Tuesday, 22 February 2011

3220 Killing Blows

Did you know that your combat log records every kill blow that happens within your current zone for both Order and Des?

I am not one to accept coonventional wisdom without some kind of look-see at the facts so in order to try and measure the power gap I have parsed my combat log from a typical night of RvR on Karak Norn and this is what is says.

The total number of kill blows was 3220. This was in 3.5 hrs so one killing blow every 4 secounds. Good stuff!!!

The combat log also tells you who was killed and by whom. There were 700 unique characters names who landed a KB or fell to one. This is more than I expected. Lots of people, or enough!, out and fighting.

I think these are very good numbers. That's lots of people playing and lots of fighting in the zones.

Next though, when an auto-attack lands the killing blow the combat log details the weapon that was being used....

[11/02/07][23:31:52] Knofel has been slashed by Aeoluw's Sentry's Blade of Reverence in The Cursed City.

From this I can work out how many people are using which type of weapon. Scenario weapons all have the same name for each RR band - e..g "Axe/blade/sword/whatver of Supremacy" for Order RR65 weapons, "Blah Blah of Malevolence for Destro RR90". There is also the "Imposing" wepaons. From this I can determine how many killbows high RR folks are getting compared to low RR.

Time for some graphs...

The graph on the left is the number of unique character names that landed a killing blow with each rank of scenario weapon and is my approximation at the renown rank distribution at the moment. There were 6 characters seen using a RR90 Scenario weapon.

The graph on the right shows the number of killing blows landed by each rank of scenarion weapons and is an approximation of which renown ranks are landing the most killing blows. RR90 characters landed 64 kill blows.

You can see killing blows for RR90 on the right hand side chart is disproprtionate to the distribution of ranks on the left and is out of sync with the normal distribution. This is show more disticntly on the chart below.

This is the ratio between the first two charts. It is number of killing blows per RR char. At RR90 a character lands 3.7 KBs on average. RR65 0.4 KBs per char on average.
This is a clear indication that RR90s are running rampant slaughtering as they go. I do not want to descredit them for their skills and abilities - because I have see some of these guys in action and they are good.
But the Kill Blow ratio chart shows a nice and steady increase up from RR65 to RR84 like you would expect. The drastic increase at RR90 is a measure of the effect of the power gap.
Remember I determined RR by KBs from scenario weapons but now knowing the names of those who use a RR90 weapon I can work out the total kill blows and their deaths. The ratio was 224 killed 30 deaths. 17 of their deaths were a single WE who I guess was out ganking. So, with her removed, the ratio would be 171 KBs to 13 deaths. Unbalanced.
The good news is there are plenty of people playing and fighting, averging a kill every 4 secs in the zones I was in. Most people are still sub 90 and their kill blow ratio is not out of hand. This means this game is still enjoyable... most of the time.
But RR90 Power means those players can run amok and when you run into them repeatedly it's probably not so enjoyable. Let's hope Myhic address this with 1.4.1.
There are a low number of kills from low RRs though so new players into T4 do need the boost they will be getting from 1.4.1, especially as player majority of players push into they high RR80s.
All being said, I still have great time out in RvR, even if we do need to pay special attention to certain players. Most of the time it's an even fight, all being said. The real problem for me is the RvR mechanic. The fun just grinds to a halt when it's siege time, but that's another story, and another patch.


  1. Interesting stuff, as always. Be interesting to see the class breakdown too - I imagine you were left with several magic-RDPS that you hadn't managed to RR-tag (because no auto-attack) so the data set will be incomplete, but good enough to start drawing conclusions.
    Probably, also worth saying the 90+ players would be roaming together mostly, so would be expected to perform well above the average.


  2. Yes, you are right Readda. Melee DPS get many more auto-attack kill blows. RDPS hardly any. So the RR data is skewed. But it's all I have so it is what it is.

    And you are right too about the RR90's banding together. You will be acquainted no doubt with Banatak and Crionic of - 46 of the 64 RR90 KBs came from them two. They were on a killing spree! Have these two and Orz always been such a force? I don't know.

    I have just realised writing this that I could do the same analysis on a combat log I have from last summer and see how it compares.

  3. Great idea. However did you consider that most RR 90 players:

    -Have more experience
    -Tend to play in groups
    -Have better target selection then new players
    -Know who is easy to kill
    -Are better at escaping/evading.
    -Play their character much more efficently then new players
    -Often times have better gear all around (Jeweley, Pocket items, weapons and armour, pots, talis)

    Not to say that 90 or 100 doesn't make a big difference cause it does. I've seen less talented players who are super powerful now due to hitting 90.

    My question to you is have you ever noticed how much better you play after you have played for a long periods of time over many days? And my second question do you think that plays a role in how well players do in WAR?

  4. As always, I'm amazed at your patience, dedication and the way you make sweet love to those numbers.

    I will remind you, however that a single player can die more than once every 4 seconds (I can atest to that personally) especially if they have a healer out for when they go down, already throwing a res before they hit the blood drenched grass.

    Also, a lot can happen in 3.5 hours. I doubt that you actually had 700 distinct players. If we were anything to go by, relogs are common when you're trying to establish a functional warband, with people logging in and out of healers, tanks and bombers.

    I salute you, my good man. It makes me feel good about myself when I read things with charts!

  5. @ Sara - I know 700 chars does not mean 700 ppl playing but I'd guess at least 500??? You are a goodie. I will be following your new blog.

    @ Anon - Quote "My question to you is have you ever noticed how much better you play after you have played for a long periods of time over many days?"

    Yes totally. Often I can miss a week of WAR and when I get back I mush keys like a nab, indeed.

    "And my second question do you think that plays a role in how well players do in WAR? "

    Yes again, totally. I mentioned in my post that I did not want to discredt the skill of the top players. In my comment to Readda that I asked if the top players have always been dominating kill blows and am left pondering by both your comments whether the new gear is going to make them even stronger or if the elite imbalance has always been there.

    I am doing a new analysis.

  6. Good to see you take the time to look at things from a statistical point of view, unlike most of us who just base posts on the "feel" of the game. But I'm sure I don't have to remind you that statistics can often be misleading.

    I share the concerns expressed by the commenters above:
    - Auto attacks will usually respresent the minority of killing blows
    - Not every career uses AA
    - AA almost won't show in larger AoE based fights
    - The sample is fairly small and from one evening (?)
    - You get players ganging up skewing the numbers

    I'd wouldn't want to draw any kind of final conclusion from these numbers. Though they do give a hint to what may be happening, we may be right about the power-gap. In which case, the latest batch of changes Mythic has lined up won't do anything to address this. They're rejiggling lower rank sets while those are clearly not an issue. Sigh.

    I, for one, can't agree with giving (the majority of) high ranked players the benefit of the doubt regarding their skill. I have had the opportunity to face many of them individually and I can't say that most of them are nothing above average. Which they certainly should be with their experience and knowledge of the game. Now I'm not saying that I'm a player of exceptional skill myself, I'm merely claiming that from tracking the past progress and success of many of the current high ranked players I can see most of them are riding the gear imbalance but are nothing special as players themselves. They can be called persistent farming little buggers but that's usually all.

    PS: I never knew you were playing on KN.