I’ll put aside the irony that many people on the right are borrowing a page that many on the left use on other issues by blaming a climate of opinion for the two Brooklyn cop murders. My own view is that the person you should blame for the murders is, um, the murderer. Jesse Walker of Reason has handled that issue quite well here.

But I want to address an issue of numeracy. The New York Post quotes retired policeman Vernon Geberth as saying that murders of two New York policemen “brought me back to the days of the 1970s when cops were routinely being assassinated.” Yes, he used that word “assassinated.” My guess is because it sounds more serious and scary than “murdered.” Either way, you’re dead.

I’ll address two questions: (1) Is the amping up of the threat scenario really justified? (2) Just how dangerous is it to be a policeman?

On (1), the evidence that we have says that it is not. A lone gunman who had murdered his ex-girlfriend decided to murder two policemen. It doesn’t sound like part of an organized plot or a broader social movement. It’s true that many people have expressed their upset about policemen. But that’s a far cry from going out and murdering them.

On (2), it turns out that being a policeman is not as risky as you might think. Take New York city as a starting point. Ed Krayewski writes:

Rafael Ramos and Wenjian Liu are the first NYPD officers killed in the line of duty since 2011.

That means that we’ve effectively had 3 years with zero murders of New York policemen. You might argue that we have a small sample size. No, we have few murders. The sample–actually, the population–of New York policemen is huge. According to the New York city government’s web site, “The NYPD’s current uniformed strength is approximately 34,500.” So that’s about 100,000 man-years of risk, and during that time 2 of them were murdered.

In The Concise Encyclopedia of Economics, in an article titled “Risk and Safety,” authors Aaron Wildavsky (deceased) and Adam Wildavsky have a table showing the fatality rates in various occupations. Check Table 2. Based on fatality data alone, being a policeman is seven times as risky as working in manufacturing. But it’s about 70% as risky as being a farmer and only 17% as risky as being a lumberjack.