Jump to content


[WN7] What is it and how does it work?

statistics stats noobmeter performance efficiency metric compare quality rating Garbad

  • Please log in to reply
2483 replies to this topic

VaporGator #2461 Posted Oct 11 2013 - 14:24

    First lieutenant

  • Players
  • 9773 battles
  • 809
  • [5PANZ] 5PANZ
  • Member since:
    09-19-2011

View PostPraetor77, on Oct 11 2013 - 01:46, said:

I know...  :confused:
I´ve also been having lag spikes, packet loss, 300ms ping, you name it. Thats why I have been playing less tanks and doing more WN8! :)

All I am hearing from your post is "denier!!"      :teethhappy:

Praetor77 #2462 Posted Oct 11 2013 - 21:10

    Captain

  • Players
  • 29723 battles
  • 1,675
  • Member since:
    05-01-2011

View PostVaporGator, on Oct 11 2013 - 14:24, said:

All I am hearing from your post is "denier!!"   :teethhappy:
:teethhappy:

Meoow #2463 Posted Oct 19 2013 - 18:13

    Sergeant

  • Players
  • 17938 battles
  • 233
  • [MMIMM] MMIMM
  • Member since:
    04-25-2011
if you pay attention, all the unicum and super unicum players get these from their 2nd or 3rd account just for showing off good stats. They don't solo pub, only platoon and company. They don't often play anything higher than tier 5-6 because that would kill their stats instantly.
Logically speaking, you can't be super good at anything that you had never done before...so yeah...I'm close to unicum by hard work and mostly solo pub. I had an extensive knowledge about tanks before this game came out for years which did help. [Content moderated] But then again, it is their way of gaming. Most of us don't time and money just to get crazy stats in a game.
Content moderated.
~Moderation Team

Edited by Irith, Oct 19 2013 - 21:49.


sr360 #2464 Posted Oct 19 2013 - 20:02

    Major

  • Players
  • 56516 battles
  • 7,085
  • [-G-] -G-
  • Member since:
    06-26-2011

View PostMeoow, on Oct 19 2013 - 18:13, said:

if you pay attention, all the unicum and super unicum players get these from their 2nd or 3rd account just for showing off good stats. They don't solo pub, only platoon and company. They don't often play anything higher than tier 5-6 because that would kill their stats instantly.
Logically speaking, you can't be super good at anything that you had never done before...so yeah...I'm close to unicum by hard work and mostly solo pub. I had an extensive knowledge about tanks before this game came out for years which did help. [Content moderated] But then again, it is their way of gaming. Most of us don't time and money just to get crazy stats in a game.
Lawl.

VaporGator #2465 Posted Oct 19 2013 - 20:03

    First lieutenant

  • Players
  • 9773 battles
  • 809
  • [5PANZ] 5PANZ
  • Member since:
    09-19-2011

View PostMeoow, on Oct 19 2013 - 18:13, said:

if you pay attention, all the unicum and super unicum players get these from their 2nd or 3rd account just for showing off good stats. They don't solo pub, only platoon and company. They don't often play anything higher than tier 5-6 because that would kill their stats instantly.
Logically speaking, you can't be super good at anything that you had never done before...so yeah...I'm close to unicum by hard work and mostly solo pub. I had an extensive knowledge about tanks before this game came out for years which did help. [Content moderated] But then again, it is their way of gaming. Most of us don't time and money just to get crazy stats in a game.
Actually that is a very valid point...  How can someone claim to be the best if they don't play the hardest tanks in Tier 10?
Perhaps average tier needs to be part of WN factoring...  separate the average tier 5 "unicums" from average tier 8 "unicums"....  If Win Rate can be a fudge factor, Average Tier should mean a lot as well because it shows the higher it is the more rounded a player you are.

Euro_Trash #2466 Posted Oct 19 2013 - 22:35

    First lieutenant

  • Players
  • 1216 battles
  • 680
  • [EUCUM] EUCUM
  • Member since:
    07-03-2013

View PostMeoow, on Oct 19 2013 - 18:13, said:

if you pay attention, all the unicum and super unicum players get these from their 2nd or 3rd account just for showing off good stats. They don't solo pub, only platoon and company. They don't often play anything higher than tier 5-6 because that would kill their stats instantly.
Logically speaking, you can't be super good at anything that you had never done before...so yeah...I'm close to unicum by hard work and mostly solo pub. I had an extensive knowledge about tanks before this game came out for years which did help. [Content moderated] But then again, it is their way of gaming. Most of us don't time and money just to get crazy stats in a game.
Content moderated.
~Moderation Team

In short, anyone with higher numbers is a phony and you are the real deal?

