Jump to content


WN9? Simple fix for WN8


  • Please log in to reply
73 replies to this topic

o7o7o7o7o7o7o7o7o7o7o7o7 #21 Posted Jul 11 2018 - 17:09

    First lieutenant

  • -Players-
  • 424 battles
  • 865
  • [ITDED] ITDED
  • Member since:
    01-23-2015

View PostNeatoMan, on Jul 11 2018 - 17:07, said:

WN8 already factors in seal clubbing.  That was the a main drive to the WN series; to create a rating system that didn't overly reward seal clubbers.  Currently seal clubbers can still earn high WN8s, but they have to be very good at it...  much better than it takes to earn the same WN8 in the higher tiers

 

really? because i can tell you tier 10 wn is the most brutal one. the problem with low tiers is the lack of damage NOW but it wasnt that way in the past. observe:

 

http://wotlabs.net/na/player/killju

 

you could farm low tiers for days. back when the game wasnt dead anyway



MacDaddyMatty #22 Posted Jul 11 2018 - 17:14

    Major

  • -Players-
  • 28131 battles
  • 6,701
  • [D-DAY] D-DAY
  • Member since:
    10-07-2016

View Postthe_dude_76, on Jul 11 2018 - 10:55, said:

 

Your stats do reflect that, anyone can look to see which tiers you play most frequently.

 

I use WN8 to track my own improvement. Even though it doesn't factor everything into the equation it is still a fair representation. Your idea would make WN8 completely worthless. At the end of the day WN8 is a representation of skill, adjusting for tier negates that entirely. The guy who free exp's past the low tiers in a rush to get to tier ten is going to be over-represented because even though he's going to suck in those tier 10 matches he'll still get the x10 bonus. The guy who does it "right" and learns the game in the lower tiers is going to be underrepresented because, even though he's more skillful than the guy who rushed to tier 10, his WN8 isn't going to reflect that. So what's the point of it then?? If I can make myself look better just by playing higher tier tanks the whole point of the system goes right out the window.

 

In fact nevermind the rerolls, giving players an incentive to skip the lower tiers in a rush to get that x10 bonus to their stats doesn't sound like a good idea.

 

I considered the guy getting 500 WN8 @ tier 10 vs the guy getting 2000 WN8 @ tier 2, hence the metric I came up with.

The hope was a concept that discouraged stat-padding and encouraged - just play the game.


 

This was inspired looking at clan applications.

Some look great on the surface, then you see they have 2000 games at tier 2, and the eyebrows hit the ceiling.

As well, the guy with the meh WN8 who has decent resents in tier 8,9,10s.

The overall WN8 does not seem to reflect this and a very different story emerges when you peel the onion.

But - I'm lazy and don't want a 2nd job when I could be playing the game...

;)


 

Besides, all of this is an exercise, as I doubt any WN developers care what I think.


 

View PostNeatoMan, on Jul 11 2018 - 11:07, said:

WN8 already factors in seal clubbing.  That was the a main drive to the WN series; to create a rating system that didn't overly reward seal clubbers.  Currently seal clubbers can still earn high WN8s, but they have to be very good at it...  much better than it takes to earn the same WN8 in the higher tiers

 

Do you think it's weighted enough?

As I mentioned above, the wave-top view rarely reflect what the player is actually doing IMHO.



Bavor #23 Posted Jul 11 2018 - 17:20

    Major

  • Players
  • 31795 battles
  • 2,980
  • Member since:
    04-21-2013

WN9 is dead due to Wargaming not changing tank IDs when they changed tank tiers.  

 

http://forum.wotlabs.net/index.php?/topic/20669-wn9-candidate-prototype/



Isola_di_Fano #24 Posted Jul 11 2018 - 17:20

    Major

  • Players
  • 22318 battles
  • 3,740
  • Member since:
    11-05-2012

I thought WN9 had been dropped ?

 

And I do not care much about stats, I play tanks I like, mostly WW2 era and a few blueprints tanks, couldn't care less about tier10.

 

Everyone has a different endgame: mine is fun !



MacDaddyMatty #25 Posted Jul 11 2018 - 17:23

    Major

  • -Players-
  • 28131 battles
  • 6,701
  • [D-DAY] D-DAY
  • Member since:
    10-07-2016

View Posto7o7o7o7o7o7o7o7o7o7o7o7, on Jul 11 2018 - 11:09, said:

 

really? because i can tell you tier 10 wn is the most brutal one. the problem with low tiers is the lack of damage NOW but it wasnt that way in the past. observe:

 

http://wotlabs.net/na/player/killju

 

you could farm low tiers for days. back when the game wasnt dead anyway

 

 

SU-26 U.S.S.R. SPG tier: 3

battles: 7193

Total battles:7580

WN8: 4695
 

GREAT GOOGLIE-MOOGLIES!!

