Again, the nature of this conversation is "why" something happens. So saying it happens and it's statistics is not a rebuttal to me suggesting a possible explanation as to why it might happen.
This will be my last response in this thread - it seems to be degrading into a flame war. If you want more info email me (my address is in my profile).
I wasn't saying 'statistics' as a magical name, I was just using it as shorthand for the explanation that it provides. Frankly, I don't have the time or inclination to type up a half hour lecture on stats (especially since there are thousands of others who have already done so far better than I could).
But, here's the 10 second explanation:
Many easily measurable outputs (height, weight, wealth, etc) are functions of a large number of roughly randomly distributed inputs. It's very unlikely that all these inputs are lined up in such a way that the observable output is maximized or minimized. It's much more likely that they will, in fact, on average cancel each other out. Given a sufficient number of trials (i.e., a large enough population), this results in a normal/guassian/bell-curve distribution of the observable output.
Given other distributions/correlations of inputs you get the other common distributions of outputs (lognormal, boltzmann, exponential, etc).
