Jump to content
RichardNixon

WN9 candidate prototype

Recommended Posts

Regarding WN9, I think RN gave all lights the base values for the tier above, before applying corrections. That was harsh as it ignored the fact that, contrary to his 'carry your weight' argument, lights have the MM weight of the tier below. Now, we just learned that some lights will all move up a tier, others wont, and new ones will be added. Buffs and nerfs will abound. This may influence when WN9 is rolled out into service. It might even need a tweak to the method first.

Share this post


Link to post
Share on other sites
2 minutes ago, Gryphon_ said:

Regarding WN9, I think RN gave all lights the base values for the tier above, before applying corrections. That was harsh as it ignored the fact that, contrary to his 'carry your weight' argument, lights have the MM weight of the tier below. Now, we just learned that some lights will all move up a tier, others wont, and new ones will be added. Buffs and nerfs will abound. This may influence when WN9 is rolled out into service. It might even need a tweak to the method first.

Somehow I think those are relatively far off from being released. At that point, the only real change with the formula should just be the MM weights (we can probably ballpark them getting the same MM as equivalent-tier mediums) and the data should catch any other significant changes... assuming someone will be updating it.

Share this post


Link to post
Share on other sites

A few weeks passed and sadly I don't see WN9 implemented on WotLabs, which is sad because this site should take the lead. Not blaming anyone, just stating the obvious.

Share this post


Link to post
Share on other sites
6 hours ago, dirtyred said:

A few weeks passed and sadly I don't see WN9 implemented on WotLabs, which is sad because this site should take the lead. Not blaming anyone, just stating the obvious.

See below

On 1/15/2017 at 10:47 AM, Gryphon_ said:

Regarding WN9, I think RN gave all lights the base values for the tier above, before applying corrections. That was harsh as it ignored the fact that, contrary to his 'carry your weight' argument, lights have the MM weight of the tier below. Now, we just learned that some lights will all move up a tier, others wont, and new ones will be added. Buffs and nerfs will abound. This may influence when WN9 is rolled out into service. It might even need a tweak to the method first.

 

Share this post


Link to post
Share on other sites
On 1/27/2017 at 6:17 AM, dirtyred said:

A few weeks passed and sadly I don't see WN9 implemented on WotLabs, which is sad because this site should take the lead. Not blaming anyone, just stating the obvious.

FWIW, it's time-consuming to write the code so that the tables, charts, graphs all work correctly. Also, @Nevermay need to write code that handles updating the historical per-player, per-tank data so that the historical trend graphs will render correctly. In addition, he has to update the code for the forum signature generator.

It's a fair amount of work. Give him time to make sure the work is done right.

On a broader note, the reason I wanted to post on this thread is to say a big THANK YOU to the folks who have contributed to the WN9 model.

I've been reading up on the excellent writeups on WN9 on jaj22's web site, and WN9 seems to address some of the limitations or issues with prior implementations of WN*:

http://jaj22.org.uk/wn9description.html

http://jaj22.org.uk/wn9faq.html

The folks here have done an impressive job of working through issues with the data

I love the concept of "Account WN9":

Quote

 

Account WN9 is a replacement for overall WN8 that throws away each player's worst tanks. The goal is to reduce reroll incentive, work around problems with historical nerfs & buffs, and make it work better as a skill metric for applications where you can't use recent WN9.

The discard level is currently set at 65% of battles (selected by tank), which was chosen by polling

 

 

Share this post


Link to post
Share on other sites

taugrim is very much corect.

God (actually @Kitten) knows I've been struggling to implement WN9. It does a lot of things in a way that WoTLabs was never designed to do nor it had the capacity to do due to the information available from the API at the time.

So please hang in there. I want WN9 as much as anyone here, I just ask for patience.

Share this post


Link to post
Share on other sites
On 2/3/2017 at 5:39 AM, taugrim said:

FWIW, it's time-consuming to write the code so that the tables, charts, graphs all work correctly. Also, @Nevermay need to write code that handles updating the historical per-player, per-tank data so that the historical trend graphs will render correctly. In addition, he has to update the code for the forum signature generator.

It's a fair amount of work. Give him time to make sure the work is done right.

On a broader note, the reason I wanted to post on this thread is to say a big THANK YOU to the folks who have contributed to the WN9 model.

I've been reading up on the excellent writeups on WN9 on jaj22's web site, and WN9 seems to address some of the limitations or issues with prior implementations of WN*:

http://jaj22.org.uk/wn9description.html

http://jaj22.org.uk/wn9faq.html

The folks here have done an impressive job of working through issues with the data

I love the concept of "Account WN9":

 

You might like WN9 and wotlabs may need to take a leadership role in trying to roll it out but it won't get used anyway. The guys at XVM aren't interested in it at all and that is because of the glaringly obvious flaw in WN9, almost NO POINT OF DIFFERENCE from WN8. When WN9 contained the platoon padding quotient the buzz around WN9 was almost Amazon cicadas level. As soon as it was removed WN9 died. It is a shame because it could have really made a difference to how WG actually looked at it's own API. We may have even got them to separate out platoon stats from solo stats. but because the powers at be relented to the padding element here at wotlabs no more platoon quotient. Why would xvm go through all this extra coding just to get a slightly different rainbow scale to the one we already have.

