Jump to content


Accuracy in WoT


  • Please log in to reply
102 replies to this topic

LordMuffin #1 Posted 05 April 2018 - 07:45 AM

    Field Marshal

  • Player
  • 48074 battles
  • 10,820
  • [-GLO-] -GLO-
  • Member since:
    06-21-2011

Currently accuracy in WoT is one of the least useful stats there is.

 

 

Setup: BIA, ventilations and food.

Strv S1 get 0.27 accuracy.

T-44 get 0.31 accuracy.

 

This small difference is hardly noticeable, and this is the difference between a supposedly very accurate tank with just a mediocre accuracy medium tank (T-44).

 

This accuracy of the STRV S1 means that you will, in a game situation (the VK is not standing still and have some slight cover covering part of lfp etc) bounce roughly 50% of your shots at 400m from a tank like the VK100P or any other kind of armoured tanks.

The reticule at this range is as big as the whole turret and a bit more off the VK100P.

 

Same sniping with the T-44 (and premium shells to get enough pen) would result in pretty much equal number of penned shots.

 

This difference in accuracy is just way to small to be a balancing factor in WoT.

 

The newly introduced Obj 268 v4 (which is supposedly inaccurate) have 0.37 accuracy.

A bit worse then Strv S1, but this is the difference between one of the most accurate tanks in the entire game, and a tank that was supposed to be let down by really bad accuracy.

The T110E4 sports a 0.33 accuracy. Difference 0.04, between a medium accurate tank and a tank with very bad accuracy.

 

In same situation as above, the Obj 268 v4 might go down to 40% pen.

 

The thing is, all accuracy stats in the game are way to compact to ever be useful as a balancing factor.

To fix this, WG should really make the difference in accuracy mean something.

 

Make accurate tanks like Strv S1/Leopard 1 and similar have 0.15 accuracy.

Give Obj 268 v4 (a very inaccurate tank) like 0.5 dispersion.

And then medium accurate tanks are at ~0.3.

 

This way, the stat called dispersion would actually be a useful stats to look at when judging a tank, instead of now being a statement about as relevant for judging a thank as the radio range is.



NoobySkooby #2 Posted 05 April 2018 - 07:51 AM

    Lieutenant

  • Player
  • 10748 battles
  • 1,744
  • [TFMB] TFMB
  • Member since:
    09-23-2011
I find that a lot of HE shells go wide, although I was in battle in my T14 and they seemed to go wide too

LordMuffin #3 Posted 05 April 2018 - 07:54 AM

    Field Marshal

  • Player
  • 48074 battles
  • 10,820
  • [-GLO-] -GLO-
  • Member since:
    06-21-2011

View Postkaptainkrunch, on 05 April 2018 - 07:51 AM, said:

I find that a lot of HE shells go wide, although I was in battle in my T14 and they seemed to go wide too

This have nothing to do with the accuracy stat, only with the distribution of shells within the reticule. There is a brumbarr thread somewhere explaining how the distribution works.



Thejagdpanther #4 Posted 05 April 2018 - 08:17 AM

    Brigadier

  • Player
  • 34032 battles
  • 4,338
  • [TKBS] TKBS
  • Member since:
    07-16-2012

You mean this "accuracy"?

 



Enforcer1975 #5 Posted 05 April 2018 - 08:22 AM

    Field Marshal

  • Player
  • 20480 battles
  • 10,537
  • [WJDE] WJDE
  • Member since:
    05-04-2014

View PostLordMuffin, on 05 April 2018 - 07:54 AM, said:

This have nothing to do with the accuracy stat, only with the distribution of shells within the reticule. There is a brumbarr thread somewhere explaining how the distribution works.

So does HE get worse distribution? 

Played my M4 Derpman yesterday and stopped counting the shells i needed to hit the bush i wanted...



STLR #6 Posted 05 April 2018 - 09:06 AM

    Warrant Officer

  • Player
  • 30399 battles
  • 730
  • [MIND] MIND
  • Member since:
    06-07-2013
does this word called "accuracy" has anything to do with wot? since 25% rng? :D

TheOddRogue #7 Posted 05 April 2018 - 09:08 AM

    Staff Sergeant

  • Player
  • 9985 battles
  • 443
  • Member since:
    08-08-2012
Anything below 0.40 dispersion coupled with good gunhandling is accurate, accurate in a relative term, in this game, the difference is not that significant between a tank with 0.3 and 0.4 accuracy provided they both have good gunhandling. 

