Reply #310: There's a big difference between +3 and +9

[joubert]

Even if a marginally greater percentage of Republicans voted on Election Day rather than voting early, it doesn't explain the magnitude of the change in the presidential race. Your theory also doesn't explain why the governor's race results didn't change. Whatever you may say about Easley's crossover appeal, his success in any given N.C. county was strongly correlated to Kerry's. The counties that went for Kerry by a small margin went for Easley by a much GREATER margin. Therefore, Easley should be up in early voting returns above his final percentage if your theory is correct.

There are explanations for why the Easley results are similar. The early voting and Election Day samples might be similar, but there might have been a shift away from Kerry in the final days. However, the exit polls of late-deciding voters suggest that the OPPOSITE of that happened - instead, late-deciding voters shifted toward Kerry. Or there could be some systematic difference in behavior between early and late voters that isn't linked to party registration. That's possible, but no one has offered any affirmative evidence to prove it.

I have no idea why you are so obsessed with this issue. No one is freaking out here - people are patiently examining some pretty interesting data that is far more solid than the vast majority of the "fraud" claims I've seen. I've seen some of your posts on other topics that are perfectly reasonable. If this issue doesn't interest you, why not just ignore it?