So, naturally, my predictions are often hopelessly wrong.  Still, I'm interested in what you're doing here.

But what are you doing?  How are you assembling these numbers?  What is your process?

I've been looking at pollster.com, and the numbers there show that in national popularity during 2008, Obama has consistently been ahead of McCain, while it's only been during the last few days that Clinton has edged out McCain.

If we combine those results with what you're showing, that seems to indicate that McCain could win the Electoral College but lose the popular vote.

In which case this time my protests would be something less than peaceful.

But I really don't want to go to jail next year unless it's absolutely necessary.  Are there any possible weaknesses in your method which might indicate that such a dreaded result will not occur?