re: #8 dangerman

if he doesnt explain the path, then he is just doing a trump -
saying lots of stuff that has no meaning or bearing on the discussion

This dKos comment thread explains it fairly well, I think.

I am well aware of Lichtman and his approach. He is not alone in attempting to have a model that allows one to make a determination solely on the basis of “data” and “facts” independent of what is happening on the ground. His model has a high degree of correlation with previous elections, but first, correlation does not necessarily mean causation, and second, the country has been increasingly bitterly divided in a way we have not seen I would argue in more than a century.

…as well as:

One of his “keys” is assuming Johnson will get over 5% of the vote. If you dont believe that his model would predict a Clinton victory.