Jump to content


The new efficiency rating explained...

New Efficiency Rating Formula Calculations

This topic has been archived. This means that you cannot reply to this topic.
48 replies to this topic

AliceUnchained #1 Posted 27 May 2012 - 11:45 AM

    First Sergeant

  • Player
  • 22611 battles
  • 2,628
  • [UNICA] UNICA
  • Member since:
    10-18-2011
Alright, here it is.

All 5 aspects have a base efficiency calculation. What is important to know, is that the formula does NOT take into account just how much you 'scored' with a specific type of vehicle: It takes the total amount, and then generates efficiency scores based on percentages.

To clarify, let's take a very simple example for damage. 5 games played, 2 with a Tier VIII medium and 3 with Tier VIII Heavy tank. 40%/60%. A total of 1000 damage, all done in the 2 games with the medium tank. The formula does not calculate efficiency based on 100% medium usage, but on the 40%/60% ratio.

Using the Average Damage calculation as described below: 1000 * 2.5 / 8 = 312.5 Base Efficiency. Then we add (1 * 1000 * 40/100)/8 = 50 for Medium Tank Efficiency and (4 * 1000 * 60/100)/8 = 300 for Heavy Tank efficiency and we get 662.5. The exact same amount as you'll get using the website.

Average destroyed:

= _Destroyed * 20

+ ( 150.376 * _Destroyed * % / 100 ) for Tank Destroyers.
+ ( 150.376 * _Destroyed * % / 100 ) for SPG's.
+ ( 100 * _Destroyed * % / 100 ) for Light Tanks.
+ ( 100 * _Destroyed * % / 100 ) for Medium Tanks.
+ ( 153.85 * _Destroyed * % / 100 ) for Heavy Tanks.

N.B. Both 150.376 and 153.85 are rounded down. Actual numbers are 150.37593984962 and 153.84615384615. I'm sure they're divisions, but I just can't figure how they have been calculated.

Average Damage:

= _Damage * 2.5 / _Avg.Tier

+ ( 3.509 * _Damage * % / 100 ) / _Avg.Tier
+ ( 4 * _Damage * % / 100 ) / _Avg.Tier
+ ( 2/3 * _Damage * % / 100 ) / _Avg.Tier
+ ( 1 * _Damage * % / 100 ) / _Avg.Tier
+ ( 4 * _Damage * % / 100 ) / _Avg.Tier

N.B. 3.509 is rounded down from 3.5087719298246.

Average Detected:

= _Detected * 20

No added efficiency for Tank Destroyers.
- ( 20 * _Detected * % / 100 ) for SPG's. That's correct, a decrease in efficiency for SPG usage.
+ ( 200 * _Detected * % / 100 ) Suicide scouting never has been more rewarding.
+ ( 100 * _Detected * % / 100 )
No added efficiency for Heavy Tanks.

Average Defense Points:

= _Defense * 30

No added efficiency.
No added efficiency.
+ ( 100 * _Defense * % / 100 )
+ ( 200 * _Defense * % / 100 )
No added efficiency.

Average Capture Points:

= _Capture * 10

No added efficiency.
- ( 10 * _Capture * % / 100 ) Indeed, another decrease.
+ ( 2/3 * _Capture * % )
+ ( 2/3 * _Capture * % )
No added efficiency.

N.B. For both Light and Medium Tanks the / 100 has been left out. Not a mistake: Simply a rewrite from (200/3)/100 = 2/3

Hopefully, seeing this it will become obvious that the 'New Efficiency Rating' is a joke: Be it intended or not.

Edited by AliceUnchained, 28 May 2012 - 10:55 AM.


GoldStar #2 Posted 27 May 2012 - 12:02 PM

    Sergeant

  • Beta Tester
  • 13906 battles
  • 491
  • Member since:
    12-06-2010
Who makes up shit like this? You must have absolutely nothing else to do make before coming up with a E-peen formula with numbers like 150.37593984962 and 153.84615384615.!!

With a job, a wife and 2 kids, I estimate the chance I come up with this shit same the same as winning the big one in the lottery.  

