Jump to content


A disturbing thing I have observed about my stats.

Warning LONG POST

  • Please log in to reply
59 replies to this topic

PowJay #1 Posted 16 November 2019 - 11:39 PM

    Major General

  • Player
  • 40500 battles
  • 5,620
  • Member since:
    09-07-2012

Everyone who knows me knows that I often bring stats into an argument discussion, but I hope that you also know that I try to demonstrate a link between more damage, more kills, more spotting and more wins, as opposed to simply picking on any random player that comes along. Some are not for turning on their opinions that everything is rigged, but we (the forum) did once win someone over to the light side and it was a great moment

 

I know I am not fantastic (not even close) but I do try to understand how this game works and I know my (tier) limits. I mainly play higher tiers for missions, although I am not too bad in some.

 

BUT....

 

Here's the thing. Why are some of my win rates far better than I appear to deserve (apart from the obvious, low-tier play)? I don't mean overall, but I mean some of my tanks are actually pretty poor and yet I keep winning. Small sample size is one thing, but what about 200 battles or more, for example? Surely a decent WR is not all just good luck.

 

I am going to discount TDs as they are easy enough to get positive (above 1) damage ratios. Sometimes three, or even two, shots is enough for Fire for Effect. I am going to look at my second-most played class: meds.

 

It appears that the highest win rate I have in a keeper med, with a fair number of battles, and in the top half of the tiers is the Cromwell. 324 battles. 57.1% WR (185 wins and 139 lost/drawn) That's a fair difference of 46 more wins than losses. But how does this come about?

 

The Wn8 is, a not entirely terrible, 1357. This is due to doing 630 HP damage per game. WHOAH! Hang on! 630 HP? This is a 700 HP vehicle! :amazed:

 

So, I know what it has to be: It has to be all those low HP tanks I am killing. Mmm. 0.87 per battle! It doesn't seem like it.


Ah, the Cromwell is fast, so it has to be that. I am scouting effectively. Well, damage with assistance is only 384, but that together with the gun damage is, at least, just over 1,000 average per battle. But still.

 

In the first instance, we do still have a small sample size and the figure of something over 1,000 games was mentioned to truly have a fair idea of ability in a vehicle, but I could understand this if I was not enjoying the tank and was hovering just over 50%, WR, but 57%?

 

Damage ratio is 0.94, but kill/death ratio is, at least, positive at 1.15, so I do kill more than I die, with a Survival rate of 24.7% (which is not my best, but far from my worst).


Medal haul from 324 battles? 1 Top Gun, 1 High Calibre, 3 Confederate, 3 Patrol Duty, 1 Defender. 1 Lehvaslaiho's medal. Only 67 Fire for Effect and 10 Hunter, but these were introduced some time after I started playing it.

 

I have 78% MoE, on the way UP, not (as with some) on the way down, so I am playing pretty well compared to others, but none of this makes sense the more I look into it. The basic principal of more than your own HP in damage does not apply and yet I have a good win rate.

 

So do I really suck that badly? How much of this WR is luck? I know that I quite enjoy playing the Cromwell, but have a look at the Cromwell B! 102 battles only, same crew, 637 damage (7 more), 0.81 kills/battle (0.06 less) and 372 average assist (12 less).

 

So I seem to have very similar performance in both tanks and yet the WR absolutely SUCKS on the Cromwell B at only 43.14% or 44 won and 58 lost.


The unfortunate thing is, that looking at a lot of my tanks with respectable win rates all display similar characteristics. Negative or low-positive damage and kill/death ratios, dull assist and less than 1 kill/battle and yet here I am sitting on a WR of 55% or greater. The T-34-100 is widely disliked, but it is one of my best tanks tier for tier. And yet I go and look and again: less than 1 to 1 damage ratio and only 1 kill/battle but I sit on 56.29% from 167 battles.

 

I do understand the simple concept of staying alive and being a threat, or at least a pain in the backside, and I am sure that there are some battles where blocked damage or simply stalling an advance has probably contributed to the win. Doing nothing obviously achieves nothing, but just being alive is a start.

 

The simple premise still stands. When I choose to use my own statistics to demonstrate the link between ability (damage/kills/assist) and win rate, I pretty much always can demonstrate that I do more of every one of those when compared to the poor player with his rigged-game accusations and 47% WR. I have also pointed out the stats of one or two far better players to show that it applies across the board.

