Normal Signals, One Big Cluster ....?

[Wonkish1]

Is anybody else having the feeling that in this election the normal signals we use to predict these things are way more chaotic than normal?

So much of it is contradictory. The supposed more R leaning polls occasionally throw up huge swings to the Obama, the more D leanings polls occasionally throw up huge swings to Romney. The party ID numbers are all over the board like we haven't seen before. Polling has gotten so volatile that one journalist for a major media company even wrote an article titled, "The day polling died".

A state that has new registration numbers looking good for one party has the other party looking increasingly better in the polls. The early voting is running also running contrary to poll momentum in certain states.

In most instances all of these things tend to move in tandem. Both 2004 and 2008 were much more stable and predictable in terms of early signals. Today it's increasingly getting more and more frustrating to try to analyze these things.

I'm sure I'm not alone on this and I'm sure it's shared on both sides.