{"id":1611,"date":"2013-07-17T21:20:23","date_gmt":"2013-07-18T01:20:23","guid":{"rendered":"http:\/\/masspirates.org\/blog\/?p=1611"},"modified":"2013-07-17T21:20:23","modified_gmt":"2013-07-18T01:20:23","slug":"not-likely-to-catch-subway-bombers","status":"publish","type":"post","link":"https:\/\/masspirates.org\/blog\/2013\/07\/17\/not-likely-to-catch-subway-bombers\/","title":{"rendered":"How Likely is the TSA to Catch a Subway Bomber? Not very."},"content":{"rendered":"

\"TSA
\n(PHOTO: REUTERS\/Jason Reed. A TSA agent dons rubber gloves at Washington Reagan National Airport. Courtesy of Boing boing<\/a>.)<\/em><\/small><\/p>\n

On July 4th, there were several
\n Restore the Fourth<\/a> demonstrations in Boston, and we Pirates were proud to participate. They were good demonstrations, and it was encouraging to see how much the issue of mass surveillance<\/a> resonated with people.<\/p>\n

After the day’s events, I came across this tweet<\/a> from @AnonInMass, regarding TSA bag searches on the MBTA. We’ve all heard about the TSA’s bag searches (and we demonstrated<\/a> against them in February). But @AnonInMass’s tweet got me thinking: how effective are these searches? In other words, what’s the probability that a TSA bag search will prevent a bomber from getting on to the subway?<\/p>\n

Let’s spend a little time trying to answer this question, even if it’s nothing more than a back-of-the-envelope estimate. This estimate involves assumptions, but I’ll try to do this in a way that casts the TSA in a positive light. In other words, let’s try to give the TSA the benefit of the doubt, and see how well they do.<\/p>\n

First, we need to know something about the number of trips taken on the T. For this information, I’ll turn to the 13th edition of the MBTA’s Ridership and Service Statistics<\/a>. Page 5 of the pdf shows the average weekday ridership of the red, orange, blue, and green lines is 719,934 (this is “Total Heavy Rail” plus “Total Green Line”). As far as I know, the TSA only conducts bag searches at subway stations; therefore we should only consider subway ridership.<\/p>\n

Next, we need to know something about how ridership is distributed throughout the day; if the TSA were trying to maximize the effectiveness of bag searches, it would make more sense for them to search during peak hours, when more people are riding the T. I wasn’t able to find a breakdown of peak vs non-peak travel, so let’s say that 40% of ridership occurs during am peak, 40% occurs during pm peak, and 20% occurs during off peak.<\/p>\n

Next, we need a notion of how long a peak travel period lasts. I’ll assume that peak travel periods are four hours long, based on the peak travel periods shown in commuter rail schedules.<\/p>\n

Here’s what we have so far: 719,934 rides per weekday, where 80% (about 570,000) occur during eight hours of peak travel. From TSA’s perspective, this is the size of the problem space.<\/p>\n

Now, let’s turn our attention to the screening process. I wasn’t able to find any information about how often bag searches occur, so for the sake of discussion, let’s assume that TSA conducts two four-hour screenings per day, each during peak travel periods. (Probably a high number, but again, we’re trying to give TSA the benefit of the doubt.)<\/p>\n

The screening process involves wiping a rider’s backpack with a cloth, placing the cloth into a testing device, and reading back the test results. If the TSA can scan four bags per minute (one every 15 seconds), then they’re scanning 240 bags per hour. 240 scans\/hour during two peak periods per day gives 240 * 4 * 2 = 1920 scans\/day.<\/p>\n

By combining this with the ridership figures we derived earlier, we get the probability that a TSA screening will catch a bomber:<\/p>\n

\n (# scans)\/(number passengers) = 1920\/576,000 = 0.33%<\/p>\n<\/div>\n

Yes, all that for a 0.33% chance of success. Of course, that only holds if the would-be bomber submits<\/em> to a scan (as opposed to, say, turning around and walking to the next T station). If you take the bomber’s motivation into account, then we can make an arguement that the odds are lower than 0.33%.<\/p>\n

So far, we’ve been discussing what TSA subway screenings are capable of doing. We should also take stock of what the screenings cannot do. For example:<\/p>\n