That is ignored, pattern matching is not good enough for you, you overcame pattern matching.

I wouldn't say that. "This looks cranky, it's probably not worth investigation further" is usually a pretty good heuristic. And, as you say, unless you actually know enough about the field to be able to be close to an expert yourself, it's often very hard to tell the difference between a logically consistent crank argument with no blatantly obvious mistakes and an argument for something that's actually correct. On the other hand, from the outside, people with minority views that are eventually vindicated also tend to look somewhat like cranks. So the only really reliable way to tell the difference between a crank and someone who should be taken seriously is to have someone who knows enough to find the non-obvious flaws actually is an expert look at the arguments.

(Incidentally, the "energy catalyzer" fails the "no obvious problems" test; if you say you have a working device and then won't let people perform independent tests on it, that's an obvious problem.)