I get your point (and it's a good one), but I don't think it's even possible to get 100% effectiveness in screening for weapons. The government can't even keep

I get your point (and it's a good one), but I don't think it's even possible to get 100% effectiveness in screening for weapons. The government can't even keep cell phones and drugs out of jail. From a Bayesian standpoint, having the ability to look out for certain individuals and subject them to more scrutiny should at least in theory (if it's done correctly) lead to better outcomes.