Hacker News Reader

Comment by JoeAltmaier | Hacker News Reader

...and twice as likely to commit suicide, than be killed by anybody.

So this is the problem with statistics; these statistics apply to the average person. But there is no such thing as an average "real person". For instance MY chances of suicide are zero; and the chances of me being killed by a terrorist are far below the average person because of where I live and where I spend my time; the same goes for being killed by a police officer. This applies to MOST people for at least the "being killed by a police officer category". I think we all know the subset of society that is most prone to being shot by police officers: those that are breaking the law.

Agree in principle. But the chance that each of us will die by suicide is not actually zero. Its more closely related to the chance we will experience debilitating depression. Its a disease issue; no one is immune.

> So this is the problem with statistics

That's not a problem with statistics, it's a problem with using particular statistics to talk about a non-particular group. If you'd adjusted for people who don't break the law, people like you, who get shot on accident by a police officer, the statistics would be perfectly fine to compare to the threat of you dying at the hands of police vs terrorists. There's no problem with statistics in the way you describe them, just with bad statistics just like there's a problem with bad anything (math, engineering, policy etc)

It would be nice to assume that were true, but is it actually the case? And what laws were the breaking? Perhaps more interesting is how close the second category comes in terms of frequency: what's the probability for a black man with a mental health issue?

Reply on news.ycombinator.com