Jump to content

Recommended Posts

From the look of it, the T49 will still farm WN8 better than any other tank out there as it sits at the sweet spot at tier 5. I average 2.5 kills and 1100 dmg/game in it.

 

Kills: (1240-1040/(5)^0.164)*2.5 = 1103

 

WN7:

Damage: 1100*530/(184*e^(0.24*5)+130) = 787

 

WN8:

Damage: 1100*630/(287*e^(0.203*5)+25) = 848

 

That's almost 2k WN8 before winrate, defense and spotted are even considered.

 

How about instead of basing the rating off current performance indicators, and then having those metrics manipulated by players going after the requisite 'best paying stats', we instead change the goalposts so that WN8 "farming" only further encourages good play. What I mean here, is I think a higher average tier should be encouraged, we've spent a while trying to remove the seal clubbers from the statistics, but why not just make low end play less attractive than high end play?

 

I threw this together in excel, mainly via trial and error to produce the numbers I wanted it to.

 

DAMAGE*(600+(8*(MAX(4,TIER)-1)))/(287*e^(0.203*TIER))

 

BASE = the base damage I used for the tier. The 4 sets of 'skill' are then calculated using base damage * dmg multipler and thrown into the damage only portion of the respective formulas. Here is a table of the quantity of "WNAllu" which is generated from the damage component of the WN formula.

 

0jw1BXU.png

Link to post
Share on other sites

See the issue with WN7 is the non-linearity of the damage curve - specifically, the positive concavity of the damage curve.

 

For example,

 

Suppose I contribute 1000 to my WN7 via damage all the time.

 

I want my average tier to be 7 over 2 games. I have 2 options.

1. Play 2 tier7 games. I do 2108 average damage, and I contribute 1000 to my WN7 via damage.

2. Play 1 tier5 game and 1 tier9 game. I do 1398 in the tier 5 game and 3256 in the tier 9 game. Individually, I contribute 1000 to my WN7. However, now my average tier is still 7, but my average damage is 2327. I now contribute more to my WN7 via damage.

