Jump to content


www.NoobMeter.com - WoT performance rating (player comparison tool)

performance efficiency rating comparison tool noobmeter www.noobmeter.com

  • Please log in to reply
1830 replies to this topic

Super_Salo7 #141 Posted Nov 19 2012 - 19:20

    First lieutenant

  • Players
  • 17657 battles
  • 848
  • Member since:
    04-09-2011

View PostNoobMeter, on Nov 19 2012 - 18:42, said:

I will write you a PM.
me 2 plz

FishingCat #142 Posted Nov 19 2012 - 19:20

    Sergeant

  • Players
  • 27184 battles
  • 113
  • [RCON] RCON
  • Member since:
    06-20-2012

View PostNoobMeter, on Nov 19 2012 - 19:01, said:


Except in case of TC/CW play, where it is hard to boost both. In fact, even just platooning with strong players will make my damage go down (while winrate goes up).


But what percentage of games are TC/CW?  Probably a very small percentage I would guess.  Even with platooning that has to be a small percentage of the whole.  So I'm not sure I can buy into that as an argument if we are making a rating system for the whole community.

Please understand that I am trying to be constructive, not critical.  As I said in my original post I doubt there could ever be a perfect rating system in this game.

FishingCat #143 Posted Nov 19 2012 - 19:21

    Sergeant

  • Players
  • 27184 battles
  • 113
  • [RCON] RCON
  • Member since:
    06-20-2012

View PostNoobMeter, on Nov 19 2012 - 19:15, said:

The adjustment is to consider SPGs as higher tiered than tanks of the same tier.

Ah, that would explain the result in that account.

TankHunter678 #144 Posted Nov 19 2012 - 19:25

    Corporal

  • Players
  • 3712 battles
  • 20
  • [4THBR] 4THBR
  • Member since:
    05-13-2011
NoobMeter
Performance rating:  1,306
Efficiency rating:  1,097

WoT-News
Efficiency rating: 885.67

So according to this, I am above average in performance, going by the "efficiency rating scale" on wot-news using your efficiency rating number I fall under average. Even though on wot-news they calculate my efficiency rating a good 200 points lower then your program does, placing me squarely in below average.

Shortcult #145 Posted Nov 19 2012 - 19:26

    Captain

  • Players
  • 16635 battles
  • 1,721
  • [-AK-] -AK-
  • Member since:
    08-21-2012
Haven't finished reading, so maybe it was already brought up, had to jot it down before it left my empty mind.  Would love to be able to look at win percentage by tier and/or type.

MIDNIGHTFENRIR #146 Posted Nov 19 2012 - 19:28

    First lieutenant

  • Players
  • 21038 battles
  • 848
  • [BULLS] BULLS
  • Member since:
    06-18-2012
sweet according to this im above average. boo yah.

Super_Salo7 #147 Posted Nov 19 2012 - 19:29

    First lieutenant

  • Players
  • 17657 battles
  • 848
  • Member since:
    04-09-2011
can i also suggest using damage dealt/damage taken ratio instead of damage dealt only..

raeff001 #148 Posted Nov 19 2012 - 19:32

    Staff sergeant

  • Beta Testers
  • 11028 battles
  • 395
  • [BOLD] BOLD
  • Member since:
    10-05-2010

View PostVerilogus, on Nov 19 2012 - 16:56, said:

You need to make it ignore the anomalies. It says my rating is 2,027 and it's obviously too low, it must be considering my tanks from the noob days like IS3, IS, KV1 etc...

LOL, was waiting for a unicum to whine because it was not raiting them like they wanted.  Thank you.

raeff001 #149 Posted Nov 19 2012 - 19:34

    Staff sergeant

  • Beta Testers
  • 11028 battles
  • 395
  • [BOLD] BOLD
  • Member since:
    10-05-2010
I'm good with mine.  Almost exactly the same.

P-1640
E-1630
WoT Calc-1609

Guess i'm a fairly consistent player.

Edited by raeff001, Nov 19 2012 - 19:52.


_VS_ #150 Posted Nov 19 2012 - 19:38

    Major

  • Players
  • 20869 battles
  • 2,736
  • [REL_3] REL_3
  • Member since:
    08-25-2011
Performance rating:  1,484
Efficiency rating:  1,390

same as WoT news

NoobMeter #151 Posted Nov 19 2012 - 19:54

    Staff sergeant

  • Players
  • 9 battles
  • 436
  • Member since:
    11-14-2012

View PostFishingCat, on Nov 19 2012 - 19:20, said:



But what percentage of games are TC/CW?  Probably a very small percentage I would guess.  Even with platooning that has to be a small percentage of the whole.  So I'm not sure I can buy into that as an argument if we are making a rating system for the whole community.
With platooning it is a fairly large percent, especially for the players with the best winrate.

But what I'm saying is that I don't think that even if they correlate perfectly that it would be a problem. Though to be honest I don't understand baseball, so maybe I didn't fully get your example. Can you translate it to basketball or hockey?


NoobMeter #152 Posted Nov 19 2012 - 19:55

    Staff sergeant

  • Players
  • 9 battles
  • 436
  • Member since:
    11-14-2012

View PostSuper_Salo7, on Nov 19 2012 - 19:29, said:

can i also suggest using damage dealt/damage taken ratio instead of damage dealt only..
Damage taken is not available in web results.

OOPMan #153 Posted Nov 19 2012 - 19:58

    Major

  • Players
  • 13872 battles
  • 4,042
  • [IOC_5] IOC_5
  • Member since:
    09-23-2011
Peformance: 1553
Efficiency: 1228

I Like It!



