Oh My GodRasmussen released his September poll of partisan ID today, and it is a bombshell. To refresh your memory, this poll is conducted monthly and asks 15,000 people what party they are affiliated with. In my Rasmussen Party ID model, I use a 3 month average of these numbers.
In August he released this poll and it showed a Republican advantage of 4 points. This poll has never showed a Republican advantage of more than a single point, and usually shows a significant Democrat advantage, since it is self identification and doesn't track likely voters.
Most of us considered the August result to be an outlier, because it was such a huge increase and goes against all history, 2010 included. I personally expected August to be influenced by the Republican convention, and that in September the Democrats would return to parity due to a bounce from their convention.
That isn't what happened. The September number confirms the Republican edge with a 2.6 point advantage. Very unexpected and very significant.
With the new September number, the 3 month average changes:
Old - New
33.8 33.8 Democrat
36.0 36.4 Republican
30.3 29.8 Independent
I can't over emphasize this. Earlier today I was arguing with CAC, who thinks a D+3 turnout is the most likely result. I've been arguing that an Even turnout is much more likely. But this poll says that R+2 or better is the most likely result.
With this new average, Romney is leading in the RCP average by 3.9%!