After reading Bang Bank you’re safe I thought it would be interesting to produce a graph of the Brady Gun score against the gun homicide rate per 100,000.
No real surprise: the gun score doesn’t seem to have much to do with your chance of being killed with a gun.
There are a lot of problems with this graph; correlation is not causation, the scale is a bit off, and California tends to dominate the eye, being all the way on the right.
But the biggest problem is that the underlying Brady score is measuring inputs instead of outputs. It is like trying to figure out if a school is any good by looking at how much money it spends instead of how well the students learn.