Voltron says: A well written, readable, article in Wired magazine explains how a mathematical abstraction became the basis for the explosion in credit derivatives. This particular formulation came about after I left Wall Street, but it falls into many common pitfalls. One of them is that correlation between two assets is not constant and in a market panic, all assets become completely correlated. Most books on "quantitative finance" only devote a few pages, at the end, on the model's weaknesses. In reality, that should be the subject of most of the book, because that's what good traders need to worry about. When I would interview job candidates, I would never ask them to "derive the model" . . . I'd ask them to list the assumptions of the model and then debate how realistic, risky and problematic those assumptions are.