 

~~~~~

 

(i do realize that given WG not really scaling DPG linearly by tiers, it makes it tough to implement a accurate stat-measuring tool)

Link to post
Share on other sites

Playing a Foch 155 sure is "hard".

 

It isn't "hard", but it punishes you harshly if you make even a small mistake.  However, on the other end of things, it rewards you handsomely when the other guy makes a small mistake.

Link to post
Share on other sites

Playing a Foch 155 sure is "hard".

 

 

 

Show me 4k+ and 2 kills+ in your Foch...didn't think so.  It's easy to do mediocre in one.

 

 

Playing a Foch is harder then running around in a vk3601 picking on some noob in a hetzer.

Link to post
Share on other sites

Show me 4k+ and 2 kills+ in your Foch...didn't think so.  It's easy to do mediocre in one.

 

 

Playing a Foch is harder then running around in a vk3601 picking on some noob in a hetzer.

 

I dont have a Foch, auto-loaders break the game and I stopped playing them.

 

And no, its not easy to do "mediocre" in it. Even the server stats show that. 

 

Sure the Obj 268 is almost as guilty, but atleast it cant kill 90% of the Tier X tanks in 10 seconds.

Link to post
Share on other sites

Allurai, I think you are going too high with that scale. Unless we diminish the weight of some other stat that will just result in everyone´s WN raising... and quite a bit too. Gotta be careful with that.

 

Regarding T49, Foch155 discussion, actually ANY tank in which you can average 2.5 kills will be OP and will definitely deviate your WN from the norm. This is true of Foch155, T57, T49, 3601, ISU-152, Obj 268, etc. etc.

 

 

Kraft, you are completely right. We can just make it as good as possible, but can´t overcome the linearity problem.

Link to post
Share on other sites

What concerns me, is when someone who is "average" like me (comparatively to the very skilled players on this forum) decides I want to go up a new tech tree.  I don't have oodles of free-xp sitting there to get past the first 5-6 tiers of a new tree.  So if/when I start a new tree (like say, when the British Arty come out, I plan to go up that tree), I will be starting at tier 2.  That will end up costing me WN7 rating because I am not skipping to tier 6 to be 'in the average'.  

 

 

As for my kills on the FCM and KV-2, where did you get those numbers?  When I am in-game, XVM shows me as 3 kills per avg on the FCM, and 2 kills per average on the KV-2.  Unless XVM is very screwed up.

 

All I can say is "Suck it up". It sucks, but if you don't have a massive amount of games played, starting new trees is going to hurt you. I haven't figured out how to seal club yet myself (The Seals usually club me), so my stats go down whenever I start a new line. You just have to offset shitty performance when you're unfamiliar with excellent performance when you figure out what the hell you're doing. Remember, every unicum out there (other than rerolls) had to fight against their gaming history to make it to unicum status. Why should we be any different?

Link to post
Share on other sites

Allurai, I think you are going too high with that scale. Unless we diminish the weight of some other stat that will just result in everyone´s WN raising... and quite a bit too. Gotta be careful with that.

 

Regarding T49, Foch155 discussion, actually ANY tank in which you can average 2.5 kills will be OP and will definitely deviate your WN from the norm. This is true of Foch155, T57, T49, 3601, ISU-152, Obj 268, etc. etc.

 

 

Kraft, you are completely right. We can just make it as good as possible, but can´t overcome the linearity problem.

 

I've tested the WN8 formula on myself and a few friends of mine and actually got some interesting results.

 

Player 1 (Just over 2100 WN7) -> Loses about 15 points when converted to WN8

Player 2 (1089 WN7) -> Gains 42 points when converted to WN8

Player 3 (1440 WN7) -> Gains 55 points when converted to WN8

Player 4 (2253 WN7) -> Gains 51 points when converted to WN8

Player 5 (2100 WN7) -> Loses 12 points when converted to WN8

Player 6 (1655 WN7) -> Gains 44 points when converted to WN8

 

Both players losing points had defense points over WN8's new 1.4 limit.  Anyone with slightly padded stats due to a high number of defense points seem to get the overall negative effect.  It seems those with more stats based on damage and not over the 1.4 limit were able to get a gain from the change.

 

WN8 certainly reflects the players ability to do damage and kills at higher tiers, but as it has been mentioned many times, it will always have vulnerabilities.  It will always be that way, we are just looking to reflect the "most accurate" stat possible right?

Link to post
Share on other sites

I dont have a Foch, auto-loaders break the game and I stopped playing them.

 

And no, its not easy to do "mediocre" in it. Even the server stats show that. 

 

Sure the Obj 268 is almost as guilty, but atleast it cant kill 90% of the Tier X tanks in 10 seconds.

 

Auto-Loaders break the game?  Really?  You really believe that?

Link to post
Share on other sites

Show me 4k+ and 2 kills+ in your Foch...didn't think so.  It's easy to do mediocre in one.

 

 

Playing a Foch is harder then running around in a vk3601 picking on some noob in a hetzer.

 

 