sr360 #2467 Posted Oct 19 2013 - 23:00

    Major

  • Players
  • 56516 battles
  • 7,085
  • [-G-] -G-
  • Member since:
    06-26-2011
He is as close to unicum as I am to AntonioHandsome in stats

arrgh #2468 Posted Oct 21 2013 - 23:32

    Private

  • Beta Testers
  • 0 battles
  • 7
  • Member since:
    01-03-2011
Nice discussion so far. (Not that I read all 130 pages, sorry for that).
I would like to throw in two arguments, don´t know if they are new.
1. I watch more and more killstealers, players that really wait and calculate with the death of their team just to throw in a shell for the last hit points, even if they reloaded long before. I would like to see a correlation with damage dealt on the killed tank. If the last shot hits by lets say 10 % of possible damage, then the real deal goes to the tank that did the most damage... even if the normal score for a kill is just 1.77 as I calculated.
2. If you don´t award cap points, you rip away a little more of the teamplay. What has the higher score, to win the game or to hunt the enemy to the death...
thx
Edit:
I missed a bracket at my calculatien, one kill is worth bout 464 points efficiency, one point of damage is worth 0.24 in x, 0.307 in tier 9 and 0.38 in tier 8... must killsteal more

Edited by arrgh, Oct 25 2013 - 16:58.


Sticky_Drapes #2469 Posted Oct 22 2013 - 00:00

    Captain

  • Players
  • 35728 battles
  • 1,348
  • [RDDT] RDDT
  • Member since:
    04-30-2011
So, which metric do I have to pad to get my WN7 up? I'm not reading 130 pages.

Pahech #2470 Posted Oct 22 2013 - 00:07

    Major

  • Beta Testers
  • 21363 battles
  • 3,833
  • Member since:
    09-28-2010

View Postclownshoes2, on Oct 22 2013 - 00:00, said:

So, which metric do I have to pad to get my WN7 up? I'm not reading 130 pages.

Play less team companies.

HypersonicGlider #2471 Posted Oct 22 2013 - 03:54

    Sergeant

  • Players
  • 39499 battles
  • 180
  • Member since:
    05-01-2012

View Postclownshoes2, on Oct 22 2013 - 00:00, said:

So, which metric do I have to pad to get my WN7 up? I'm not reading 130 pages.
WN7 will be obsolete soon.
To answer your question, kills is the most weighted in WN7 (not anymore in WN8 tho)

bjshnog #2472 Posted Oct 22 2013 - 06:41

    Major

  • Players
  • 18386 battles
  • 2,567
  • [LABS] LABS
  • Member since:
    07-13-2011

View Postarrgh, on Oct 21 2013 - 23:32, said:

1. I watch more and more killstealers, players that really wait and calculate with the death of their team just to throw in a shell for the last hit points, even if they reloaded long before. I would like to see a correlation with damage dealt on the killed tank. If the last shot hits by lets say 10 % of possible damage, then the real deal goes to the tank that did the most damage... even if the normal score for a kill is just 1.77 as I calculated.

2. If you don´t award cap points, you rip away a little more of the teamplay. What has the higher score, to win the game or to hunt the enemy to the death...

1. Kills are kills and damage is damage. In WN8, damage plays a bigger role.

2. Most (definitely >50%) cap points are gained when the battle is already won and the player just wants some extra experience by sitting in the enemy cap while the last enemy dies on the other end of the map. It's just valuable enough for WN.

Noklith #2473 Posted Dec 29 2013 - 11:29

    Corporal

  • Players
  • 16052 battles
  • 29
  • Member since:
    06-12-2013

LOL, THIS is why online gaming has become a joke.  You all want to stick math, so called logic, and whatever else into a simple stress release online game.  Simple gd common sense debunks your whole Goodwill Hunting movie into BS.  Why?   Because WoT MM doesn't take into account of ANY of your WN7 super computer math equation.  You mean to tell me if the 65%er gets stuck into a match with 40% players against all 50%ers he is gonna win all the time?  No way.  MM matches tanks and tiers only and with T7 against T9 or above its a joke.  At the current state of MM, most T7's can barely pen T9 frontal.  And to think a T7 can flank T9's on most maps is a joke.  There are a ton more T9/10's now days and they should be in their own matches.  All other matches should be plus 1 / minus 1.  Plus the amount of people that play now that are causal players is way more then your "evolutionary algorithms' allows, and casuals really don't give a crap about stats, thus ruining those that do and rely on MM.  But I'm sure you know that considering this thread is over a year old.  Not to mention the simple fact that most maps are nothing but camp-fests and ALL high rate WN7 players pad their stats playing T6 matches.  You said yourself that "Players with a considerable number of battles who have an average tier lower than 4 are heavily penalized for sealclubbing."  So T6 matches, which all major clans filter have no effect and pad stats.  Awesome try to come up with something, but the reality of things is, at the end of the day its a GD online game full of half wit players that YOU can and should NEVER try to over-fAqn-analyze.  Just go have fun and go "Shooting tanks for the heck of it" and if you get bored of it "after a while" go outside and do something.  There are a ton of more variables that could go in effect to measure a "good" player, none of which you have mentioned. Only in-game data that you stated has limitations that are huge factors does Wn7 factor in.  Seems pretty simple to me, but maybe I'm just not a smart as you math wiz are.