:ohmy:


Edited by MacDaddyMatty, Jul 11 2018 - 17:35.


MacDaddyMatty #26 Posted Jul 11 2018 - 17:24

    Major

  • -Players-
  • 28131 battles
  • 6,701
  • [D-DAY] D-DAY
  • Member since:
    10-07-2016

View PostIsola_di_Fano, on Jul 11 2018 - 11:20, said:

I thought WN9 had been dropped ?

 

And I do not care much about stats, I play tanks I like, mostly WW2 era and a few blueprints tanks, couldn't care less about tier10.

 

Everyone has a different endgame: mine is fun !

 

True dat.

 

But as I said - this is for (competitive) clan applications.



Buttknuckle #27 Posted Jul 11 2018 - 17:40

    Major

  • Players
  • 47560 battles
  • 2,587
  • [GOONZ] GOONZ
  • Member since:
    03-19-2013

If you are looking for a wave-top level indicator of a players abilities you should use PR. Its better at that than WN8.



Isola_di_Fano #28 Posted Jul 11 2018 - 17:48

    Major

  • Players
  • 22318 battles
  • 3,740
  • Member since:
    11-05-2012

View PostMacDaddyMatty, on Jul 11 2018 - 11:24, said:

 

True dat.

 

But as I said - this is for (competitive) clan applications.

 

 

Sorry that was not specified in your first post.

 

For those interested in competitive play it might be interesting however good luck on figuring out metrics that everyone will agree with !


Edited by Isola_di_Fano, Jul 11 2018 - 17:52.


StrachwitzPzGraf #29 Posted Jul 11 2018 - 17:58

    First lieutenant

  • -Players-
  • 38006 battles
  • 510
  • [PSYCO] PSYCO
  • Member since:
    01-01-2015

View PostMacDaddyMatty, on Jul 11 2018 - 09:27, said:

I think XVM is toxic to the game (or more specifically WN8 as well as the "chance to win" BS).

 

But I also believe in constructive criticism, not just wailing and gnashing of teeth - so here is my recommendation for you guys to flame me with.

;)

Seriously though, as with all my posts, (constructive) input is encouraged.

 

My recommendation for a stat tracker:

 

- WN8 encourages stat padding / seal clubbing.

Instead of a blanket "performance rating", include the tank's tier into the equation by the tier's factor. 

i.e:

W8 of 1000 at tier 1: 1000 x 1 = 1000.

W8 of 1000 at tier 10: 1000 x 10 = 10000.

 

This will:

- change the range of rating, but

- will more accurately reflect the players activity, and

- discourage seal-clubbing / stat padding lower tiers.


 Also, nuke the "chance to win" BS.

Hey Aslan, get rid of your chance to win mod.

It is NOT accurate and causes undue turmoil.
 

Ok...

I'm ready.

:harp:
 

 

I object - my personal tracking of the WN8 "balance" shows that when a team is as little as 2% different in total WN8 - that difference is 85% accurate in predicting the outcome... so, sure sometimes you lose with a "winning team" and visa-versa ... but 85% accuracy is not exactly "BS".

 

Still game you +1 -- anyone trying to think outside the box to get a better representation of player ability is at least THINKING... and I appreciate that.



MacDaddyMatty #30 Posted Jul 11 2018 - 18:00

    Major

  • -Players-
  • 28131 battles
  • 6,701
  • [D-DAY] D-DAY
  • Member since:
    10-07-2016

View PostStrachwitzPzGraf, on Jul 11 2018 - 11:58, said:

 

I object - my personal tracking of the WN8 "balance" shows that when a team is as little as 2% different in total WN8 - that difference is 85% accurate in predicting the outcome... so, sure sometimes you lose with a "winning team" and visa-versa ... but 85% accuracy is not exactly "BS".

 

Still game you +1 -- anyone trying to think outside the box to get a better representation of player ability is at least THINKING... and I appreciate that.

 

:honoring:

Thanks for the input!



_Kradok_ #31 Posted Jul 11 2018 - 18:08

    Captain

  • -Players-
  • 29585 battles
  • 1,364
  • [_OOF_] _OOF_
  • Member since:
    10-09-2014

Interesting suggestion but remember this is a GAME, and people will play as they see fit.  If I feel like playing NOTHING but tier 3's, I will. "Stat padding" and crap like that doesn't even enter my equation -- MY enjoyment is my own, defined BY ME solely...  I'm a 1400+ overall player, 1800 recents….. but I don't care.  I'm in a decent clan, I pull my weight and outside of clan activity - they don't throw out ANY demands other than enjoy the game, don't screw over the clan and show up when you're needed.

