[DividesByZero]

It might be a small niggle, but the author is pretty dismissive of microeconomic theories of risk.

Despite what the author writes - "Of course statistically there’s no difference, so this isn’t a question of math or economics or intelligence; it’s a measure of your attitude towards risk.", risk aversion is a well studied concept in economics.

http://en.wikipedia.org/wiki/Risk_aversion

Most people have a preference curve that is risk averse, and for these actors it makes perfect economic sense to take the 10 and not the 20. 'Statistically identical' they most definitely aren't, just the expected value.