I remember leading up to Election Day his predictions were giving her 66% and Trump 33% (IIRC). Which means if you roll a die and it comes up as 1 or 2 then Tru

I remember leading up to Election Day his predictions were giving her 66% and Trump 33% (IIRC). Which means if you roll a die and it comes up as 1 or 2 then Trump wins (which isn't all that unlikely) Many people mistook these numbers to mean Clinton was going to get 66% of the vote and thus were expecting a landslide, but that is not what his numbers meant. Now I wasn't reading any of the analysis he or his team was putting out, but the modeling seemed reasonable to me.