bjshnog #2474 Posted Dec 29 2013 - 12:05

    Major

  • Players
  • 18386 battles
  • 2,567
  • [LABS] LABS
  • Member since:
    07-13-2011

View PostNoklith, on Dec 29 2013 - 21:29, said:

LOL, THIS is why online gaming has become a joke.  You all want to stick math, so called logic, and whatever else into a simple stress release online game.  Simple gd common sense debunks your whole Goodwill Hunting movie into BS.  Why?   Because WoT MM doesn't take into account of ANY of your WN7 super computer math equation.  You mean to tell me if the 65%er gets stuck into a match with 40% players against all 50%ers he is gonna win all the time?  No way.  MM matches tanks and tiers only and with T7 against T9 or above its a joke.  At the current state of MM, most T7's can barely pen T9 frontal.  And to think a T7 can flank T9's on most maps is a joke.  There are a ton more T9/10's now days and they should be in their own matches.  All other matches should be plus 1 / minus 1.  Plus the amount of people that play now that are causal players is way more then your "evolutionary algorithms' allows, and casuals really don't give a crap about stats, thus ruining those that do and rely on MM.  But I'm sure you know that considering this thread is over a year old.  Not to mention the simple fact that most maps are nothing but camp-fests and ALL high rate WN7 players pad their stats playing T6 matches.  You said yourself that "Players with a considerable number of battles who have an average tier lower than 4 are heavily penalized for sealclubbing."  So T6 matches, which all major clans filter have no effect and pad stats.  Awesome try to come up with something, but the reality of things is, at the end of the day its a GD online game full of half wit players that YOU can and should NEVER try to over-fAqn-analyze.  Just go have fun and go "Shooting tanks for the heck of it" and if you get bored of it "after a while" go outside and do something.  There are a ton of more variables that could go in effect to measure a "good" player, none of which you have mentioned. Only in-game data that you stated has limitations that are huge factors does Wn7 factor in.  Seems pretty simple to me, but maybe I'm just not a smart as you math wiz are.

 

Holy crap... Not only do you have zero clue what you're talking about, but you're also way behind the times.



sr360 #2475 Posted Dec 29 2013 - 13:39

    Major

  • Players
  • 56516 battles
  • 7,085
  • [-G-] -G-
  • Member since:
    06-26-2011
He is right though, in that he's not very smart. The rest, of course, is laughably wrong.

TheKilltech #2476 Posted Jan 07 2014 - 17:51

    Corporal

  • Players
  • 49 battles
  • 84
  • Member since:
    02-26-2013

i very much like the new WN8 approach.

 

there is only one minor thing i can suggest to make it a bit more precise (though it this might be something to consider for WN9). the problem with the relative rSTAT is that it still is quite vehicle dependent. doing 1.5 of the expected damage on a hellcat is much less impressive then the same value on an AMX 40 (the damage distributions for these tanks differ strongly in their broadness). ideally it would be the best to take the quantiles for each stat: for the hellcat 1.5 rDamage might be just better then 70% of all players on that tank (guessing / just for example) but on the AMX 40 it might already be better then 90%. of course using quantiles isn't an option because for each stat and tank one would need to know the inverse of the cumulative distribution for that stat instead of just one value (expStat). still one can improve the calculation by using one additional value, the variance of the distribution for each stat.

 

so instead of using:

 

rStat = avgStat / expStat

 

i'd suggest using:

 

rStat = (avgStat - expStat) / sqrt(varStat)

 

that would however require to calculate the varStat values for each tank and stat anew as well as the weighting factors for the WN formula.

also ideally the weighting factors in the WN formula should be different for each tank because different tanks contribute differently to the win chance - though that is probably overkill considering the stats are done per tank already, 


Edited by TheKilltech, Jan 07 2014 - 17:52.


