Statistics aren't there to make us feel differently about a loss.  They do help to put fear in perspective.  There is something inherently more terrifying about a plane crash than a car crash, but that doesn't mean it's actually more dangerous on a plane.  Statistics just remind us of that, I don't see it as belittling the loss, a life is still a life and even one life lost is too many.

Who flies?  Does everybody fly?

No.  The higher you are up on the economic ladder, the more likely you are to fly.  Poor people fly far more rarely, if at all.  People tend to see more of others in their own socioeconomic class.  So - the people on that flight were most likely in the upper half of the income scale.  People in the same income class associate with each other more than those in a different class.  So "everyone in Seattle" is likely "everyone I know in Seattle".  Who do they know?  Most likely, people in their own economic class which is subset of the Seattle population.

The best example of the this is viral marketing in theory and in practice.  In theory, if you email three people something and then they each email three people and so on, you get a rapidly expanding geometric progression.  1, 3, 9, 27, 3 to the power of n.

In reality, some of those people know each other, so within a few degrees, people will be sending each other the same emails and many people will receive multiple copies.  I imagine it as a mandelbrot image where there are multiple degrees of association (school, church, work, domicile, income, sports) inside and relatively few types of association outside.