Apparently I've fallen victim to the 'no tone of voice in electronic posting thing.'
or maybe I just did some super crappy writing. David and Peetzakilla: I'm sorry for not being more clear.
David wrote:
And that pretty much says it all, IMO. Let's not care what the facts are, let's not engage in a reasoned and reasonable analysis.
Which is fine to say, but not at all what I'm talking about. My fault for writing while bouncing the baby. I may deserve that, so, wincing, I'll try to be more clear.
What I'm saying instead is that there are places which are 'safer' and places which are 'less safe' but the boundaries can be quite close statistically. If you look at the rest of my post, I think you could see (reading it charitably, maybe) that what I'm talking about has some thought and analysis behind it. Maybe it really is that terribly written.
One other poster got what I was talking about with lotteries and expected payouts (if you'll forgive my petulant whining). I don't want to live in Darfur. But the argument from common sense that Peetza refers to is problematic, because common sense lies to us quite often and we are in general blind to its deficiencies.
Anyway: you consider your stats and you make your plans, but you make your plans based on stats that don't really matter. You make your move to Smallville where it's super safe, quit wearing your gun and then Joe Nutso shoots up the church, or feed store or whatever and you're unarmed because you believe your stats and didn't realize you were in a lottery. This is why I say "We don't care what the crime rate is" because of the asymmetries in payouts. There are places where you can expect to 'win' the crime lottery a lot more often--statistically, NYC is more dangerous than Blacksburg, VA. Murder rate is triple--but forcible rape is a third of Blacksburg's. Robbery, though, is ten times worse in NYC. Littleton, CO is safer than NYC in a lot of ways. But statistically, due to massacres, Blacksburg and Littleton are worse for college and high school students.
So that's why I don't care about the stats in aggregate: because the analysis might not actually matter. Don't think that I'm rejecting analysis outright, David--I'm trying very hard to think about what sort of analysis might actually matter. Risk management isn't about "what my risk 99% of the time"; it's about risk of ruin, and it's not clear to me that my risk of ruin is any greater living in Cincinnati, OH than it is in NYC. Cincinnati's murder rate is about quadruple NYC's (using the per 100,000 rates.) I lived in NYC for some time, and there are areas I would not go to, just as there are in Cincinnati--so my position, while that of a skeptic, is not that of a totally naive skeptic. But I'm just considering stats. If I add in 9/11, NYC has a higher murder rate over a longer time period than Cincinnati. But of course I shouldn't do that--it's common sense not to. Or is it?
(all crime stats, by the way, coming from Area connect. If'n you have better sources, educate me. Cincy vs. NYC located at:
http://blacksburg.areaconnect.com/crime/compare.htm?c1=Cincinnati&s1=OH&c2=New+York&s2=NY)
Peetza: here's a small attack on the problem of common sense. It's not attacking you since it's a total strawman; it's simply illustrative:
A:
Everybody knows NYC is a dangerous place--that's just common sense.
But, not quite A: Except NYC is actually much safer than national averages are for crime stats (click the link.)
So which is it that we are to believe (if in fact you think A was actually common sense)? You say you know the numbers put out by the Feds. OK. What time period? And why that time period? Do they chuck outliers? Is it as reliable as, say, the CPI is for inflation?
David, does this seem more reasoned and reasonable? Maybe problems in epistemology don't belong here but they seem germane to the OP.
