Jump to content
Never

Expected Values updated to v27

Recommended Posts

6 hours ago, TheMarine0341 said:

Im just waiting for WN9, I'll be 395+ (or about 2650 WN8 equivilent) overall

Sorry to bursty your bubble, but the color scale on the page itself is out of date.

You should be looking at this:

base 300 base 320 base 450 base 650
<100 <100 <150 <200
100 100 150 200
140 150 210 300
180 200 270 400
220 250 340 500
270 300 410 600
320 350 480 700
370 400 550 800
420 450 620 900
470 500 700 1000

Share this post


Link to post
Share on other sites
32 minutes ago, Orrie said:

Sorry to bursty your bubble, but the color scale on the page itself is out of date.

You should be looking at this:

I was basing it on the website for calculating WN9, as the scale isn't officially official yet

Share this post


Link to post
Share on other sites
3 hours ago, Orrie said:

Blame @RichardNixonfor being lazy and not updating it.

I hadn't decided which scale won until now :P

The 650 base is only an approximation anyway, so even if the scale is decided then I can't use it yet. Final multipler result depends on how I interpret the other polls and which WN8 version I use for comparison (as v27 apparently pushed everyone up, and condition was purple WN8 = purple account WN9). Likely range at the moment is 660-680.

 

6 hours ago, canadiantrex said:

What is the post 8.0 top guns bullshit? I started pkaying in 8.5 and it's assuming some of my top guns were before the 8.0 patch?

No. It's a debug message that's displayed regardless of whether there is an adjustment. 131 is the same as your total top guns, hence no adjustment.

 

Share this post


Link to post
Share on other sites

http://www.wnefficiency.net/wnexpected/

 

When sorting by delta damage, we can see the tanks that were most affected by the change.  

 

First, we can see that the T-22 had its' expected massively raised by 359, probably in line with its' 9.14 performance.  It'll be interesting to see weather this drops back down slightly with the recent nerf.  

The 260 finally had its' expected raised, and now stands nearly 300 above the IS-7

The Skoda T 50 and 50/51 both had their expected go up by around 200, which is well deserved; mayhaps WG will finally admit they (especially the Skoda) are overperforming now.  

WHAT?!?! WHY DID THE EXPECTED ON THE T-55 GO UP BY NEARLY 200?  It is just a T-54 slightly better gun handling that trades a huge portion of its' mobility for the buff

907 had its' expected go up by almost 150, which I have no strong opinion about; I seem to do better in that than my other RUmeds

AMX 30 prot had a big jump and has very high expected for a T9 tank now, which I am suprised about... I consider that to be one of the worst tanks I've ever played.  It's gun handling is entirely too bad to have an ammo rack that weak. About the only thing good about it is the fantastic shell velocity.  Then again, I might just be bad at the tank.  

Also of interest, the IS-5 got a noticeable bump, as did the 95e6 and M60.  The e5 also had a minor bump.

 

 

The other end of the scale is dominated by arty, many of which had their expected damage cut by 100 or more.  Of interest among them are the CGC, T92, and 53/55, all of which got a big bonus.  

Additionally, the 183 and 4005 both had their expected reduced slightly.  

Of interest, the Waffle got a decent drop of nearly 100 after its' removal.  

Share this post


Link to post
Share on other sites
17 minutes ago, TwixOps said:

WHAT?!?! WHY DID THE EXPECTED ON THE T-55 GO UP BY NEARLY 200?

Probably 100-150 points for having no stock grind, and a bit more for recency bias. The IS-3 vs IS-5/IS-6 relationship is nearly as far off as T-54 vs T-55A.

The arty cut was semi-manual, by the way. The automated method locks some arties to pre-8.6 results. This time, post-8.6 data was manually blended in.
 

 

Share this post


Link to post
Share on other sites
11 hours ago, TwixOps said:

http://www.wnefficiency.net/wnexpected/

 

When sorting by delta damage, we can see the tanks that were most affected by the change.  

 

First, we can see that the T-22 had its' expected massively raised by 359, probably in line with its' 9.14 performance.  It'll be interesting to see weather this drops back down slightly with the recent nerf.  

The 260 finally had its' expected raised, and now stands nearly 300 above the IS-7

The Skoda T 50 and 50/51 both had their expected go up by around 200, which is well deserved; mayhaps WG will finally admit they (especially the Skoda) are overperforming now.  

WHAT?!?! WHY DID THE EXPECTED ON THE T-55 GO UP BY NEARLY 200?  It is just a T-54 slightly better gun handling that trades a huge portion of its' mobility for the buff

907 had its' expected go up by almost 150, which I have no strong opinion about; I seem to do better in that than my other RUmeds

AMX 30 prot had a big jump and has very high expected for a T9 tank now, which I am suprised about... I consider that to be one of the worst tanks I've ever played.  It's gun handling is entirely too bad to have an ammo rack that weak. About the only thing good about it is the fantastic shell velocity.  Then again, I might just be bad at the tank.  

Also of interest, the IS-5 got a noticeable bump, as did the 95e6 and M60.  The e5 also had a minor bump.

 

 

The other end of the scale is dominated by arty, many of which had their expected damage cut by 100 or more.  Of interest among them are the CGC, T92, and 53/55, all of which got a big bonus.  

Additionally, the 183 and 4005 both had their expected reduced slightly.  

Of interest, the Waffle got a decent drop of nearly 100 after its' removal.  

