The quanititative data isnt flawed. It's a very good way to aprehend a plane's impact on the Arenas, which is all the perk status is about. Especially when the sample has a broad base such as all kills / deaths from a whole year.
What about a spit16? This plane can dominate almost every other plane in the game the way it's modeled now. Why is the kill/death so low? Because it's so popular you can get a kill and die instantly. Newbie pilots crash them before getting gear up, or auger from inexperience, or rip the wings off.
The impact on the arena isn't told through the kill/death ratio. When there's a fight involving 4 spits, and all of them are spit16s, to me that says a lot more than a kill/death ratio. Regardless of the outcome (win lose or draw) the number of them used is the impact on the server, not the number used vs the number dead.
The interpretation of the numbers is flawed, not so much the numbers themselves. Too much importance is being placed on data that doesn't support it.
And of course you will have to add some common sense when interpreting the data.. For example planes with huge bombloads suffer from being used the kamikaze way vs fields, cv's, wirbels (the 14 has no ords)
Free planes with a "1337" tag are attracting newbies (The Spit 14 is not free)
Popular planes with a huge "usage" suffer somewhat from a lower overall skillbase of their pilots (Spit 14 is very rare)
You can also check plane vs plane numbers.. A long term sample shows that the Spit 14 has a hard time vs it's German equivalent in the MA
All of these are the very reasons I listed above as to why kills/deaths do not illustrate the impact of a plane on an arena nor the capability of a plane.
You can't say "you need to use common sense" because the argument isn't based on common sense, rather it's based on numbers. So you're saying "10/5 =2, but we're going to use common sense to tell you what the value of 2 is" -- it doesn't really work that way. 2 is 2. Either it means 2 or you're using the wrong argument.
I'm personally undecided on whether the spit14 should be unperked, but his constant shouting of this flawed argument gets old very quick.
He needs to take a class on research or on writing a thesis paper. He's using incompatible methods and conclusions.