i explained how your interpretation of your evidence was bad. the stats themselves aren't bad; i wholeheartedly believe they accurately depict damage done by the given classes. but your interpretation, which you distilled down to "drk is #1 in 2 out of 4 fights, therefore DRK>WAR in about half the end-game content" is wrong.
yeah, if all that exists is the #1 spot then you're right. but the fact is, all of those people, all 800 of them, have cleared their respective savage fights. once a fight is cleared, the rest is just gravy.
given that there are statistically more wars than drks in this sample, and that the average damage the war does is higher compared to the drk, it's a safe bet to say that war deals more damage, overall, than drk does.
my spoiler was me posting a personal story about a bard friend who, truth be told, probably spends too much time browsing yaoi porn on tumblr while she does literally anything else on her computer. including playing yaoi VN's. she's not a bad bard though. she's actually quite good when she's paying attention. we just make fun of her as being a bad bard, and i like grinding her gears because her reactions are always a hoot. anything else that you read into that statement is all on you.
again, that's on you. i didn't say that any tank was bad, or any tank was better than another. the one thing i did say was that war outdoes drk in dmg. which is statistically correct, overall. this doesn't mean they're aren't exceptions, but when you look at the aggregate whole, war is the tank that puts out the most damage on average. exceptions to the rule do not invalidate the rule.
again, the stats themselves are bad per se, but the way you interpreted them was. the stats are simply an accurate representation of damage per second done by given characters in a given fight over the course of a certain duration. that is literally all they are. you cannot look at those stats and say "drk beats war in 50% of the content" because you could just as easily say "drk loses to war in 50% of the content" and ultimately, both statements are meaningless. however, what you can do is get an aggregate of the given data, set limitations to it, and then parse it out from there to draw some limited conclusions.
my response, using this data, isn't that "war>drk" or some bs. my response, as noted above, is that war is present about 2.6 times as often than drk is in savage A9-A12, and tends to, on average with these given stats, pull higher dps than drk does. now, these are generalizations, meaning they aren't true in all cases. but they are true in most cases, and this is supported by the data.
this right here is the key difference; the conclusion i'm drawing is pretty limited. the conclusion you drew (DRK>WAR in about half the end-game content) is ridiculously broad, and the data you provided for support does not pan out in support of your conclusion.
nigga please. i got walls of text so large trump is gonna use my Holla Forums posts as the foundation for the Great Internet Border Wall he plans on erecting around brazil and france.
in retrospect, yeah. i think it's a matter of degree. the jump from dzemael or stone vigil to AV is pretty drastic. by comparison, going from AV to, say, dusk vigil is a comparatively larger leap, with many more moving parts. hell, even going from AV to old school AK was a big jump; i can't even remember how many times i had to solo the demon wall as my warrior because people couldn't understand the concept of "stand directly in front of the boss."
i also kinda have nostalgia glasses about AV, since i ran that place so damn much back when i first hit 50
i have already won.