What these sigmas refer to is the number of standard deviations from the mean of a probability distribution at which an outcome will, probabilistically speaking, occur, with each higher sigma (number) being increasingly less likely than the last. According to Mr. Viniar, what happened in 2008 was “comparable to winning the lottery 21 or 22 times in a row.” LTCM’s ten sigma in 1998 was, likewise, an event that should have occurred roughly three times in the life of the universe. That these two events happened a mere nine years apart shows us that such claims are nonsense. It also tells us why Nassim Taleb has a huge problem with the idea of risk management and financial engineering.

Mark Blyth, Austerity, loc. 104