You missed North Dakota, though it doesn't really matter since your whole concept here is flawed.
All predictive models have flaws. I claim no talent as a prophet.
(I knew that I was off by one state in my count -- and an easy State to ignore. Thank you for that correction).
...This model might show who is having problems (or more problems) at the moment -- when there is a poll. It is no better than the poll, of course.
You cannot deny that Pat Toomey is in far worse shape than Chuck Schumer -- right?
Bold analysis.
It's very bold actually. Schumer is not inevitable, Anything Can Happen In Politics™, and the election is still 21 months away!