Share this post


Link to post
Share on other sites
8 minutes ago, Acumen said:

You might like WN9 and wotlabs may need to take a leadership role in trying to roll it out but it won't get used anyway. The guys at XVM aren't interested in it at all and that is because of the glaringly obvious flaw in WN9, almost NO POINT OF DIFFERENCE from WN8. When WN9 contained the platoon padding quotient the buzz around WN9 was almost cicadas in Amazon level. As soon as it was removed WN9 died. It is a shame because it could have really made a difference to how WG actually looked at it's own API. We may have even got them to separate out platoon stats from solo stats. but because the powers at be relented to the padding element here at wotlabs. Why would xvm go through all this extra coding just to get a slightly different rainbow scale to the one we already have.

There's virtually no difference between WN7, 8, 9 and PR at the account level. You can only shake out so much from the API.

Your thought about platoon vs solo is just wishful thinking though. It's been years for spotting damage. The API makes them no money , nor risks any. There is no reason for them to touch it. Plus the platoon estimates weren't integral to the method improvements in 9 , nor did they end up being terribly discerning.

WN9 works very nicely for recents, due to the extremely diligent sampling and it'd be nice to see PR used for overalls. After all PR is supported by the only people getting a paycheck.

Share this post


Link to post
Share on other sites
Just now, Ezz said:

We being the API.

I thought the API included it now, but there were issues with not having data from all time, so the decision was made to continue to exclude assisted damage?  I did read that somewhere, but.... not true?

Share this post


Link to post
Share on other sites
Just now, moogleslam said:

I thought the API included it now, but there were issues with not having data from all time, so the decision was made to continue to exclude assisted damage?  I did read that somewhere, but.... not true?

Since 8.8 i think. But yeah, it would be nice if someone tried to make a 'recent' that included it. Not sure if anyone looked much into it tho.

Share this post


Link to post
Share on other sites
2 minutes ago, Ezz said:

We being the API.

assistance damage can now be accessed by API it has been for a while but I think it is only recent or something so no historical values or something like that. Because you know historical values of wn8 are so accurate. Like my tier 9 batchat stats which I have a wn8 of 2445 or something but have never owned or played.

Share this post


Link to post
Share on other sites
1 minute ago, Ezz said:

Since 8.8 i think. But yeah, it would be nice if someone tried to make a 'recent' that included it. Not sure if anyone looked much into it tho.

Would be real nice..... I'd die happy :)

Share this post


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

almost NO POINT OF DIFFERENCE from WN8.

Huh? It cuts LT padding and stops punishing arty players. It also works around buffed/nerfed tanks. Thats definetely a good amount of difference and one that is quite important in the light of the recent drastic tank changes.

Share this post


Link to post
Share on other sites
6 minutes ago, Jaegaer said:

It cuts LT padding

Mostly arbitrarily from what i gather. Just as above, without spotting assistance the whole light class was always going to be out of whack.

Share this post


Link to post
Share on other sites

No, spotting damage is not in a usable form in the API. The only place it tracks spotting damage is your overall account statistics since 8.8... Which means it's useless. No per tank spotting damage, never has been, and evidently WG doesn't care enough to make it happen. 

Share this post


Link to post
Share on other sites
1 hour ago, Kitten said:

No, spotting damage is not in a usable form in the API. The only place it tracks spotting damage is your overall account statistics since 8.8... Which means it's useless. No per tank spotting damage, never has been, and evidently WG doesn't care enough to make it happen. 

Okay, good to know.  I'll go back to blaming WG then :)

Suppose they did start including assisted damage.  Would it then be possible to include it in a "from this point forth" WNx metric, or would not having it from the beginning of time still pose a problem?

Thanks

Share this post


Link to post
Share on other sites
17 hours ago, Jaegaer said:

Huh? It cuts LT padding and stops punishing arty players. It also works around buffed/nerfed tanks. Thats definetely a good amount of difference and one that is quite important in the light of the recent drastic tank changes.

If it can't use spotting damage then all you are doing is gimping the lights and quite frankly fuck arty.

Share this post


Link to post
Share on other sites
1 hour ago, Acumen said:

If it can't use spotting damage then all you are doing is gimping the lights and quite frankly fuck arty.

WN9 doesn't have any arbitrary LT anti-padding measures, so nothing gets intentionally gimped.

It just assumes not all tanks' dmg scale similarly with skill, unlike WN8.

Share this post


Link to post
Share on other sites
17 hours ago, moogleslam said:

Okay, good to know.  I'll go back to blaming WG then :)

Suppose they did start including assisted damage.  Would it then be possible to include it in a "from this point forth" WNx metric, or would not having it from the beginning of time still pose a problem?

Thanks

If WG time stamped the API the sampling gymnastics of WN9 would be simplified nicely. Recents would be able to be valid. Account overall would be a very complicated formula, in that it'd need different parts for different eras.  Spotting would be then potentially useable, and evaluated (assuming per tanks). But there is literally "nothing" we can do about shitty record keeping - and by nothing I mean RN set up a shadow server to pull and store API intervals to faux up timestamped data.

That's why I just say "use PR" in every other post. Their messy pile of data, let them fight it.

Share this post


Link to post
Share on other sites
19 minutes ago, CraBeatOff said:

That's why I just say "use PR" in every other post. Their messy pile of data, let them fight it.

For years the lack of overarching metric engendered a view that WG could not, or would not, come up with a decent ability analogue. It will take time to convince the community otherwise.

Share this post


Link to post
Share on other sites
36 minutes ago, Ezz said:

For years the lack of overarching metric engendered a view that WG could not, or would not, come up with a decent ability analogue. It will take time to convince the community otherwise.

It's been out 1 month shorter than WN8...of course the did fuck up the first edition badly...

Share this post


Link to post
Share on other sites

From recollection the prominence of hit rate and battle count were the hairiest of chin moles. But since then they've even managed MoE which ticks many a box in many a circle.

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...