October 27, 2018

612 words 3 mins read

I was wrong! And here's the lesson

A couple of days back I said the mail “bombs” looked like a false flag. But I left open the possibility of a crazy Republican. Boy, am I glad I gave myself that out.

I should have paid more attention to Scott Adams in those days. Scott used pure statistical reasoning to conclude that the packages probably came from a crazy Trump supporter: base cases. There are more crazy people than there are brilliant criminal strategists who hate Trump.

If the bomber theories come down to either a crazy Republican or a super-clever Democrat false flag mission, one of those things is far more common than the other.— Scott Adams (@ScottAdamsSays) October 25, 2018

Adams also pointed out another tell indicating a Trump supporter: DeNiro. A Democrat strategist wouldn’t think of DeNiro the way an obsessed Trump fan would.

The biggest clue to the bomber’s identity is that one bomb went to Punchy De Niro. Punchy is far more likely to be on a nutjob-Republican-bomber list than on a Bernie-supporter-false-flag list.— Scott Adams (@ScottAdamsSays) October 25, 2018

So, I was wrong, and Scott Adams was right. What makes my error more egregious is that I use base-case forecasting to determine probabilities all the time. This time, though, I ignored my training and went with my hope. I hoped the failed bomber was an anti-Trump zealot and not a crazy Republican. I let my emotions drown out the voice of reason, tamping down any temptation to think statistically. I used reason in the form of a theory-of-mind analysis, but only after I’d emotionally decided who was responsible. So, my biased reasoning tended to support my emotional preference, aka confirmation bias:

  • This event so close to the election would only hurt Republicans, therefore the culprit must not be a Republican but a Democrat seeking to hurt Republicans
  • The devices were sent in packages that look like the dictionary definition of “suspicious package.” That tells me the sender wanted people to think the packages were suspicious. In acting, this is called “indicating,” and it means telegraphing a message clumsily. If you want to take someone by surprise, you don’t telegraph your blow. Therefore, whoever sent the packages must have meant no harm to the recipients but only to seek attention to the event itself.
  • The bombs were either hoaxes or assembled by an idiot. They weren’t really bombs so much as things made to look like bombs. Again, telegraphing. This supported my theory of attention-seeking vs. mayhem-seeking.
  • Lack of attempt at concealment. The bomber must have known he’d be quickly caught. This told me he either didn’t care about the consequences or expected special treatment if caught. The third possibility was that he was mentally feeble and didn’t realize his bombs and packaging were like breadcrumbs leading to his crazily stickered white van. (And why do these crazy people always drive white vans?)

When you look at that list of reasons, it makes that I’d think the bomber was a Democrat. But only if the base case favors a Democrat or is, at least, neutral.

The moral of the lesson (which I am likely to ignore in the future) is: do the math first. Then do the theory of mind work. When you start with theory of mind, you must first have a mind in mind. This means you go with your gut and your biases and your prejudices. When that happens, every calculation you make seems to validate your biases and nothing contradicts them.

Nothing, that is, until the FBI arrests the guy who is a poster child for the base case.

P.S. Did you ever hear of Adam Carolla’s “Florida or Germany” theory?