacePalming said:

I don't think he is talking about Captain skills, but how well you, as a player, perform in the ship you have chosen for the given battle.

 

Thank you.  You get a gold star for reading comprehension.

 

product_73477_1_orig.jpg

Share this post


Link to post
Share on other sites
3,206
[GWG]
[GWG]
Alpha Tester, In AlfaTesters
15,615 posts
9,014 battles

I have bad memories of when WoWP had an undocumented skill factor in the MM and many players would end up waiting a long time to get a match.

Share this post


Link to post
Share on other sites

  • Recently Browsing   0 members

    No registered users viewing this page.

×