While I'm very, very anti-TSA and this sort of thing wouldn't surprise me a single bit, saying you "found patterns in complaint data" is silly. Such data is ob

While I'm very, very anti-TSA and this sort of thing wouldn't surprise me a single bit, saying you "found patterns in complaint data" is silly. Such data is obviously biased from the start; we can do far better than this, with simple actual statistics on persons examined. But in the end, this is what you get when your "random screenings" aren't actually random in any way, not recorded for future analysis, and (due to the previous point) not audited for bias. You can make arguments why we wouldn't actually want random screenings in this stuff, but I think we all know that it's pure security theater regardless.