 

 

 

My Su-18 is calling.  :teethhappy:



goldfinger_555 #32 Posted Jul 11 2018 - 18:19

    First lieutenant

  • Players
  • 26861 battles
  • 668
  • [FFWC] FFWC
  • Member since:
    10-17-2013
I'd like to go the other way and remove all stats from the game for 6 months. Than give everyone a survey in their garage (like they do sometimes now) and let them respond to whether they want to continue the game without stats or bring them back. They would still keep a running tally of them incase the folks wanted them back, but they would be hidden.

Devildog8 #33 Posted Jul 11 2018 - 18:25

    Major

  • Players
  • 10547 battles
  • 6,478
  • [XILES] XILES
  • Member since:
    12-26-2011

View PostMacDaddyMatty, on Jul 11 2018 - 08:27, said:

I think XVM is toxic to the game (or more specifically WN8 as well as the "chance to win" BS).

 

But I also believe in constructive criticism, not just wailing and gnashing of teeth - so here is my recommendation for you guys to flame me with.

;)

Seriously though, as with all my posts, (constructive) input is encouraged.

 

My recommendation for a stat tracker:

 

- WN8 encourages stat padding / seal clubbing.

Instead of a blanket "performance rating", include the tank's tier into the equation by the tier's factor. 

i.e:

W8 of 1000 at tier 1: 1000 x 1 = 1000.

W8 of 1000 at tier 10: 1000 x 10 = 10000.

 

This will:

- change the range of rating, but

- will more accurately reflect the players activity, and

- discourage seal-clubbing / stat padding lower tiers.


 Also, nuke the "chance to win" BS.

Hey Aslan, get rid of your chance to win mod.

It is NOT accurate and causes undue turmoil.
 

Ok...

I'm ready.

:harp:
 

 

all aslians has now is a W8 meter which tells you the percentage of W8s which shows the percentage of which team has higher W8 than the other

537dL #34 Posted Jul 11 2018 - 18:33

    Sergeant

  • -Players-
  • 20076 battles
  • 188
  • Member since:
    11-23-2016

stock xvm does not include a chance to win stat

 

there are as many toxic [edited] not using xvm as there are using it

 

you play the game your way, I'll play it mine

everybody wants to be a cop these days


Edited by 537dL, Jul 11 2018 - 18:34.


BillT #35 Posted Jul 11 2018 - 18:45

    Major

  • Beta Testers
  • 27049 battles
  • 4,494
  • [F-3] F-3
  • Member since:
    08-13-2010

View PostMacDaddyMatty, on Jul 11 2018 - 09:27, said:

My recommendation for a stat tracker:

 

- WN8 encourages stat padding / seal clubbing.

Instead of a blanket "performance rating", include the tank's tier into the equation by the tier's factor. 

i.e:

W8 of 1000 at tier 1: 1000 x 1 = 1000.

W8 of 1000 at tier 10: 1000 x 10 = 10000.

 

This will:

- change the range of rating, but

- will more accurately reflect the players activity, and

- discourage seal-clubbing / stat padding lower tiers.

 

 

As already pointed out, this approach really rewards Tier 10 play. You can easily improve your score just by running as much Tier 10 as possible -- even if you play 10% worse than you play at Tier 9, you still get a higher rating.

 

And it's not like Tier 10 is the hardest tier to play. You're never low tier in a Tier 10 tank.  Tier 8 should get the biggest multiplier, IMO. 

 

But most important is an assumption you haven't stated.  What is the purpose of this WNx statistic?  You mentioned clan applications later on, but do you really need a different stat for that? You're already looking at more than Wn8; you're checking average tier, games played in Tier 10s, and probably recent trend in stats.  I don't think you're ever going to create a single stat that tells you everything you want to know about a player.

I bring this up because I don't think a new stat would decrease seal clubbing. I think the good clans have already figured out how to identify and reject seal clubbers, just as you have.  And I think seal clubbers do it mostly because they like to win games, and because they want purple win rate stats.  So no new calculation you add is going to discourage them.   As for stat padders, there too I think the current statistics give clans all the tools they need to weed out poseurs, and people will figure out how to pad any statistic you choose to use.  With your proposal, the way to stat pad is to play the easiest or least popular Tier 10 tank constantly, and spend cash to buy credits so you never have to play lower tier. 

 

I suggest a much simpler way to eliminate the objectionable low-tier stat padding: a WNx stat that ignores everything below tier 6 or even 8, so games played in low-tier tanks just don't count.   Any clan that cares about stats really wants to know what you can do in high tier tanks, right?



MacDaddyMatty #36 Posted Jul 11 2018 - 19:20

    Major

  • -Players-
  • 28131 battles
  • 6,701
  • [D-DAY] D-DAY
  • Member since:
    10-07-2016

View Postgoldfinger_555, on Jul 11 2018 - 12:19, said:

I'd like to go the other way and remove all stats from the game for 6 months. Than give everyone a survey in their garage (like they do sometimes now) and let them respond to whether they want to continue the game without stats or bring them back. They would still keep a running tally of them incase the folks wanted them back, but they would be hidden.