Edited by TheOddRogue, 05 April 2018 - 09:09 AM.


HaZardeur #8 Posted 05 April 2018 - 09:10 AM

    Second Lieutenant

  • Beta Tester
  • 33257 battles
  • 1,157
  • Member since:
    08-14-2010
The "accuracy" values we see in the garage are "snakeoil"... much more relevant are the hidden soft stats.

Eaglax #9 Posted 05 April 2018 - 09:24 AM

    Lieutenant Сolonel

  • Player
  • 18973 battles
  • 3,481
  • Member since:
    01-12-2012

View PostSTLR, on 05 April 2018 - 09:06 AM, said:

does this word called "accuracy" has anything to do with wot? since 25% rng? :D

 

25% rng has nothing to do with accuracy....

Slyspy #10 Posted 05 April 2018 - 09:37 AM

    Field Marshal

  • Player
  • 14193 battles
  • 16,542
  • [T-D-U] T-D-U
  • Member since:
    12-07-2011

View PostEnforcer1975, on 05 April 2018 - 08:22 AM, said:

So does HE get worse distribution? 

Played my M4 Derpman yesterday and stopped counting the shells i needed to hit the bush i wanted...

 

No.

mortalsatsuma #11 Posted 05 April 2018 - 09:47 AM

    Lieutenant

  • Player
  • 13088 battles
  • 1,593
  • Member since:
    06-13-2014

I'm pretty certain that the Obj 268 V4 has the worst accuracy out of all the tier 10 TDs yet it has one of the best overall hit rates out of all of them.This was from a recent QB video discussing if the vehicle is OP or not (spoiler: it is and I didn't need his opinion to inform me of that). 

 

Basically accuracy is totally irreverent in this game but I feel at this point i'm just stating the obvious.  



Temid23 #12 Posted 05 April 2018 - 09:55 AM

    Warrant Officer

  • Player
  • 51305 battles
  • 655
  • [G-STR] G-STR
  • Member since:
    12-20-2013

Or Wg could just flat out list hidden parameters in the client so everyone could see why inaccurate tank from certain nation are actually more accurate then most accurate "sniper" tanks in the game.

 

I can already see "it will confuse new players" argument from WG but whats better? Confuse players or lie and false advertise to them.



XxKuzkina_MatxX #13 Posted 05 April 2018 - 10:19 AM

    Second Lieutenant

  • Player
  • 47690 battles
  • 1,379
  • [RU18] RU18
  • Member since:
    04-02-2016
Accuracy or dispersion of a certain gun in this game is a failed attempt on part of WG to appear sophisticated. It means absolutely nothing and i imagine it makes the game "interesting" in their opinion. My most hated aspect of this game is the accuracy model. You get your strv 103b with 5 skills crew, vents, food and directives and it's "bellissimo", now go snipe with your paper TD and miss at least 50% of your shots!!! Now imagine that 10 times worse with less accurate tanks.

UrQuan #14 Posted 05 April 2018 - 10:30 AM

    Lieutenant General

  • Player
  • 19398 battles
  • 6,115
  • [T-D-U] T-D-U
  • Member since:
    08-19-2011

I agree with you LordMuffin, the shell distribution matters more then accuracy when it doesn't differ much. My KV-4 topgun always felt pretty accurate despite it being 0,38, helped by the shell distribution preferring the center (shots going to the edge of the aimcircle seem to be a rarity on it). It tends to hit where I aim (aiming is needed on that gun tho, if you don't wait to aim, the shot does tend to go everywhere).

The German tanks got better accuracy on paper, but I found it's often offset by the worse shell distribution, which in turn leads that their actual accuracy doesn't differ much from what I experience from the KV-4. Only german guns that felt really accurate is the Stug topgun & Lowe gun, with those, I can snipe at pixels if I want. rest doesn't feel that accurate compared to the accuracy stat shown.

 

Edit: as pointed out, below, it's more a case of confirmation bias, as i'm simply so used to that topgun, vs other tank guns.


Edited by UrQuan, 05 April 2018 - 12:20 PM.


Dava_117 #15 Posted 05 April 2018 - 10:32 AM

    Major

  • Player
  • 18519 battles
  • 2,731
  • [B-BAS] B-BAS
  • Member since:
    12-17-2014

View PostTemid23, on 05 April 2018 - 09:55 AM, said:

Or Wg could just flat out list hidden parameters in the client so everyone could see why inaccurate tank from certain nation are actually more accurate then most accurate "sniper" tanks in the game.

