What’s In A Number?

Is using a killboard an accurate measure of PVP skill?  Using my own stats as an example, I have 55 ships killed worth 1993.21 million Isk with 10 ships lost worth 42.46 million Isk.  This gives me a 97.91% efficiency rating, which at a glance seems impressive.  The losses that I have inflicted upon others greatly outweigh my personal losses correct?  Well let’s take a look at what goes into that number. 

First off, yes I fit and fly cheap ships in militia.  Not because I am concerned about boasting about my 97.91% efficiency rating.  But rather to minimize the financial impact of a loss so that I don’t mind taking more and more risks with my ships.  You might label me as cheap as I don’t spend the extra millions of Isk to snag tech II modules or high meta-level options.  Really I just hate seeing Isk go to waste, and increasing my personal DPS by 5%, when most of my fights are gang warfare, doesn’t seem to make that much of a difference.  Also, now when I get blown up my loot is of little value to the looter.

My biggest problem with using the individual efficiency rating as a mark of skill is that it inflates the value of your kills while keeping the value of your losses constant.  For example, a gang of ten pilots catch and kill a 10 million Isk ship.  Each player then gets credit for the full 10 million Isk kill for their individual win/loss ratio, while the overall campaign kill board only gets credit for a 10 million Isk kill.  Therefore an additional 90 million Isk in kills was just credited on the kill board to all of the individual pilots involved than was actually lost by the opposition. 

The end result is what you will currently find on the Minmatar Militia kill boards.  Our current campaign for September 2009 against the Amarr gives us a 49.24% efficiency rating.  Yet if you click on any of the pilots involved in the various kills, you will find that nearly everyone is boasting an efficiency rating of over 80% with many players well over 90%.  So where are the pilots with the horrible loss ratios that are bringing down our overall numbers against the Amarr?  Well you are looking right at them! 

Going back to my 10 million Isk ship that was popped by ten pilots, if each individual pilot was credited with their portion of the kill then you would have each individual pilot credited with only 1 million Isk for their individual efficiency ratings.  This way the ratios are not distorted by large gang inflation and individual rating would more greatly produce ratings that show individual performance. 

Using myself as an example, I arrived at an average of 14.3 other players present on each of the kills that I was involved in.  I have had as few as two on a mail and as many as 30.  Once we reduce my total amount killed by dividing it amongst the other pilots present I arrive at a personal efficiency rating of about 66.8%.  Nothing worth boasting about but at least I am not below 50%.  

The kill board also creates other problems with losses that are not tracked but can have a large financial impact.  On an encounter with an enemy carrier a pair of our frigates made it a point to take out as many of the enemy pilot’s fighters as possible.  They managed to take out four fighters before they were popped.  These fighters have a cost of around 28-30 MILLION ISK EACH.  These two frigates, which maybe had a complete cost of around one million, managed to inflict around 120 million Isk in damages to the Amarr that day, yet were credited with no kills.  

Not that I am advocating things like drones and such showing up as kill mails, but I do think it’s worth pointing out that large financial loss can be inflicted on your opposition that does not show up on a kill board. Fun Fact: If the fighters are blown up while in the carrier they count on that mail, but not in space while they are actually fighting you! 

Logistics pilots also get the short end of the stick.  Their remote repairing modules and drones keep other ships in the fight and keeps up the steady stream of Pew-Pew.  They can be the pin holding the entire fleet together and often are themselves the target of enemy fire.  Some pilots in the militia channel boast about being creative with this problem and have developed various work arounds, such as carrying DPS drones instead of logistics drones or fitting a single long range weapon and shooting at primary targets to get credit for being in on the kill.  

I wish that I could come to some solution or conclusion that would make the kill board a more effective gauge of individual player skill and contribution in PVP engagements.  Dividing up the kills by the number of players actually present on the mail does appear to present a more accurate snap shot of an individual player, while encouraging smaller gang operations.  But, clearly in the case of support roles many pilots who fly crucial fleet logistics ships are continually left out in the cold.


~ by Centuri on September 16, 2009.

2 Responses to “What’s In A Number?”

  1. You run into this on any MMO where people try to track stats. Generally it’s a dps parser, maybe a heal parser, rather than the killboard, but it’s the same principle. There are things left out, given too much weight, etc.

    For example, on a heal parse… how often do they show state effect cures? Rarely, if ever. Yet if a healer isn’t curing those effects, you’re wiping the group or raid. Let’s say you’re in a raid and you have a healer devoted simply to curing and not healing…. he’s not on the parse as a result, but he’s absolutely critical to the success of what you’re doing.

    Sounds similar to the logistics thing you mentioned in your post — the “healer” isn’t on the parse (killmail) even though without him you’d have lost. Sad, but… that’s how the cookie crumbles.

  2. Well the kill board as a gauge really rips into logistics pilots specifically. Not only are most not getting credit for kills, but all off their losses are tabulated. The best and most dedicated logistics pilots out there must look terrible on paper (kill boards).

Comments are closed.

%d bloggers like this: