Jailer's very accurate observation is similar to my theory about crime and local news. If there are 100 violent crimes on a given day in New York City, the 3 or 4 most serious or significant get on the news during the 10 minutes or so (usually the first 10 minutes) devoted to local crime news. Then if, over time, violent crime decreases by 90%, 3 or 4 of the 10 or so crimes left still make the news. The result is that people, who are not very interested in statistics, still for the most part think the City is a very dangerous place, if their perspective is the "evil box" . If their perspective is personal knowledge of victims, let's say that 15 years ago I might have known personally 1 person per year who was a victim of a mugging, and now I know zero. What does that reasonably tell anyone about the risk of violent crime?> Absolutely nothing! The sample is far too small to be meaningful, yet most people base their perceptions either on personal experience or their view of the world through the Box. Neither is particularly useful.

The plain fact is that my neighborhood in Manhattan has about the same crime rate as Peoria. I doubt that many people living outside of New York City understand that.