 

The point of this thread is for those who like to talk about rigging to say that I am a bad player who is lucky to try and understand why I can be so meh! in some tanks and yet have reasonable positive WR. DAMN, but I hope it isn't just small sample size, because I am going to have a lot of 50% tanks when I get them to over 1000 battles. :teethhappy:

 

Footnote. To all those great players who want to troll me, it is hardly going to work when I am open and honest about all of this. The manner in which I use my stats for demonstration purposes is still valid unless you are one of the "rigged" brigade.

 

I am quitting now, but I will be happy to respond from Sunday. G'night all.



_cro_magnon #2 Posted 16 November 2019 - 11:48 PM

    Major

  • Player
  • 33358 battles
  • 2,854
  • [-MM] -MM
  • Member since:
    10-21-2012
TL;DR: rigged

NUKLEAR_SLUG #3 Posted 16 November 2019 - 11:50 PM

    Brigadier

  • Player
  • 34502 battles
  • 4,861
  • [FISHY] FISHY
  • Member since:
    06-13-2015

Maybe it's as simple as those are tanks that just particularly fit your play style so you do over and above what would be expected in them in comparison. 

 



Bulldog_Drummond #4 Posted 16 November 2019 - 11:55 PM

    Field Marshal

  • Player
  • 33576 battles
  • 10,751
  • Member since:
    08-10-2014

View PostPowJay, on 16 November 2019 - 10:39 PM, said:

 

 

I am quitting now, but I will be happy to respond from Sunday. G'night all.

 

All the answers to your questions are at

 

http://forum.worldoftanks.eu/index.php?/topic/670185-the-art-of-coarse-tanking-ch-1/#topmost


Edited by Bulldog_Drummond, 16 November 2019 - 11:57 PM.


BravelyRanAway #5 Posted 16 November 2019 - 11:55 PM

    Field Marshal

  • Beta Tester
  • 24491 battles
  • 12,252
  • [H_I_T] H_I_T
  • Member since:
    12-29-2010
There is also the point of you drawing fire in fast tanks that would normally be aimed at your team.....being hard to hit can be just as good as bouncing enemy shots in a Maus >>> which can result in less overall damage to your team >>>> which can lead to a win.

Bulldog_Drummond #6 Posted 16 November 2019 - 11:56 PM

    Field Marshal

  • Player
  • 33576 battles
  • 10,751
  • Member since:
    08-10-2014

Chapter 2 - Choosing a Coarse Tank

 

Coarse Tankers, although always enthusiastic and trying their best, are, unfortunately, completely useless.  Do not despair, for 95% of players are Coarse Tankers, so your skills will be at least equal to theirs.  What we need is an edge, and this is where choice of tank is critical.

 

The worst mistake a novice Coarse Tanker can make is to listen to experts and good players.  They will tell you about cunning tactics, the need to get to high levels to improve skills, about aiming for weakspots, that the Skorpion G is over-powered, and similar rubbish.  Ignore such advice, it is all rot as far as the Coarse Tanker is concerned.  You have only one aim, and that is to win, with the minimum amount of skill and effort.  Time wasted on watching experts on stream making things look easy, in the vain hope you could ever be like them, would be more profitably spent in the pub.  

 

So, given that you want to win, and that your skills are and always will be mediocre, the key decision will be choice of tank.

 

Once again, ignore the experts.  They know nothing about Coarse Tanking and their advice on tank selection is worthless.  What you need is a range of tanks in which any fool can and will harvest wins.  The golden rules for tank selection are:

 

1) Rarity.  Find some tank that is seldom seen.  It doesn't matter (within reason) how bad it is.  After a few dozen games you will get the hang of it.  As they have seldom or never seen it before your opponents will be clueless about its strengths and weaknesses.  This is the Zulu Principle - spend half an hour reading about the Zulus and you will know more about them than 99.99% of people on the planet.

2) Buy a rubbish tank with good gold ammo.  There are plenty of tanks which are universally despised but which have transformational gold ammo, e.g. the DW2 or PzIID or Pz38(t)NA.  Let them laugh at your Douchewagen.  The boot will be on the other foot after you have shredded them with 2100 dpm and 130 pen.  What you do not do is to buy a universally acknowledged OP tank like the KV1 since although it might look good on paper there are so many out there that everyone knows its weak spots.

3) Stick with the tanks that win for you.  If you can get 55% wins in a tank, there is something about it that works for you.  Keep it.  Do not be seduced by the lure of grinding further, that will only end in tears before bedtime.

4) Know your limitations.  Most Coarse Tankers struggle at higher tiers.  Get to a tier that works for you and stick to that as your limit.