Super_Salo7 #154 Posted Nov 19 2012 - 19:58

    First lieutenant

  • Players
  • 17657 battles
  • 848
  • Member since:
    04-09-2011

View PostNoobMeter, on Nov 19 2012 - 19:55, said:

Damage taken is not available in web results.

it is available in WoT Stat. just think about it, if you do 800 damage and die in T10 heavy its nothing, but if you do the same it t4 med it a whole new story. I dont know how hard would it be to pull this data from a dossier file but i think it will be a better stat to use.

AquilaAce #155 Posted Nov 19 2012 - 20:04

    Staff sergeant

  • Players
  • 15527 battles
  • 461
  • Member since:
    02-22-2011
Peformance rating: 1771
Efficiency rating: 1652

I'm not entirely convinced...

However, it appears to me Performance is generally 100-200 above efficiency for reasons I don't know (although there are exceptions). Even our forum-famous 39%er has a higher performace rating than efficiency (but 868 under this system is still terribad).

There also seems to be some merit to this system that no one has mentioned: evaluation of low-tier noob farmers. The only clan I know off the top of my head that does this is UMB. Its players have absurdly high efficiency ratings and win rates but fall into the 1100-1300 range using Performance (just from a quick sampling of several of their members). I have a certain new "elite" player on ignore who plays Tier 1s and 2s, jsnazz. His Performance = 1170. Average, barely. I'm liking this aspect of Performace rating; it puts noob farmers in their place.

EDIT: Something I just thought of. To the OP: I know Performace is calculated across all vehicles, but is there an aspect of the algorithm you use that could be applied to only a single tank? If possible, I'm wondering if it would be a more precise evaluation of skill than efficiency is currently for a single tank.

Edited by AquilaAce, Nov 19 2012 - 20:10.


NoobMeter #156 Posted Nov 19 2012 - 20:04

    Staff sergeant

  • Players
  • 9 battles
  • 436
  • Member since:
    11-14-2012

View PostSuper_Salo7, on Nov 19 2012 - 19:58, said:



it is available in WoT Stat. just think about it, if you do 800 damage and die in T10 heavy its nothing, but if you do the same it t4 med it a whole new story. I dont know how hard would it be to pull this data from a dossier file but i think it will be a better stat to use.
Actually, I think it doesn't matter for the team if you take 0 damage and deal 1000 damage and your team wins, or if you take 800 damage and deal 1000 damage and your team wins.

This is the sort of stat FishingCat's example talks about - we would be doing double accounting.

NoobMeter #157 Posted Nov 19 2012 - 20:06

    Staff sergeant

  • Players
  • 9 battles
  • 436
  • Member since:
    11-14-2012

View PostAquilaAce, on Nov 19 2012 - 20:04, said:

Peformance rating: 1771
Efficiency rating: 1652

I'm not entirely convinced...

However, it appears to me Performance is generally 100-200 above efficiency for reasons I don't know (although there are exceptions). Even our forum-famous 39%er has a higher performace rating than efficiency (but 868 under this system is still terribad).
The reason is that the scales between the two ratings are not the same. They are apples and oranges. Each player has an apple and an orange. You can compare two player's apples (which of them are bigger) or two players oranges (which of them are bigger).

But comparing one player's apple with his orange is not very useful, even if often they are of a similar size.

Quote

There also seems to be some merit to this system that no one has mentioned: evalulation of low-tier noob farmers. The only clan I know off the top of my head that does this is UMB. Its players have absurdly high efficiency ratings and win rates but fall into the 1100-1300 range using Performance (just from a quick sampling of several of their members). I have a certain new "elite" player on ignore who plays Tier 1s and 2s, jsnazz. His Performance = 1170. Average, barely. I'm liking this aspect of Performace rating; it puts noob farmers in their place.
Good point.

NoobMeter #158 Posted Nov 19 2012 - 20:13

    Staff sergeant

  • Players
  • 9 battles
  • 436
  • Member since:
    11-14-2012

View PostTankHunter678, on Nov 19 2012 - 19:25, said:

WoT-News
Efficiency rating: 885.67
I'm getting you at 1096.45 at wot-news.

Super_Salo7 #159 Posted Nov 19 2012 - 20:14

    First lieutenant

  • Players
  • 17657 battles
  • 848
  • Member since:
    04-09-2011
the new PR award players who play higher tier tanks as they by default deal more damage. (am i wrong? have not seen the formula). what if i prefer T7-T8 tanks? changing damage to damage dealt/taking ration would fix this.

hammer91 #160 Posted Nov 19 2012 - 20:29

    Staff sergeant

  • Beta Testers
  • 35833 battles
  • 286
  • [SG] SG
  • Member since:
    10-18-2010

View PostNoobMeter, on Nov 19 2012 - 18:06, said:

Actually the aim was to make it mostly damage slanted. However, people doing high damage have high winrate, so the correlation between damage and winrate is very much there, so it appears to be quite winrate slanted.

That being said, if you only get pulled by your CW/TC mates while having high winrate, but not doing damage then you will get a poor performance rating.

Oh....I see now.

Since my winrate of 63% is artificially inflated by platooning/TC/CW (even though I play 98% solo), my poor 2200+ average damage per battle is used to correct my overall performance rating to slightly below average were it properly belongs.

That makes sense.

Like I said, please get XVM to adopt your system so I can get my red label!  I will be happy to let others experience the joy of that purple color in solo pub matches for a change.




3 user(s) are reading this topic

0 members, 3 guests, 0 anonymous users