The new PPP poll of Connecticut was conducted September 24-26, 2012, and finds Romney matching his worst performance against Obama out of the 4 polls PPP has taken in CT since the Fall of 2011. Obama leads by 13 points, 54-41%, which is only half of the 22.4 point margin he carried the state by in 2008 (61-38%). A glance at PPP's internals reveals that Connecticut likely voters are self-identifying in a very similar fashion to 2008; party ID for the poll is 43% Democrat, 26% Republican, 32% Independent, where as party ID in CT in 2008 was 43D/27R/31I. So it's pretty safe to assume that if we were to re-weight this CT poll to 2008 numbers, the results would be almost identical, given the similarity of the partisan identity. But what would happen to PPPs CT numbers if we were to apply a 2010 turnout model (which was more GOP friendly, and was 40D/28R/33I)?
As you can see, Obama's 13 point lead is reduced to a 10 point lead (53-43%). And that's under a very GOP friendly turnout scenario. But what happens if we apply an even more GOP friendly turnout scenario (such as the 37D/30R/33I electorate CT experienced in the 2004 election)?
Sheesh. Lets be honest folks. We're not going to get 2004 turnout in November, barring some major Obama blunder in the next 5 weeks. So much for this poll from last September. But even under the best circumstances, Obama still manages 51% in the state, to Romney's 44%. So, at least according to PPPs numbers, ignore those polls that make you wonder if Connecticut is competitive this fall.