5) Do not despise HE.  With a derp gun you will do okay even if you can't aim for toffee and know nothing about weak spots.

6) Armour is your friend.  Leave the soft-skinned tanks for the clever kids at the back of the class.  The only exception is if the tank has binos, camo, a good gun, and is invisible when sitting in a bush.

7) Any tank with preferential match making will give you a welcome edge.



PowJay #7 Posted 17 November 2019 - 12:00 AM

    Major General

  • Player
  • 40500 battles
  • 5,620
  • Member since:
    09-07-2012

Still up because the missus has decided that she wants a lift home (soon, thankfully). I hadn't noticed the Coarse Tanker thread before- probably because I was on holiday that week. 

 

A quick glance says quite a lot. GJ, BD. I'll have a look.



Balc0ra #8 Posted 17 November 2019 - 12:56 AM

    Field Marshal

  • Player
  • 74388 battles
  • 21,280
  • [WALL] WALL
  • Member since:
    07-10-2012

The thing about classes is, that you can just explore to find what works for you. Meds have several types. Support snipers, flankers, brawlers, and hybrids as in they can scout as well as bully. Cromwell is definitely one of those for me. And even tho they are the same, they might not end up the same as you don't have the same teams etc on both, even tho you have the same average damage and XP more or less on both. But with 100 battles, a small win streak can still turn that around still.

 

I love the B and the Cromwell. But for me, I have a bit better stats in the B with over 1K games in it. Tho to be fair it's done more platoon games vs the normal B. I doubt the WR would be that big if I did not have 300 ish 3x B platoons with it.

 

As for the T-34-100 I hated it, I had a bad streak in it. It's one of those tanks I've only aced on loses. But if it works for you, perhaps explore that style more. Much like the T-43.



BR33K1_PAWAH #9 Posted 17 November 2019 - 08:55 AM

    Lieutenant

  • Player
  • 7634 battles
  • 1,729
  • [WJDE] WJDE
  • Member since:
    04-11-2018

WN8 values are set by WN devs and they are artificial. What i mean is that before WN devs set WN* values for a particular tank they make their subjective opinion of how strong is it. So for a parple wn8 on Chieftain they will set higher DPG threshold than for a parple IS7.

And winrate is just a winrate.

Which brings to a situation where our player will have green WN8 in Chieftain with 60% winrate and parple WN8 in Black Prince with only 51% wins.

 


Edited by BR33K1_PAWAH, 17 November 2019 - 08:55 AM.


PowJay #10 Posted 17 November 2019 - 10:02 AM

    Major General

  • Player
  • 40500 battles
  • 5,620
  • Member since:
    09-07-2012

View PostBR33K1_PAWAH, on 17 November 2019 - 07:55 AM, said:

WN8 values are set by WN devs and they are artificial. What i mean is that before WN devs set WN* values for a particular tank they make their subjective opinion of how strong is it. So for a parple wn8 on Chieftain they will set higher DPG threshold than for a parple IS7.

And winrate is just a winrate.

Which brings to a situation where our player will have green WN8 in Chieftain with 60% winrate and parple WN8 in Black Prince with only 51% wins.

 

I understand what you say- I have seen comment that state that the Wn8 for a particular vehicle is being adjusted. I also am of the understanding that WR is part of Wn8, and if this is true, then the Cromwell Wn8 is boosted by the simple fact that it has a fair WR.

 

It just came as something of a shock recently that while I do have a winner here, this (and a considerable number of my tanks (not TDs) do not have the suggested own HP in damage and yet they do have a secure win rate: by which I mean that it would take more than a bad couple of days to turn that + into a -

09:05 Added after 2 minute

OMG. The shame. I've been "liked" by qpranger. :(

 

I get the message: time to quit the forum for good. 

 

 



AXIS_OF_RESISTANCE #11 Posted 17 November 2019 - 10:14 AM

    Staff Sergeant

  • Player
  • 4049 battles
  • 426
  • [S4LT] S4LT
  • Member since:
    12-21-2017

T6 is such an easy winrate pad with type64,85m.. etc almost everyone has bad view range add to that the camo and your'e invisible with much more room to move ,new players spending 500k on optics and using food plus a good crew to hit that 445 mark is not going to happen also you can pretty much pen everything on same tier with prem ammo from a t6 light\med ,on high tiers it becomes near impossible to pen armor .

Cromwell B  102 43.14% what happened with the cromwell b could explain the winrate mystery on the regular cromwell ?


