I think they are adjustable - or at least they can vary between countries: I've got a belt which has mostly been ok in UK airports, but has alarmed a few times in mainland Europe. In the US it seems to guarantee me an alarm (it alarmed the first 2 times I used precheck, since then I have always removed it, and I've worn it once recently to test again - but departing from LHR recently I didn't have issues).

FWIW there are two separate indications: there's metal being deteted, and there's a random selection. They make different tones. At SEA recently I went through, had the (metal) beep, went back and took off my belt, went through again: different tone, the agent emphasised that this was a random (and told the second agent manning the AIT the same thing).

Yes, that's a perfect example of randomness.

Note, I don't think your combinatorics are quite correct. You'd probably want to multiply that by 8, since the non-selected event could have happened at any of 8 positions, i.e. 8 permutations. But it's a bit late for me so I might be being stupid right now.

It has been way to many years since Probability and Statistics, so I am not even going to try for a precise analysis. Suffice it to say that the odds of 7 searches happening over 8 trials is very low indeed. I'd give the odds of being searched at less than 5%. So ehallison's numbers are probably on the high side :) (if one could say that with a straight face).

But lets stay with the 10% odds of being searched at any one entry. So there is a 90% chance of not being searched. The odds of not being searched seven consecutive times is around 47% if my late-night math skills are working. I'd say that most of us beat those odds, but we don't question whether the randomness of the situation exists.

When dealing with probabilities and extremely large numbers of travelers, all sorts of anomalies/outliers will exist. The rare person who gets a bunch of secondaries, and the rare one who has never gotten one.

Some WTMDs at some airports in the world have adjustable levels of "random" alarms for WTMD users. Some airports have had the frequency of the "random" selection for additional checking set at 10-25%. Other airports haven't used the "random" selector to even go after 10% in this way.

I don't know the setting most frequently wanted by the TSA, but various airports have had the setting changed from time to time.

Binomial Distribution

This distribution applies when the probability of success ("being selected") is the same for each independent trial. Like I said earlier, when the probability changes from event to event, the math is more arduous but can be calculated.

A common fallacy among non-statistician types is that the presence of unlikely events is evidence of a non-random selection process - yet Powerball has winners when the odds of winning are greater than 180,000,000:1. Except for the few examples where the selection process was manipulated by cheaters, lottery ball systems are designed to be truly random.

There is also the fallacy of dependent trials. Just because you were selected last time, and the time before that, and the time before that, and the time before that (however unlikely) does nothing to make selection on the next trip any more or less likely.

And to be really accurate, the selection algorithm is not truly random at all but is likely based on a seeded pseudorandom structure of some sort. Hopefully the WTMD algorithm is statistically random, but I'm sure that info is SSI.

Your lottery example is spot on. To add to the mystery of random selection: Some people win the lottery multiple times:

http://www.cbsnews.com/news/new-york...e-gas-station/

http://www.philly.com/philly/news/lo...y_history.html

The presence of unlikely events may be evidence of a non-random selection process. And it's the presence of unlikely events that helps some to determine if the selection process is random or not truly so.

A TSA staff member once explained to me that the percentage of random alarms is variable. I do not know if it changes by shift, airport, checkpoint, or all of the above.

The sensitivity of the metal detector is also variable. It is my understanding that the designated PreCheck detectors are set at a much lower sensitivity. I started to take out my large set of keys, including an electronic transmitter. The agent said, "believe me; you can leave those in your pocket." She was right.

The problem arises when a single metal detector serves multiple purposes. The same outfit, accessories, etc. will alarm at one checkpoint but not at another based on the designation of that particular machine.

I've experienced random quite a few times. It used to just be the hand swab, then it was the full body scanner, then back to the hand swab. As usual: they're "consistently inconsistent."

I know that the TSA relies on a lack of predictability, but I fail to see how any of these techniques prevent an act of terrorism. The TSA's answer, of course, would be that this is SSI. That means, "we don't know; we have no data."

SEA Precheck
 

I have never had any problem going through security at any airport except SEA. Every time I go through, I get the "you have been selected for additional screening" from the TSA agent. Anyone else find this happening to them at any particular airport?

Random (zone 5) selection with TSA Pre✓
 

Today, I went through Pre✓ at SFO international. I was randomly selected ("zone 5" is the term I keep hearing). I had to wait for someone, then he swabbed my hands, then I left.

An hour later, I had to go back to check-in.

Coming back in, random selection, zone 5. This time, they swabbed one of my two bags and I was on my way.

However, while today was the first time that was enough to prompt me to write this post, I am probably around 60% for this with Pre✓.

Does anyone have any idea how this works?

I tend to leave everything on my body (including some medical equipment) that might increase the amount of metal passing through the WTMD, but it is NOT the same thing that happens when you walk through with a phone.

I still find it's easier (in the sense that shoes and hoody stay on, laptop stays in bag, etc.) than the regular priority line, but if I'm just doing something stupid to trigger this, I'd like to know :)

Thanks

The metal detector has to beep and it is a high pitched beep much different from the one that is used when metal is detected by the detector. I had this happen this month, had to take wallet out, belt off, coat off, go through the <redacted> (AIT) and I was off. It was annoying but hasn't happened since.

You're not doing anything stupid. A few years back, around the start of PreCheck, the WTMDs were updated to include a random number generator in the firmware.

Yes, some people are receiving it more than once more than, but it's pure random coincidence. There is no secret hidden trigger that a TSO can operate to set it off.

A bump to the machine no longer makes any of them beep?

Most of these alarms were never set off that way, but at least previously it was a way.

What makes you say that it was "random"? If it was the odd beep from the metal detector, that is the so-called randomizer and it has nothing to do with you or with Covenant, the private security company at the SFO checkpoints. If you hit the randomizer twice in a row, that is simply bad luck.

If you are being diverted to a secondary for other reasons, there are other reasons.

If all that is happening is a hand / bag swab, more likely the former.

It's definitely the "random" check, but I've gone through at SFO probably 50 times this year, and I've hit the random check 25 times. It's not just about hitting it twice in a row.

I understand probability, which is why I've waited so long to post about this.

There's very little variance in what's on my body when I pass through the WTMD each time, which is why I'm wondering exactly how the system's implemented. If it's some probability factor multiplied by the amount of metal it detects on you, for example, then that could explain why I'm hitting it more than most.

Otherwise, I guess I'll resign myself to getting my hands or bag swabbed every other trip :p

No, that still happens, but that will only produce a regular metal alarm.

What I'm meaning to say is, not a single TSO can trigger the monotone random alarm.