Phooca #3 Posted 27 May 2012 - 12:12 PM

    First Sergeant

  • Player
  • 279 battles
  • 1,601
  • Member since:
    08-05-2011
While I like methods of the new system there are two points I think are utterly retarded

1: Damage done is not equal per tank type , why should TDs get a massive bonus to efficency for 500 dmg as opposed to a medium tank . Damage is damage regardless of tank type

2. Stopping a cap , mediums get a huge bonus for stopping the cap ,why once again it doesnt matter who stops it

Other than that I like the improvments such as capping and spotting meaning relativly little efficency for an arty player

ECBBailoutFund #4 Posted 27 May 2012 - 12:15 PM

    Junior Sergeant

  • Player
  • 17542 battles
  • 169
  • [ISM] ISM
  • Member since:
    07-18-2011
Sorry, does this not mean Heavy hitters get better eff rating because of damage?

I am not a fan of capping etc...but it was still part of the game.  Why promote camping by giving them on par eff points when all they do is sit on their behinds in base?

That formula for me just penalises the lower levels far too much.

Tanku49 #5 Posted 27 May 2012 - 12:17 PM

    Sergeant

  • Player
  • 18917 battles
  • 458
  • Member since:
    06-11-2011
how does it factor in ..held left flank for 7 minutes, was bottom tank as a slow TD, used my pimped out marder with 100% crew and half a ton of perks, died taking the hits to save the arty, dont use prem account so has to use the cheaper gun and ammo at tier 8, only gets to play Saterdays, and all the other relevent factors ?

loituma #6 Posted 27 May 2012 - 12:17 PM

    First Sergeant

  • Beta Tester
  • 1207 battles
  • 2,481
  • [NOFUN] NOFUN
  • Member since:
    12-14-2010
Open Lianna's thread, use forum search function to look up "troll" and see the related replies

anakreonts #7 Posted 27 May 2012 - 12:20 PM

    Sergeant

  • Player
  • 20489 battles
  • 364
  • [EXEED] EXEED
  • Member since:
    07-12-2011
Gj. Happy to see that people wants to know more than average player.
To second poster, I guess you never had any hobby... For someone it's plane modelling, for another one it's numbers. No one makes you to calculate, so don't trash others people work

Edited by anakreonts, 27 May 2012 - 12:21 PM.


Grabarz19_PL #8 Posted 27 May 2012 - 12:20 PM

    First Sergeant

  • Beta Tester
  • 10978 battles
  • 4,430
  • [NOFUN] NOFUN
  • Member since:
    09-04-2010

View PostZombot101, on 27 May 2012 - 12:17 PM, said:

how does it factor in ..held left flank for 7 minutes, was bottom tank as a slow TD, used my pimped out marder with 100% crew and half a ton of perks, died taking the hits to save the arty, dont use prem account so has to use the cheaper gun and ammo at tier 8, only gets to play Saterdays, and all the other relevent factors ?
It factored in on your Global Win Ratio. It's also shown for a reason.

Stock tanks are always a downside. If you're a good player you will easily fix the rating as soon as you upgrade your tank :)

Tanku49 #9 Posted 27 May 2012 - 12:34 PM

    Sergeant

  • Player
  • 18917 battles
  • 458
  • Member since:
    06-11-2011

View PostGrabarz19_PL, on 27 May 2012 - 12:20 PM, said:

It factored in on your Global Win Ratio. It's also shown for a reason.

Stock tanks are always a downside. If you're a good player you will easily fix the rating as soon as you upgrade your tank :)

Firstly i have no issue with people wanting to play a game to achive a ranking, BUT if your going to make that ranking mean anything you have to have a level playing field, each player needs to be judged on the same basis. In WoT that doesnt happen, and i can not see a simple way to achive this, butoth Win ratio and Eff are unable to reflect the conditions under which each player achives thier score.

TBH the best current measure of a player is how they perform in CW battles, there the variables are much less.