tbajas #2477 Posted Feb 10 2014 - 05:58

    Corporal

  • Players
  • 4168 battles
  • 12
  • [ITD] ITD
  • Member since:
    02-28-2012

Hello Praetor77,

For the first 2 years of my WOT experience I was playing on an [extremely] low end system, only achieving framerates of about 5-15 FPS on minimal settings, and I could tell just how negatively effected my performance by playing on my friends gaming PC's and feeling like I could actually put a good shot into someone. Now that I finally have a good gaming PC I have felt a fantastic increase of my performance, however I'm still marked by the 'Bad' WN7 rating when anyone looks at my stats, which is less than satisfying. While I'm certainly not a great player, whenever I platoon with a friend or clan member they always say that I'm pretty good, and those that have looked at my record prior to are presently surprised at my performance. So with this in mind, I was wondering if there is any way to implement a new factor that takes into account your average FPS. I mean, we all know that FPS plays a role in what you can do in certain situations, and my low fps had been the bane of  my WOT experience from the beginning. Making it difficult to properly aim at times (and forget driving, shooting, and repairing modules in any sort of simultaneous action). And although I'm aware this new factor wouldn't effect my previous performance's it would certainly make low-end users happy, and assist people coming into the realm of PC gaming (not to mention WOT) for the first time on a low-end system. So there's an idea. I'm sure there would be a lot of balancing necessary to make it fair for everyone, but I'm sure that I speak for most people who came into (or still play) this game on a low end PC that it may be a nice buffer. Thanks for the read and have a nice day!

-tbajas 



bjshnog #2478 Posted Feb 10 2014 - 06:28

    Major

  • Players
  • 18386 battles
  • 2,567
  • [LABS] LABS
  • Member since:
    07-13-2011

View Posttbajas, on Feb 10 2014 - 15:58, said:

Hello Praetor77,

For the first 2 years of my WOT experience I was playing on an [extremely] low end system, only achieving framerates of about 5-15 FPS on minimal settings, and I could tell just how negatively effected my performance by playing on my friends gaming PC's and feeling like I could actually put a good shot into someone. Now that I finally have a good gaming PC I have felt a fantastic increase of my performance, however I'm still marked by the 'Bad' WN7 rating when anyone looks at my stats, which is less than satisfying. While I'm certainly not a great player, whenever I platoon with a friend or clan member they always say that I'm pretty good, and those that have looked at my record prior to are presently surprised at my performance. So with this in mind, I was wondering if there is any way to implement a new factor that takes into account your average FPS. I mean, we all know that FPS plays a role in what you can do in certain situations, and my low fps had been the bane of  my WOT experience from the beginning. Making it difficult to properly aim at times (and forget driving, shooting, and repairing modules in any sort of simultaneous action). And although I'm aware this new factor wouldn't effect my previous performance's it would certainly make low-end users happy, and assist people coming into the realm of PC gaming (not to mention WOT) for the first time on a low-end system. So there's an idea. I'm sure there would be a lot of balancing necessary to make it fair for everyone, but I'm sure that I speak for most people who came into (or still play) this game on a low end PC that it may be a nice buffer. Thanks for the read and have a nice day!

-tbajas 

 

Not even close to possible. Wargaming does not track your framerate, neither does your WoT client, and there is virtually zero chance that anything like that would be implemented. Plus, different people deal with drops in framerate in different ways. I, for example, have played at 10fps and found no decrease in performance at all. I see you have only 2317 battles. If you play a decent amount of battles, your 'bad' stats should be washed away quickly if you are as decent as you claim you are (based on the stats I can see on Noobmeter, you are definitely improving).

Also, this thread is about WN7, which was replaced with WN8 a few months ago. ;)



tbajas #2479 Posted Feb 10 2014 - 06:59

    Corporal

  • Players
  • 4168 battles
  • 12
  • [ITD] ITD
  • Member since:
    02-28-2012

View Postbjshnog, on Feb 10 2014 - 06:28, said:

 

Not even close to possible. Wargaming does not track your framerate, neither does your WoT client, and there is virtually zero chance that anything like that would be implemented. Plus, different people deal with drops in framerate in different ways. I, for example, have played at 10fps and found no decrease in performance at all. I see you have only 2317 battles. If you play a decent amount of battles, your 'bad' stats should be washed away quickly if you are as decent as you claim you are (based on the stats I can see on Noobmeter, you are definitely improving).

Also, this thread is about WN7, which was replaced with WN8 a few months ago. ;)

That's too bad. So I suppose the framerate indicator is just for your own knowledge. Thanks for the info, I though WN8 was still in beta.  



nathanater72 #2480 Posted Apr 25 2015 - 18:33

    Private

  • Players
  • 8042 battles
  • 6
  • [CDEV] CDEV
  • Member since:
    01-03-2013

https://www.youtube....h?v=IN9J-f3ci0M

 

watch otter... just watch







Also tagged with statistics, stats, noobmeter, performance, efficiency, metric, compare, quality, rating, Garbad

1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users