Well, while CW was still alive I was working on some statistics regarding CW online war, and at the moment when I was finish.. CW was closed :(..
I didn't know was it for good until I have read it in post by Ming. I was waiting for a new map (kuban?) but now I see that I will have to wait for
some new war to emerge :P.. So, I have decided in meantime to post here what I have found out digging the depths of CW database dump that was so kindly
handed to me by Jirka (aka Vogon). Now , I feel it is too bad that I will not post it in CW forum, because data that I will present would be
commented best there, but I hope that some CW player are reading SimHQ also ..(as I see some of them do..:))
Also I hope that this will be interesting and for the rest of ppl.
First off all, to fully understand all this data you really need to be a CW player. CW had its history of Tank Vs Tanks battles, "balancing" issues,
etc, etc.. anybody who was in bellum also knows about what I'm talking here :o). And very important thing - war was decided on ground. Who had better
tanks , and better killing of tanks (in any form) was a winner.
Statistic is a very very tricky thing and it can show some good explanations of why did some things happened and also it can also be very delusional.
So I will show you some raw numbers and my explanation why I think some thing are like they are (IMHO of course..).
ALL data is regarding TANK LOSSES. How many tanks were lost by both sides and what has caused that loss. Because what is counting in CW is if tanks will win or loose,
I have focused only on this. Also it would take too much time to take ALL data out (like ALL ground units losses), and that data is not so important for achieving a victory (putting tactics a side for now). At the end basically what is important is will your tanks win the battle or not.
Ok let's start
---
First thing that you have to know is how much was some map played. So you can then understand and take in account why some planes are showing so much more
then some other types.
As you can see most of the time we were playing Moscow map (1941 and 1942) - some 60% in total. So we were dealing most of the time with early planes and
tanks (well.. here comes tank 'balancing' aka give-me-more t-34s -> but that is another issue)
----
Let me remind you, all data is regarding TANKS.
VVS Total losses 
OKL Total Losses 
here you can see looses by human (flyers), Air AI (ai planes), Ground AI(killed by enemy tanks) and percentage.
Basically I will make one remark here that goes to total game play.
TO MUCH AI KILLS (as you see on both sides!). Human decides in only (average) 25% of cases if some sector will be won!! Well.. that was the biggest surprise for me..
Basically we (players) are doing only some collateral damage in simulation of tank battles (and in a bad one - tanks are in IL2 very very badly simulated).
Also you can see that VVS gets 63% kills only by tanks! Later you can read why. Also VVS has only 9% kills by VVS air AI, but most of the time while VVS Air AI was
over the enemy tanks they were already dead (killed by VVS tanks and VVS PL).
OKL has 27% of kills by human and VVS 28%, later you will also see that most of OKL missions (52.53%) were flown by planes that cannot carry any bomb or rocket load (attack on tanks or defending our own bombers while attacking tanks (then you can take bomb and also attack tanks)). With F2 or F4 you can kill only light tanks (BT7 or T60(harder..)) while VVS had bombs and rockets for the same job.
27% of kills by OKL Air AI (over 9% of of VVS AI) is explained then by two things:
1. OKL tanks could not kill much VVS tanks -> OKL AIR AI gets them
2. OKL players could not kill much enemy tanks(in 52%) -> OKL AIR AI gets them
-------
Tanks Vs Tank lossesVVS losses by OKL tanks and VVS loses by OKL tanks (by type - kill ratio)
OKL losses by VVS Tanks 
OKL has lost 24913 tanks and VVS has lost 16274 due to enemy tank fire.
That is 1,53 kill ratio in favor of VVS tanks! Well, what can we say :)) VVS tanks are better :). I will go a little more in depth later.
Most of the time in CW we were arguing about tanks, and in the end general idea was this :
There was much more VVS tanks in real life during WW2. And of course I agree, BUT what was not taken in consideration by VVS players is that OKL tanks
were much more sophisticated killing machines because of superior tactics and training and quality. So we have seen that VVS tanks were added in numbers not taking in
account that IL2 is NOT a TANK simulation. And while some of ppl were talking Stalin’s words: "Quantity has a quality all its own" this kill ratio speaks different.
It means that Russian tanks not only had a quantity but had a *quality* also. So in my opinion it was totally unfair to add more tanks on disposal for VVS.
So.. what is the name of the silver bullet? :))) You can see it by yourself. All mighty T-34. What happened regularly was that that even only ONE t-34 could kill all
9 attacking tanks (check kill ratios of T-34). And because most of the time we were playing early maps, OKL did not have ANY type of tank that could successfully fight with T-34, witch has been available in too much numbers (regarding Moscow 1941 and 1942) in CW.
---------
Tank losses by Air AIVVS Tank losses and OKL Tank losses
Here German AI has 27% vs 9% of VVS air AI. This is big difference. There are few factor we should reconsider (I have sad a few words regarding this already.)
VVS has a better tank kill ratio, so most of the OKL tanks are already dead when VVS AI comes over them. Also VVS human pilots when escorting VVS bombers had a possibility to take rockets or bombs and help in busting tanks. OKL in at least 50% of time couldn't help much to their bombers (F2 and F4 cannot carry any bomb or rocket load), and most of VVS tanks were still alive (our tanks are weaker and couldn't kill them, especially a lot (regarding to 1941 and 1942) of T-34 and KV1) when OKL has arrived over them (target rich environment for AI).
On later maps situation for OKL gets much more better (possibility to carry bombs), but remember that 60% of time CW was played on early maps.
Also it is very hard to get a complete picture here because I cannot count kill ratios, because there is nowhere in database some info how many AI planes were used (not by type or any other data), so I only have data how many objects has AI killed.-------------
Tank losses by playersVVS tank losses, Kill ratios by plane
OKL tank losses, Kill ratios by plane
Ah, finally, some human action!
here I will leave a judgment to you.. you have all the numbers presented. just keep your eye on how much times plane was used and for what.
EBA - escort bombers attack (mission is to escort your bombers attacking enemy tanks in enemy territory (together with your tanks))
EBD - escort bombers defense (escorting your bombers defending your territory (sometimes with defending tanks))
ESU - escort SUM planes (transports)
ET - escort tanks (when your tanks are attacking)
INT - intercept (taking down enemy SUM)
BA - bomber attack (attacking enemy territory (tanks) flying bomber)
BD - bomber defense defending your territory (against tanks) flying bomber)
ART - playing as non pilot (shooter for example)
SUM - flying SUM plane (was only in few missions)
in table, column "No." is how many planes has been used by human players.
---
well boys and gals.. this was let's face it OKL view of the things :))), now I would like to hear a VVS opinion ((Ming - *pick me!* *pick me!!* *me!* *me!* *me!*) . .anybody?.. else?.. mph..ok Ming.. :P ), but the main thing is that now, AI's part in winning is TO BIG! Way too big.
Few thing I would like to see in any future online war.
In bellum and CW to win a square you had to kill certain number of tanks, and that is fine by me. But the main problem is as we see that tanks are mostly killed by AI.
So what needs to be is to remove tank battles. You can say that tank battles can be balanced – well good luck in doing so. This statistic can help , because you can see some kill ratios. But.. then..
So when one side is for example defending a territory, they should have no or only minimal number of defending tanks. Just a camp that should be defended, and ALWAS a possibility to choose human controlled tactical bomber (IL2,Stuka, bf110 - depends of side of course) , and then failure or success of attack would depends only on human flyers (BELLUM had similar concept). Less AI controlled bombing missions also.
That was in short :p.
well.. don’t know anymore, my brain has stopped
I'm sick, lying in bed with fever..
so ... over and out from me for now
driver
Previous Topic
Index