Grabarz19_PL #10 Posted 27 May 2012 - 12:37 PM

    First Sergeant

  • Beta Tester
  • 10978 battles
  • 4,430
  • [NOFUN] NOFUN
  • Member since:
    09-04-2010

View PostZombot101, on 27 May 2012 - 12:34 PM, said:

Firstly i have no issue with people wanting to play a game to achive a ranking, BUT if your going to make that ranking mean anything you have to have a level playing field, each player needs to be judged on the same basis. In WoT that doesnt happen, and i can not see a simple way to achive this, butoth Win ratio and Eff are unable to reflect the conditions under which each player achives thier score.

TBH the best current measure of a player is how they perform in CW battles, there the variables are much less.
If you have at least half accurate fair rating counted on everybody it works. I use it long enough to know. Keep an eye on my threads, I'm keeping my results and as soon as I hit a decent amount I'll post a thread about it. I'd love to see what you write then :)

It's all about the probability.

Tanku49 #11 Posted 27 May 2012 - 12:57 PM

    Sergeant

  • Player
  • 18917 battles
  • 458
  • Member since:
    06-11-2011

View PostGrabarz19_PL, on 27 May 2012 - 12:37 PM, said:

If you have at least half accurate fair rating counted on everybody it works. I use it long enough to know. Keep an eye on my threads, I'm keeping my results and as soon as I hit a decent amount I'll post a thread about it. I'd love to see what you write then :)

It's all about the probability.

Actually, personaly, i think its little to do with probability, yes there are random factors, but there are a large number of non random ones in play as well, and a number of the random ones can be reduced in influence, eg if you play Tier 10 MM will not keep putting you in as the bottom tanks in random battles.

If the game was chess then a ranking system would be easy, but with WoT you end up ranking those that play the pawns, the Knights, the queen, the king on the same basis, which simple doesnt make sense. Yes a pawn might end up playing a vital role, but its no matter who good the pawn it is still does not have the same options as the queen.

Grabarz19_PL #12 Posted 27 May 2012 - 01:01 PM

    First Sergeant

  • Beta Tester
  • 10978 battles
  • 4,430
  • [NOFUN] NOFUN
  • Member since:
    09-04-2010

View PostZombot101, on 27 May 2012 - 12:57 PM, said:

Actually, personaly, i think its little to do with probability, yes there are random factors, but there are a large number of non random ones in play as well, and a number of the random ones can be reduced in influence, eg if you play Tier 10 MM will not keep putting you in as the bottom tanks in random battles.

If the game was chess then a ranking system would be easy, but with WoT you end up ranking those that play the pawns, the Knights, the queen, the king on the same basis, which simple doesnt make sense. Yes a pawn might end up playing a vital role, but its no matter who good the pawn it is still does not have the same options as the queen.
And that's why it's the top tier tanks skill level that counts the most. XVM also takes that into account you know.

Right now I got 51 battles which proved the new XVM's "probability of victory" rating to be 71% accurate (I know it's too few battles still). That tells a thing.

Tanku49 #13 Posted 27 May 2012 - 01:43 PM

    Sergeant

  • Player
  • 18917 battles
  • 458
  • Member since:
    06-11-2011

View PostGrabarz19_PL, on 27 May 2012 - 01:01 PM, said:

And that's why it's the top tier tanks skill level that counts the most. XVM also takes that into account you know.

Right now I got 51 battles which proved the new XVM's "probability of victory" rating to be 71% accurate (I know it's too few battles still). That tells a thing.


I think we are moving away from ranks into the realm of playing the game, the importance of rankings is they are the reason some people play the game, they like to get the high score, and more power to them, they will devote themselves to that goal, its how they have fun. But in WoT you have a lot, i would guess more than 50% who are much more casual players, the fun it being in a game about tanks and getting to shoot at other tanks, and yes of course winiing the battle is more fun, but getting a high score is a by product of the fun. Currently these two dissimilar populations are judged using the same measure. This cuts both ways.

As to the XVM predictor, tbh i think i can guess-timate who will win about 50% from the opening moves made in the game, and that doesnt add to the fun. Secondly if XVM is that good, why bother to play any game it tell you you are not going to win, you might as well say yourself the cost of ammo and consumables. To me much of the fun is seeing the battle unfold around me.