 

Not a bad idea.

I'm also for opt-out, where they are tracked out of game but hidden in game.

 

 

 

View Post537dL, on Jul 11 2018 - 12:33, said:

stock xvm does not include a chance to win stat

 

there are as many toxic [edited] not using xvm as there are using it

 

you play the game your way, I'll play it mine

everybody wants to be a cop these days

 

Play as you like - amen and absolutely.

Not sure what the rest of your post has to do with the thread...


 

View PostDevildog8, on Jul 11 2018 - 12:25, said:

 

all aslians has now is a W8 meter which tells you the percentage of W8s which shows the percentage of which team has higher W8 than the other

 

Yup already addressed.

 



MacDaddyMatty #37 Posted Jul 11 2018 - 19:25

    Major

  • -Players-
  • 28131 battles
  • 6,701
  • [D-DAY] D-DAY
  • Member since:
    10-07-2016

View PostBillT, on Jul 11 2018 - 12:45, said:

 

As already pointed out, this approach really rewards Tier 10 play. You can easily improve your score just by running as much Tier 10 as possible -- even if you play 10% worse than you play at Tier 9, you still get a higher rating.

 

And it's not like Tier 10 is the hardest tier to play. You're never low tier in a Tier 10 tank.  Tier 8 should get the biggest multiplier, IMO.

 

But most important is an assumption you haven't stated.  What is the purpose of this WNx statistic?  You mentioned clan applications later on, but do you really need a different stat for that? You're already looking at more than Wn8; you're checking average tier, games played in Tier 10s, and probably recent trend in stats.  I don't think you're ever going to create a single stat that tells you everything you want to know about a player.

I bring this up because I don't think a new stat would decrease seal clubbing. I think the good clans have already figured out how to identify and reject seal clubbers, just as you have.  And I think seal clubbers do it mostly because they like to win games, and because they want purple win rate stats.  So no new calculation you add is going to discourage them.   As for stat padders, there too I think the current statistics give clans all the tools they need to weed out poseurs, and people will figure out how to pad any statistic you choose to use.  With your proposal, the way to stat pad is to play the easiest or least popular Tier 10 tank constantly, and spend cash to buy credits so you never have to play lower tier.

 

I suggest a much simpler way to eliminate the objectionable low-tier stat padding: a WNx stat that ignores everything below tier 6 or even 8, so games played in low-tier tanks just don't count.   Any clan that cares about stats really wants to know what you can do in high tier tanks, right?

 

Yes, but I'm LAZY.

Not really but it is (IMHO) a lot of work to peel the onion on someone when reviewing a clan application.

I just am looking for an easier way to do it without taking away my gaming time.


 

2nd part got my brain working... and is interesting!

Is a way to filter WN8 by tier?

Filtering out tier 1-5 would save me MUCHO time.


 



SkaerKrow #38 Posted Jul 11 2018 - 19:35

    Captain

  • Beta Testers
  • 17301 battles
  • 1,492
  • Member since:
    09-24-2010
It seems pretty reasonable to me, and reduces the statistical incentive for seal-clubbing.  

spud_tuber #39 Posted Jul 11 2018 - 20:18

    Major

  • Players
  • 53417 battles
  • 6,795
  • Member since:
    08-26-2013

View PostMacDaddyMatty, on Jul 11 2018 - 12:25, said:

*snip*

2nd part got my brain working... and is interesting!

Is a way to filter WN8 by tier?

Filtering out tier 1-5 would save me MUCHO time.


 

Not that I've seen.

 

WoTNumbers might, as the original program had the option just to look at the stats for certain tiers.

 

Clantools I think has also the ability to look at individual tiers, and gives a quick overview of people's 8 and 10 stats. 

 

Lastly, you could talk to silversound about whether he might be willing to add it to the site he's developing.  Or for that matter, contact one of the existing site managers and see what they might be willing to add upon request.



WangOnTheLoose #40 Posted Jul 11 2018 - 20:21

    Major

  • -Players-
  • 26428 battles
  • 2,131
  • [TG] TG
  • Member since:
    09-06-2014

View Postdnaman, on Jul 11 2018 - 10:30, said:

Agree that the stat padding at lower tiers should somehow be penalized.  My average battle tier is a ton higher than some of these unicum wannabes.  I usually don’t play lower unless I’m with my kids or grinding a new line.

 

Also chance to win is garbage and I am actively trying to convince my father to drop that mod...

 

Best way to take care of that is not include anything in Tier I,II, and III.  Problem solved.




3 user(s) are reading this topic

0 members, 3 guests, 0 anonymous users