PowJay #12 Posted 17 November 2019 - 10:43 AM

    Major General

  • Player
  • 40500 battles
  • 5,620
  • Member since:
    09-07-2012

View PostAXIS_OF_RESISTANCE, on 17 November 2019 - 09:14 AM, said:

Cromwell B  102 43.14% what happened with the cromwell b could explain the winrate mystery on the regular cromwell ?

It had crossed my mind. 426 battles in both. 229 won and 197 lost drawn. This is 53.7% which I'd see as much more realistic. 



Solstad1069 #13 Posted 17 November 2019 - 10:58 AM

    Colonel

  • Player
  • 45138 battles
  • 3,645
  • Member since:
    06-15-2013

Your recent WN8 is over 1400 and in a tier 5 or 6 battle this is very good. 90% of the players are worse and if you play with 5 skill crew the carry potential is high.
If you play tier 8-10 you are just one of the guys and need to be very lucky to carry anything.
 

Even with low survival stats you might be alive long enough for your team to do the rest.
At higher tiers this tactic will never work, the tanks have much more HP and people know when and how to take cover.


Edited by Solstad1069, 17 November 2019 - 11:00 AM.


BR33K1_PAWAH #14 Posted 17 November 2019 - 01:56 PM

    Lieutenant

  • Player
  • 7634 battles
  • 1,729
  • [WJDE] WJDE
  • Member since:
    04-11-2018

View PostPowJay, on 17 November 2019 - 12:02 PM, said:

I also am of the understanding that WR is part of Wn8, and if this is true, then the Cromwell Wn8 is boosted by the simple fact that it has a fair WR.

 

It is, but only by a small margin.



HassenderZerhacker #15 Posted 17 November 2019 - 02:32 PM

    Lieutenant Сolonel

  • Player
  • 31730 battles
  • 3,413
  • [1DPG] 1DPG
  • Member since:
    09-09-2015

View PostPowJay, on 17 November 2019 - 10:02 AM, said:

I understand what you say- I have seen comment that state that the Wn8 for a particular vehicle is being adjusted. I also am of the understanding that WR is part of Wn8, and if this is true, then the Cromwell Wn8 is boosted by the simple fact that it has a fair WR.

 

It just came as something of a shock recently that while I do have a winner here, this (and a considerable number of my tanks (not TDs) do not have the suggested own HP in damage and yet they do have a secure win rate: by which I mean that it would take more than a bad couple of days to turn that + into a -

09:05 Added after 2 minute

OMG. The shame. I've been "liked" by qpranger. :(

 

I get the message: time to quit the forum for good.

 

I'm convinced every account has some "blessed" and "cursed" tanks.

Can't explain how it exactly works, but it seems my performance can differ greatly even in tanks that are nearly identical.

 

I was quite ok in my T-34 but couldn't get the Type T-34 to work.

 

I have tanks in the garage which I feel to be bad, and I do badly in them, but they have a winrate of 60%

 

Others I feel the winloserate is undeserved. Take my IS-7, 126 battles in it, 35% winrate. Even if I don't play it too well, 35% seems totally uncalled for. I'm convinced this tank is cursed in my account.

I do more damage in my IS-7 than in my GW E100 arty (which sucks), but the arty has over 50% winrate.

In the ST-I, I do about the same damage as in the IS-7, but the ST-I has 56% winrate.

Same for the Obj 257.

 

It's all a bit strange.

 


Edited by HassenderZerhacker, 17 November 2019 - 02:39 PM.


shikaka9 #16 Posted 17 November 2019 - 02:38 PM

    Second Lieutenant

  • Player
  • 72145 battles
  • 1,306
  • [-SJA-] -SJA-
  • Member since:
    02-27-2013

Everyone: Oh no, he brought stats again :facepalm:

 

:trollface:



_PzKpfw_VII #17 Posted 17 November 2019 - 03:00 PM

    Private

  • Player
  • 2481 battles
  • 27
  • [B-MBT] B-MBT
  • Member since:
    02-07-2016

funny that i come across this thread while ive been questioning why the hell my Panther II stats are looking funky as hell... 

