Originally Posted by
saulblum
Please quit it with the canard that airports are teeming with individuals looking to cause mass destruction.
There is not one person yet who has posted that there is a zero risk - all have acknowledged that there is a risk and that some security is necessary. Not my canard - yours.
Originally Posted by
saulblum
Checkpoint procedure could randomly assign half of the passengers to undergo no screening at all with no ill-effect on security.
It doubles the risk. Simple math. If you were correct, "...with no ill-effect on security," then no security would be equally as efficacious.
Originally Posted by
saulblum
A passenger is either a threat or they are not. You know, kind of like being a little bit pregnant.
Actually, a passenger is either a terrorist or not is the correct analogy. Absent a test, one cannot establish the "threat" status. One can, however, establish the relative risk of being (or not being) a terrorist.
Originally Posted by
saulblum
When over 99.9% of passengers present no threat whatsoever,
I understand probability - please do me a favor though - find me the 0.1%. That's the whole purpose of this isn't it? Ensuring that the 0.1% are found before they can become martyrs.
Originally Posted by
saulblum
how can you with a straight face say that the "risk of passengers varies"?
Because I understand probability theory? Each passenger has an assigned probability of being a terrorist and that absent any additional information that probability is the population probability. Add additional predictors and the probability no longer is the population probability but rather the probability of the refined cohort. If you can refine multiple cohorts you now have a distribution of probabilities across the population. This distribution defines the variance - its math, not politics or emotion. Granted, what those factors are is up for debate - though if you've ever flown ELAL you get a sense that there are valid factors.
Originally Posted by
saulblum
And let's look at some of the PreCheck perks: You can leave your shoes on. So the presumption is that someone who has not had a background check is more likely to be hiding a bomb in their shoes. Yet exactly one passenger out of the tens of billions of passengers who have flown worldwide the past decade has tried to blow up his shoes. But yep, we can conclusively say that someone who hasn't been convicted of a crime is less likely to do so.
Risk-based security my a**.
You are arguing the process not the concept. Let me reiterate, TSA is the most ineffectual piece of government monstrosity every devised by man. It is, unfortunately, what we are asked to tolerate in the interest of aviation safety. I'm all for getting rid of every single smurf. You?
But in the meantime, lets try the pregnancy game.
You have 100 people in a room. We know that one of them is pregnant - which one is it?
Each one gets two sequential blood tests, a urine test, and an ultrasound. We find that the pregnant person is person #46. We have assumed that every person in the room has an equal chance of being pregnant - equal "risk" so to speak.
Can we do better?
Of course. In the room of 100, there were 50 men and 50 women - do we need to test the men? Likely not. So instead of 100 comprehensive exams, we only do 50, the men save time and money.
In the room of 50 women - one is still pregnant. Ten are post menopausal - hence, they have a very low risk of being pregnant. Leaves us 40. Of those 40, 10 are pre-menarch - not a non-zero risk but certainly lower than the population of all women taken together. Leaves us 30. Of those, 20 insist that they use birth control. We know that BCPs are about 97% effective so their chances are less than those who do not take BCPs.
What do you do? You could test all 50 because they are women. Certainly each one doesn't have the same chance of being pregnant, but as women its not totally biologically impossible. On the other hand, you could only test those defined in medicine as being of child bearing potential, or you could only test those of child bearing potential not reporting the use of birth control.
In the end you have 100, 50, 30, or 10 people who qualify for the intensive high cost time consuming evaluation because they represent those at the highest probability of being pregnant.
Lets adapt this to the TSA:
One could forgo any testing and wait to see if one of the people gives birth - by default, that's the pregnant one - i.e. the terrorist.
One could test everyone with a simple inexpensive blood test - with a known error rate. This is WTMD orr WWM, or whatever. Or, one could test everyone using BKSTR and secondary patdowns and body cavity searches - depends how much sensitivity you want in your test, how much time you have and how much money you want to spend.
CAUTION - THESE ARE ONLY EXAMPLES AND DO NOT REFLECT ANY PERSONAL BIASES OR PREJUDICES - THESE ARE EXAMPLES!
One could test only the women - applied to TSA test only those who are non-frequent fliers or those not qualified under TT programs. Sort of like your 50% rule.
One could test only those of child bearing potential - i.e only non-citizen non-FF non-GEers.
One could test only those who don't use birth control - i.e. non-citizen non-FF non-GEers of Arab descent on student visas.
What TSA has done is said we can differentiate risk based on two criteria - Frequent Flier Status or background characteristics. This is no different than me as a doctor not doing an expensive and definitive range of r/o preg labs and clinical evaluations on an 85 year old man complaining of nausea in the morning. Just because he has nausea (getting on an airplane) its not due to pregnancy (being a terrorist).
Whether or not the criteria are correct is immaterial to the argument that differential risk exists and the cost of differentiation should be born by the individual.
Your arguments that each individual has an equal risk and moreover that non-screening in a proportion of the population because of this particularly because of your assertion that the risk, though non-zero is infinitesimally small is non-consequential to airline security are fallacious.
Its not about policy or process - its simple math and probability theory and market rationality