30prot is easily the best tier 9 medium in the game that's not Russian. only topped by t54 with heat spam.

it has gun depression over the e50 as well, which some will argue is the best tank. oh well. cant make everybody happy. I know some people who play cancer got a jump of over 200 in overalls. seriously, where the fuck is the justice and logic here

what a joke but who even cares anymore

 

Share this post


Link to post
Share on other sites
On 6/21/2016 at 6:25 PM, breeeze said:

0RNOgQM.png

+73:serb: I dare anyone to beat that

 

80 pts, thanks to my early WoT days playing cancer and lights. lol.. up 400 pts in CGC, and 240 in T92.  I haven't even owned that since 2013.  And now that I'm purple I'm feeling the increased XVM focus... RIP arty karma.

rwUZIgA.png

4n4zi5X.png

Share this post


Link to post
Share on other sites

Hello, just joined forum to figure out why WN8 had jumped 30+ pts (overall) last week. I guess this is why. Not real happy with WN9 results (from above link), took me down a color tier, but I guess it isn't official yet...

Share this post


Link to post
Share on other sites
2 hours ago, _Legion_of_Boom_ said:

Not real happy with WN9 results (from above link), took me down a color tier, but I guess it isn't official yet...

That's probably not going to change. The issue is that the WN8 expected value method was somewhat tier-biased. Players generally play lower-tier tanks earlier in their careers, and that dragged down the expected values. For tier 5-6 non-prems the error's around 10-20%. Your top four tanks (T-150, Stug III, T67, KV-13) are all strong WN8 padders.

 

Share this post


Link to post
Share on other sites
10 hours ago, RichardNixon said:

That's probably not going to change. The issue is that the WN8 expected value method was somewhat tier-biased. Players generally play lower-tier tanks earlier in their careers, and that dragged down the expected values. For tier 5-6 non-prems the error's around 10-20%. Your top four tanks (T-150, Stug III, T67, KV-13) are all strong WN8 padders.

 

That's just silly! 20% is A LOT! When shall we see WN9 implemented instead of WN8? Any time soon?

Share this post


Link to post
Share on other sites
21 hours ago, RichardNixon said:

That's probably not going to change. The issue is that the WN8 expected value method was somewhat tier-biased. Players generally play lower-tier tanks earlier in their careers, and that dragged down the expected values. For tier 5-6 non-prems the error's around 10-20%. Your top four tanks (T-150, Stug III, T67, KV-13) are all strong WN8 padders.

 

Maybe so... I don't see the T-150 as a WN8 "padder" (I won't argue the others 1 way or the other), I have loved it since before they nerfed the KV-1S and everyone said the T-150 sucked. I am actually pretty proud of my 60% winrate in it. That is w/o platooning and mostly w/o gold. 

What do you meant by "For tier 5-6 non-prems the error's around 10-20%."


Not really sure how, when you are being compared to the damage others are doing in the same tanks, one tank can be better/different than another. I would think "expected value" should/would be a per tank evaluation, relative to everyone else. There really isn't anything more important than win % and damage a close 2nd (IMHO).

I am really not that worried about it. That was also "recent" as opposed to overall, and apparently overall will commonly be higher. 

What is "recent"? Last 500 games or something?

 

 

Share this post


Link to post
Share on other sites
5 hours ago, _Legion_of_Boom_ said:

Maybe so... I don't see the T-150 as a WN8 "padder" (I won't argue the others 1 way or the other), I have loved it since before they nerfed the KV-1S and everyone said the T-150 sucked. I am actually pretty proud of my 60% winrate in it. That is w/o platooning and mostly w/o gold.

The T-150's actually the strongest tier 6 tank, but reputation doesn't always follow. It's especially underrated by WN8 because it has a long & horrible stock grind, and it wasn't a popular elite tank, at least historically.

The first principle of WN9 is that should be similarly difficult to attain the same value in any tank, assuming that you play it elite with a decent crew. Otherwise the metric creates artificial incentives and padding opportunities.

Tier 5-6 sealclubbing wasn't particularly popular for WN8 padding (at least after the Hellcat nerf), but once you fix the more obvious padding opportunities (eg. E-50, lights) it's the next problem down the line.

Share this post


Link to post
Share on other sites
On 6/22/2016 at 9:20 AM, RichardNixon said:

That's the WN8 method, and it gets all those tanks the wrong way around, mostly due to recency bias. Interval method is WN9.

The 907's advantage (if any) is down to survivability, but the detailed API stats are incapable of measuring it properly. Here's a hint from raw average survival rates though:

  dmg_dealt dmg_taken survival
Object 907 2447.9 1663.6 0.374
Object 140 2014.9 1652.2 0.281
M60 2342.4 1687.4 0.349
M48A1 1983.9 1732.9 0.285
 

I still think that M60 should not get a higher expected value than M48.

Share this post


Link to post
Share on other sites
On 6/22/2016 at 7:01 AM, bluestealth said:

Hooray, I am no longer position 69  :doge:

Are you position 420 now?

-----------------------

edited to contribute somewhat to thread:

1201 wn9 on my m46 KR. Guess that's the new violent stats pusher. Also not super uni overall FeelsBadMan.

@Orrie can you lower the expected of only my 907 (not everyone else's). My ammorack is cursed. It should be decreased by 150 imo. Thanks.

Also, good work bois.

Share this post


Link to post
Share on other sites
16 hours ago, weenis said:

Are you position 420 now?

-----------------------

edited to contribute somewhat to thread:

1201 wn9 on my m46 KR. Guess that's the new violent stats pusher. Also not super uni overall FeelsBadMan.

@Orrie can you lower the expected of only my 907 (not everyone else's). My ammorack is cursed. It should be decreased by 150 imo. Thanks.

Also, good work bois.

ur bad :ohnyes:

 

also

 

907 has high survival rate because better players play it and die less, just look at any good super uni's survival rate

Share this post


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...