 

Spoiler

 

despite my Panther II being really weak, sub 1k efficiency which brings a tear to my eye... my WR has climbed from the early 50s% at around 20 odd battles to 60% and CLIMBING at 90 battles, which will probably plateau around the 59-62% WR after ive played over 200 battles and likely remain that way for good. i dont believe in rigging, MM tries to keep everyone at 50% as it states in the patents so its down to the players actions which influence whether their WR increases or declines. but this Panther II is just an enigma to me, i honestly cannot understand how i could play poorly in such a tank yet my WR increase by roughly 10%. at the moment im just wondering if this is just luck and then see my WR in the Panther II drop back down to 50-53% after ive played another 100 battles but if thats not the case then ill have to go back to rethink on how MM is working when i go into the battle queue


Edited by _PzKpfw_VII, 17 November 2019 - 03:04 PM.


HassenderZerhacker #18 Posted 17 November 2019 - 03:12 PM

    Lieutenant Сolonel

  • Player
  • 31730 battles
  • 3,413
  • [1DPG] 1DPG
  • Member since:
    09-09-2015

View Post_PzKpfw_VII, on 17 November 2019 - 03:00 PM, said:

funny that i come across this thread while ive been questioning why the hell my Panther II stats are looking funky as hell... 

 

Spoiler

 

despite my Panther II being really weak, sub 1k efficiency which brings a tear to my eye... my WR has climbed from the early 50s% at around 20 odd battles to 60% and CLIMBING at 90 battles, which will probably plateau around the 59-62% WR after ive played over 200 battles and likely remain that way for good. i dont believe in rigging, MM tries to keep everyone at 50% as it states in the patents so its down to the players actions which influence whether their WR increases or declines. but this Panther II is just an enigma to me, i honestly cannot understand how i could play poorly in such a tank yet my WR increase by roughly 10%. at the moment im just wondering if this is just luck and then see my WR in the Panther II drop back down to 50-53% after ive played another 100 battles but if thats not the case then ill have to go back to rethink on how MM is working when i go into the battle queue

 

many people on this forum will want to make you believe it's luck !

 

BTW: this is not played in a platoon, right?


Edited by HassenderZerhacker, 17 November 2019 - 03:13 PM.


PowJay #19 Posted 17 November 2019 - 03:17 PM

    Major General

  • Player
  • 40500 battles
  • 5,620
  • Member since:
    09-07-2012

View PostHassenderZerhacker, on 17 November 2019 - 01:32 PM, said:

 

I'm convinced every account has some "blessed" and "cursed" tanks.

Can't explain how it exactly works, but it seems my performance can differ greatly even in tanks that are nearly identical.

 

I was quite ok in my T-34 but couldn't get the Type T-34 to work.

 

It's all a bit strange.

 

Take a look at my T-34 and Type T-34 if you care to. The former I've played twice. Once was as a more experienced player for the new "A" line. Both times I've skipped the end of the grind with Free XP. The latter is a favourite of mine. 40.2% WR vs 56.4% 

 

I own nearly every tier V TD and have played every single tech-tree one to a positive WR. The one with the highest WR is the SU-85, but it does have one of the lower average damage rates.

 

The StuG IV has a not-terrible 525 HP average damage but (without checking) I am pretty sure that it's the ONLY tier V TD I have with a negative WR. It's also worth remembering that it has preferential MM and if it doesn't see tier VII then that damage is more significant as a percentage of the enemy health pool.

 

View Postshikaka9, on 17 November 2019 - 01:38 PM, said:

Everyone: Oh no, he brought stats again :facepalm:

 

:trollface:

While some agree and some disagree about stats/rigging and all that jazz, I can demonstrate that I consistently perform better than someone with 46% WR who is arguing about rigging.

 

But although I am overall classed as a "good" player on Noobmeter, it is only up to VII really and even that can be a stretch. The point in this thread is that my stats are actually quite surprising in that I do have some good win rates across the tiers where performance appears to be decidedly average. 

 

I am trying to make some sense of the hidden factor that seems to lead to a win. Blocking damage is one. I had a game in the E75 where I was useless on damage but took fire from four tanks for some time until I was destroyed. We won that, and I can see that tying up several vehicles while my allies were more constructive destructive probably helped to win.

 

Bravely' pointed out that you can achieve a similar result by attracting fire in a fast tank which may be one reason for more success in the Cromwell.



_PzKpfw_VII #20 Posted 17 November 2019 - 03:17 PM

    Private

  • Player
  • 2481 battles
  • 27
  • [B-MBT] B-MBT
  • Member since:
    02-07-2016

View PostHassenderZerhacker, on 17 November 2019 - 03:12 PM, said:

 

many people on this forum will want to make you believe it's luck !

 

BTW: this is not played in a platoon, right?

no platoons, in the 27k+ battles ive played since 2013, ive platooned for less than 500. ZERO platoons with this meme-roll






1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users