Generalwulf #14 Posted 27 May 2012 - 04:48 PM

    Senior Sergeant

  • Player
  • 10801 battles
  • 923
  • [GUARD] GUARD
  • Member since:
    08-11-2011
we like numbers, numbers are good :Smile_popcorn1:

"N.B. Both 150.376 and 153.85 are rounded down. Actual numbers are 150.37593984962 and 153.84615384615.

I'm sure they're divisions, but I just can't figure how they have been calculated."

numbers that "rounded" are okay, and OP clearly states limitations to/of the research.

Whether or not WG got the formula right is an entirely different question :Smile_trollface-3:

tip #15 Posted 22 June 2012 - 07:38 AM

    Junior Sergeant

  • Player
  • 40415 battles
  • 213
  • [HAPE] HAPE
  • Member since:
    02-08-2011

View PostAliceUnchained, on 27 May 2012 - 11:45 AM, said:


Hopefully, seeing this it will become obvious that the 'New Efficiency Rating' is a joke: Be it intended or not.

Yep, it is not efficiency rating (ER) but good behavior rating (GBR).

Let’s say you’re a SPG and the last surviving tank of your team. It is a common situation with SPGs only on both sides at the end of the battle. Don’t go cap (punished) and don’t try to defend base because you’ll have to spot them when coming (punished). Just go somewhere and sit there and let them cap.

I hope someone will provide mod with old formula if new one will be used with the XVM. I can already see people doing things by the book and by the book only.

Disclosure: my ER is greater than my GBR.

Bobbydancer #16 Posted 22 June 2012 - 08:03 AM

    Senior Sergeant

  • Player
  • 39363 battles
  • 942
  • [GGG] GGG
  • Member since:
    04-05-2011

View PostGoldStar, on 27 May 2012 - 12:02 PM, said:

Who makes up shit like this? You must have absolutely nothing else to do make before coming up with a E-peen formula with numbers like 150.37593984962 and 153.84615384615.!!

With a job, a wife and 2 kids, I estimate the chance I come up with this shit same the same as winning the big one in the lottery.  

Drop the wife thing and get to preciate some good numbers. She's obvious draging you down..

Leadbucket #17 Posted 22 June 2012 - 08:21 AM

    First Sergeant

  • Beta Tester
  • 14510 battles
  • 6,161
  • [T-D-U] T-D-U
  • Member since:
    12-29-2010

View PostGoldStar, on 27 May 2012 - 12:02 PM, said:

Who makes up shit like this? You must have absolutely nothing else to do make before coming up with a E-peen formula with numbers like 150.37593984962 and 153.84615384615.!!

With a job, a wife and 2 kids, I estimate the chance I come up with this shit same the same as winning the big one in the lottery.  
Some people do art as a hobby, some play chess, some play sport, some play Sodoku or other mind challenging games and others just sit and vegetate in front of the telly.. to each they're own.
Just because you find math either difficult or boring does not mean everyone else thinks the same way.
AliceUnchained's interest in math will most likely keep his brain sharp and in nice working order as he gets older. :Smile_great:

Edited by Leadbucket, 22 June 2012 - 08:23 AM.


japtank #18 Posted 22 June 2012 - 09:03 AM

    Sergeant

  • Player
  • 12713 battles
  • 283
  • Member since:
    04-20-2012
Victory on a battlefield has nothing to do with stats, but with clearsight and courage which can't be modeled nor translated in figures.

Making up coefficients is fine and all, as far as no one pretends they mean anything else than they are: figures.

Zamial #19 Posted 22 June 2012 - 10:34 AM

    First Sergeant

  • Player
  • 12159 battles
  • 2,831
  • Member since:
    03-29-2011
Since there is no way at all to use the stats to truly show how good or bad a player is why release one?  You are just confusing people.

zomos #20 Posted 22 June 2012 - 10:56 AM

    Sergeant

  • Beta Tester
  • 9845 battles
  • 308
  • Member since:
    01-04-2011
No decreases please and put some bonus for defence with TD, HT and SPG (smaller than more mobile tanks maybe a 1/4 of the med?) Meds role is to go back cause they are fast, HTs will need to be very fast to do so(IS8?). It will be OK then.

Only possible decreases can be team damage and team kills. But i don`t know if it is possible...