Yes, precisely. The problem with Wall Street is that the models started failing at some point, but the people implementing them didn't understand them well enough to realize why and, therefore, dismissed the results. They just kept blindly doing the same thing, convinced that the data must be wrong. This is a classic instance of Emerson's observation that "a foolish consistency is the hobgoblin of little minds". The models were fantastic for a long time, but the fundamental assumptions and behaviors changed and the model failed to account for that. It's a huge error in application.