 

I can already see "it will confuse new players" argument from WG but whats better? Confuse players or lie and false advertise to them.

 

That's due to a combination between the probability distribution inside the reticule and confirmation bias. All the tanks have the same chance to hit the center of the reticule (around 30% IIRC), so both KV-2 and STRV 103B have 30% chance to actually hit where they're aiming. But you expect the STRV to hit more, so remember better the missing shot in it, while you also remember more that 600m, 30km/h snapshot in KV-2. So you get to the wrong conclusion that KV-2 is more accurate than STRV, in example.

vcristi #16 Posted 05 April 2018 - 10:38 AM

    Staff Sergeant

  • Player
  • 65870 battles
  • 319
  • [3NRGY] 3NRGY
  • Member since:
    08-10-2011

Accuracy in a game with +/-25% RNG, lol.

Beside that the way of applying accuracy in game is plain wrong, the hits get randomly distributed in the circle, right?

But the circle have almost same diameter at 300m or at 30m - the only reason that you hit more when targets are close is just because the target is simply bigger and fills up the circle.

Maybe WG can explain how a shell can curb his trajectory so hard that in 10m it goes at the edge of the dispersion circle? 

This game has very little to do with real life physics - this accuracy implementation, totally unrealistic tank driving/handling and flipping tanks like pancakes are just laughable.



evilchaosmonkey #17 Posted 05 April 2018 - 10:43 AM

    Lieutenant

  • Player
  • 16618 battles
  • 1,745
  • [EIGHT] EIGHT
  • Member since:
    05-04-2013

The system needs to be clearer as to what real game play influence accuracy has.  Static target accuracy is reflected in the number, chance to hit a tank on the move far less so.

 

Without factoring shell velocity, it is little more than useless as a stat for a large proportion of real game play targets.

 

We need to look at hit percent for each gun first, then accuracy IMO.

 

edit: that of course, depends on the tank it's used on and whether it's designed for close up brawls or longer distance encounters.


Edited by evilchaosmonkey, 05 April 2018 - 10:46 AM.


Jigabachi #18 Posted 05 April 2018 - 10:57 AM

    Field Marshal

  • Player
  • 17858 battles
  • 18,492
  • [T-D-U] T-D-U
  • Member since:
    08-12-2011

View PostSTLR, on 05 April 2018 - 09:06 AM, said:

does this word called "accuracy" has anything to do with wot? since 25% rng? :D

There is no 25%RNG on accuracy or dispersion.

 

@ thread: I see the problem, but I'm not sure if another set of nerfs and buffs (be it accuracy or dispersion) is a good idea. Wouldn't it be better to balance via aimtime, so that Russian tanks certain tanks with big badaboom have to aim longer in order to hit further away targets and to expose themselves longer to prevent constant snapshots from behind cover?



Baldrickk #19 Posted 05 April 2018 - 11:01 AM

    Field Marshal

  • Player
  • 29506 battles
  • 13,746
  • [-TAH-] -TAH-
  • Member since:
    03-03-2013

View PostUrQuan, on 05 April 2018 - 10:30 AM, said:

I agree with you LordMuffin, the shell distribution matters more then accuracy when it doesn't differ much. My KV-4 topgun always felt pretty accurate despite it being 0,38, helped by the shell distribution preferring the center (shots going to the edge of the aimcircle seem to be a rarity on it). It tends to hit where I aim (aiming is needed on that gun tho, if you don't wait to aim, the shot does tend to go everywhere).

The German tanks got better accuracy on paper, but I found it's often offset by the worse shell distribution, which in turn leads that their actual accuracy doesn't differ much from what I experience from the KV-4. Only german guns that felt really accurate is the Stug topgun & Lowe gun, with those, I can snipe at pixels if I want. rest doesn't feel that accurate compared to the accuracy stat shown.

Unless there is a hidden stat that no-one knows about, shell distribution is the same for all vehicles.

The only difference being the area the shells are distributed over (accuracy/dispersion stat) and the penalties to said stat when moving or firing.



IncandescentGerbil #20 Posted 05 April 2018 - 11:03 AM

    Second Lieutenant

  • Player
  • 35666 battles
  • 1,443
  • Member since:
    11-24-2015
I always found one of the most accurate tanks was the kv2. You have to factor in preferred shell types when looking at notional figures like shell deviation. Type 5 is perhaps the ultimate. It hardly ever misses.

Edited by IncandescentGerbil, 05 April 2018 - 11:03 AM.





1 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users