During WWII the U.S. Navy conducted a study to try to figure out the best place to armor planes in order to minimize the number that got shot down. So they looked at all the bullet holes in returning aircraft, and got this diagram. Some people figured that they should armor the planes where there were all the red dots, because that’s where the planes were getting shot. Abraham Wald disagreed. See, he figured that what they were looking at was all the places a plane could get shot and still make it back. Thus, they needed to armor all the places where there were no bullet holes.
This is an example of what’s known as the survivorship bias. It’s a logical error of concentrating on the things that made it past some selection process, while overlooking those that did not, which can lead to incorrect conclusions.
So, let me ask you? What exactly does this chart prove? Because it certainly doesn’t prove that “the cost-effectiveness of Ganker [sic] is extremely unbalanced and there is room for rectification.”
- He picked 3 types of fits, while making no mention of what classifies as one of those fits, or why he chose them.
- He’s looking only at ships that were ganked, which means his data is subject to the survivorship bias.
- His chart fails to account for failed ganks, lost ships to anti-ganker activity, lost haulers (depending on the looting strat used), and failures to secure the loot drop.
- Bonus points for making a mistake in thinking that alphas can use large T2 guns.
- Bonus points for pulling his damage numbers out of his ass. Interestingly, they weren’t too high, but too low. And that was even when I tried assuming that he was basing his damage calculations off of what an alpha can do. So, I have no idea where he got his numbers from. I dunno, perhaps he assumed there was no concord pull? Whatever, I’m not going to spend the next hour trying to figure out what he did.
- And then, on top of all of that, he just straight up made a leap of logic in order to get to where he wanted to go.
And these are just the problems I’ve found with one chart and it’s accompanying analysis.
His observations were far from spot on, and his analysis is most certainly not proof that ganking needs to be nerfed. What his post is, however, is grade A misinformation. It gives the appearance of being a logical argument, and the fact that it falls apart under scrutiny is irrelevant, because it will still be effective. And if you want proof of that, look no further than yourself.