Lets be intellectually honest... the F155 is BROKEN. We all know it. Lets not get drug into the minutia of arguing weather we should reward tier 10 more or less based on 1 tank. I mean... the T49@Tier5 is awfully good too...it doesn't get hate from the elite because playing the T49 well translates into playing ALL vision control tanks well.

Link to post
Share on other sites

I have questions about the proposed defense mechanics for Win8.

 

* I average 2.59 Defense points already so I'm already being 'penalized'?  What kind of difference would that make in Win8?

 

* Isn't the defense bonus capped at 100 WIN in a game?  It seems as if that's an awful hard thing to 'farm'.

 

I play quite a bit of solo games and quite a bit of medium tanks as well.  As a result I personally have flex saved many games by getting back to our cap (since no one else on team is aware/alive of the fact).  This is simply my play style and I have never farmed defense points.

 

Am I being penalized for having strong map awareness/flex ability or am I missing something?

 

*edit* for the Tier 1-3 thing does it take into account the most played tanks or your highest WIN7 tanks?  I haven't played a whole lot of Tier 1-3 so I'm assuming it won't make much of a difference (Although I love my M2 light and my PZ1 C)

Link to post
Share on other sites

Swami you are hardly getting punished by losing a mere 15 points.  In addition, defense points are easily farmed by almost the exact method you describe in your play style.  Other than you usually allow one side to fall with intent to get the defense points.

However, in all honesty, I find top players to be one of the few to ever be willing to get defense points. Whether it be for how the battle developed and now demands it, or by simply following the major force and flexing when necessary to get those sweet defense points :thumbup:

Link to post
Share on other sites

I'm not sure how it is gonna work and I'm not complaining.  I'm just looking for some answers that's all ;)

 

I'm going to play my style of game no matter how stats are calculated.  I admit I used to grab some extra cap points but gave that up a few months ago.  It felt pretty lame to worry about keeping my eff above 2100.

 

In the end, good players will have good stats.  It's pretty much that simple.

 

 

However, in all honesty, I find top players to be one of the few to ever be willing to get defense points. Whether it be for how the battle developed and now demands it, or by simply following the major force and flexing when necessary to get those sweet defense points :thumbup:

 

I guess I get more D points than other top players due to my majority solo play and extreme lack of faith in my teams.  I ALWAYS assume that they will fail and will fall back to defend the base the second I see a flank fall.  It's obviously situational and sometimes it makes more sense to out cap them but I hate getting stuck in those situations.  I'd rather bear the burden of saving it myself.

 

Link to post
Share on other sites

I've tested the WN8 formula on myself and a few friends of mine and actually got some interesting results.

 

Player 1 (Just over 2100 WN7) -> Loses about 15 points when converted to WN8

Player 2 (1089 WN7) -> Gains 42 points when converted to WN8

Player 3 (1440 WN7) -> Gains 55 points when converted to WN8

Player 4 (2253 WN7) -> Gains 51 points when converted to WN8

Player 5 (2100 WN7) -> Loses 12 points when converted to WN8

Player 6 (1655 WN7) -> Gains 44 points when converted to WN8

 

Both players losing points had defense points over WN8's new 1.4 limit.  Anyone with slightly padded stats due to a high number of defense points seem to get the overall negative effect.  It seems those with more stats based on damage and not over the 1.4 limit were able to get a gain from the change.

 

WN8 certainly reflects the players ability to do damage and kills at higher tiers, but as it has been mentioned many times, it will always have vulnerabilities.  It will always be that way, we are just looking to reflect the "most accurate" stat possible right?

 

Exactly.

 

I have questions about the proposed defense mechanics for Win8.

 

* I average 2.59 Defense points already so I'm already being 'penalized'?  What kind of difference would that make in Win8?

 

* Isn't the defense bonus capped at 100 WIN in a game?  It seems as if that's an awful hard thing to 'farm'.

 

I play quite a bit of solo games and quite a bit of medium tanks as well.  As a result I personally have flex saved many games by getting back to our cap (since no one else on team is aware/alive of the fact).  This is simply my play style and I have never farmed defense points.

 

Am I being penalized for having strong map awareness/flex ability or am I missing something?

 

*edit* for the Tier 1-3 thing does it take into account the most played tanks or your highest WIN7 tanks?  I haven't played a whole lot of Tier 1-3 so I'm assuming it won't make much of a difference (Although I love my M2 light and my PZ1 C)

 

Most played tanks. If it is 3 or 4 or 5 is still under discussion. The normalizer is based on the NUMBER of games played on those most played tanks.

 

