I looked at the internals to see the D/R splits, and they did seem to be overweighted. Even so, they only moved the needle in Florida and Colorado averages, which is still a big deal.
In my model, it dropped Trump's probability from 44% to 38%.
-PJ
123 posted on 09/29/2016 8:06:07 AM PDT by Political Junkie Too
(If you are the Posterity of We the People, then you are a Natural Born Citizen.)