Quants had almost nothing to do with this crisis. The problem wasn't that the computers knew too much, it was that they knew too little. They were fed best-case

Quants had almost nothing to do with this crisis. The problem wasn't that the computers knew too much, it was that they knew too little. They were fed best-case-scenario risk analyses. Of course the loans looked good if you assumed the housing market was going to keep going up.