Regarding defense, we are trying to measure map awareness and such, but it just doesnt seem to scale linearly above 1.4 def/game. You can be extremely map aware and not have avg def above 1.4, is kinda the way it goes.

Link to post
Share on other sites

Thanks for the clarification.  Also, I'll try to stop 'doing it wrong' lol.

Link to post
Share on other sites

Needing some feedback here.

 

It seems we generally agree the new dmg curve is better, and I am pretty sure it will be best to use top 5 tanks instead of top 3 for the normalizer, as I have seen an alarming amount of players have over 150 games in at least 5 tier 1-3 tanks. That will require re-calibration of the curve, which should take a few hours tops.

 

Another thing we agree on is the reduction in weight for defense.

 

 

Something which should be rediscussed is tier-based penalty to frags and spots, and some way to test the formula on a reasonable dataset before releasing it.

Link to post
Share on other sites

Since I'm curious, but also lazy, is there an easy, plug in a few numbers and get a result formula out yet?  Or better yet, something automated?  Looked through around 3 pages of this thread without finding one.

Link to post
Share on other sites

Nope. For the normalizer you need amount of games on top 3-5 most played tier 1-3 tanks. So need to access that through API to calculate WN8.

 

Thats why I need some kind of database to test the formula on...

Link to post
Share on other sites

I took a look at the tier 10 and tier 6 tanks ingame, XVM tracks averages and top 1% (I believe it's top 1%, it could be a different percentile), and using the top 1% numbers, it seems the top 1% scores higher WN8 at tier 6 than at tier 10.

 

On average, top 1% scores ~0,1 kills per game more than tier 10, and their damage is worth ~150 WN8 points more than the top 1% tier 10 players. spotting also tends to be higher for top tier 6 players than tier 10s

 

All in all, plotting a rough average into excel for top tier 6s and top tier 10s gives the tier 6 players ~200 WN8 more than the tier 10 players. Given that this is the same percentile of skill, shouldn't it give the same WN8 score?

I only looked at tier 10 and 6, I don't know is 5, 7, 8 and 9 paint a different picture.

 

Since everybody looks at 60 day stats, if I decided in 60 days I wanted to apply to a clan with a certain WN8 requirement, It would be easier for me to reach that requirement the more tier 6s I played, obviously they wouldn't take someone who plays exclusively 6, but if I play 30% tier 6 and 70% tier 10 it would be easier for me to reach my goal than if I played 20%/80%

 

All this of course assumes that XVM's numbers are accurate, which I can't verify since I don't know where they come from, it also assumes that the top1% at tier 6 is the same level of skill as the top 1% at tier 10, which I think is a reasonable assumption seeing as it is literally the top 1% numbers for both brackets. I hope this makes sense, even reading it back it seems a bit confusing.

Link to post
Share on other sites

I took a look at the tier 10 and tier 6 tanks ingame, XVM tracks averages and top 1% (I believe it's top 1%, it could be a different percentile), and using the top 1% numbers, it seems the top 1% scores higher WN8 at tier 6 than at tier 10

 

I would actually expect this for a very simple reason:

 

The competition at tier 10 is MUCH tougher than at tier 6.  Therefore it is easier to "farm" WN7 at tier 6 than at tier 10.

 

If you were to compare me to an average tier 1 newbie with 100 battles, I am a tank god. But we don't want the numbers to reflect this because the level of competition is so low that even average experienced players would be WAY ahead of the competition.

 

If you were to compare me to an average player at tier 6 (high red) - I am no longer a tank god.  But I am leaning purple/blue, depending on the tank.

 

And finally, if you were to compare me to an average player at tier 10 (low green), I am no longer very dangerous. I am still dangerous, but I do not give my team a nearly absurd advantage - compare to my T49, where I easily singlehandedly carry my team solo 10% of the time.  In fact, i am merely above average in the tier 10 environment, and have been a complete windowlicker compared to many people I have run in to.

 

In short:  The top 1% performance at lower tiers is going to be more impressive than the top 1% performance at higher tiers due to level of competition and experience of enemy tanks. The higher skill gap is going to generate more impressive average performances for elite players.

Link to post
Share on other sites

In short:  The top 1% performance at lower tiers is going to be more impressive than the top 1% performance at higher tiers due to level of competition and experience of enemy tanks. The higher skill gap is going to generate more impressive average performances for elite players.

 

I agree with everything you said, but shouldn't the rating reflect this? If it is easier to perform impressively at tier 6, then it should give you the same score as what level of performance is equally easy at tier 10. Obviously the world isn't perfect, but ~200 points difference between top 1% seems pretty significant.

 

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...