You can't talk about odds without talking about returns. 1-in-5 is f'ing fantastic if a $1 bet pays $1000.

True, but if someone is going to go to the trouble of planning and financing an attack, I imagine they'd want a better than 20% chance of success. Especially since one attacker being caught could ruin the chances of all the others.

WHY? Really, how much do you think a PETN bomb costs? Re-read the article: nothing that happens in response to these attacks fails to help the terrorists. They send 5 people and one blows up a plane: huge positive ROI. They send 5 people and nobody blows up a plane, but commercial air travel is disrupted to the tune of hundreds of millions of dollars: huge positive ROI.

You have a good point, if you measure it in terms of financial damage, the terrorists win either way.

A side point: I suspect bombs are at bottom of the price chart in terms of financing an attack. More likely, the most expensive part is finding someone stupid enough to blow themselves up, and getting them to the right place at the right time. The cost of doing this undetected would probably increase non-linearly with the number of people involved in your attack.

Which is why Khalid Sheikh Mohammed never gets on the suicide flights. Terrorist management is expensive. Terrorist suckers are cheap, and if the history of the Iraq and Palestinian experience is an indicator, will get cheaper as AQ gets more operational experience.

The good news for our side, is that these people have to be simultaneously stupid enough to blow themselves up, yet bright enough to make it through security. This is why El Al's approach to security (personal interviews with all passengers) is effective - you'd have to be close to genius level to get through their screening without triggering a (real-world, not TSA-Pansy) pat-down and luggage check - at which point, if you were that bright, you probably wouldn't be volunteering for kaboom duty.

Or they could send someone through multiple times, giving them a prepared briefcase each time, but only put a bomb in it the last time, but telling them that there was a bomb each time; eventually, even an idiot will get used to it, and stop triggering their screening. Hell, depending on what sort of records they keep, they might eventually interpret his nervousness to just be who he is (dislikes flying, say), and wave him through anyways.

If my (discredited by most on HN, but I still hold to it) theory regarding El Al security screening is correct, they aren't principally looking for people who are nervous, or acting guilty (though they'll pull those people aside as well) - rather they are trying to identify individuals that fall within the category of people that might bomb El Al, and then pull them aside for selective screening.

Your average Muslim individual would have little chance convincing a screener they were an observant Jew, and would always fail regardless of how many times they attempted.

By the time they had learned a patter that would get them through security, they would probably be bright enough to say "Hey, why am I wasting my life by blowing myself up? I should probably _run_ operations instead of being in them"

The supposed success of El Al isn't (in my opinion) that they can stop a broad community of attackers, rather it is that they are effective in profiling quickly, and passing through 95% of the passengers quickly, and only subjecting 5% to selective screening.

I'd be interesting in hearing of the experiences of any Palestinian or Muslim passengers that have flyed on El Al for anecdotal confirmation of this.

Defenders must prevent every possible attack.

Attackers must evade the defenses just once.