I totally agree, this is more of a case of simple over-confidence and exessive risk taking than lack of mathematical understanding. Regardless of the quality of

I totally agree, this is more of a case of simple over-confidence and exessive risk taking than lack of mathematical understanding. Regardless of the quality of the model you are using there are still countless other factors that may impact your prediction quality. To me, the only way to mitigate this is to take it slow and plan ways for failure correction. Zillow is an example of greed and impatiance winning over common sense and risk-mitigation. If anything this could have been because of _too much_ higher level statistics and math, as those kinds of things can often shroud underlying problems.