Thursday, September 27, 2012

Latest NBC/Marist crosstabs indicate NC could go either way in November, depending on the electorate

 NBC/Marist's poll of likely voters in North Carolina should give the Romney campaign some heartburn - Obama currently leads 48-46% in a state he only carried by 0.3 pts in 2008. The former loyally Republican presidential state had shown signs of reverting back to it's old self in polling earlier this year, but has taken a turn for the worse since the DNC. The poll was taken September 23-25th.

If you believe the new NBC/Marist poll crosstabs, then North Carolina could be anywhere from a moderate Obama victory to a moderate Romney victory, depending on your favorite turnout model. Since past exit polls are all we have to go by, we'll use their partisan identification numbers to do some re-weighting (just for fun, of course ;).

The Marist NC poll's party identification for the likely voter sample is 39% Democrat, 31% Republican, and 30% Independent (D+8). That's a net 3 point lower turnout for Democrats in the Tar Heel state than turned out in 2008 (when partisan identification was measured at 42D/31R/27I; D+11). What would North Carolina look like this November, using Marist's numbers, and 2008's electorate?

So....ummm, yeah. Pray for low Dem turnout, right? Obama is essentially leading by 5 pts, 50-45%, matching the Marist numbers up against a 2008 electorate. That is definitely not a path to 270 for Romney. BUT, suppose turnout defies current poll results, and looks more like the NC electorate from the 2004 election (I would use the 2010 NC Senate exit poll party ID numbers, but neither CBS nor CNN have them available. If anyone knows them, please fill me in). That year, as Bush carried the state 56-43%, partisan identification broke down like this: 39D/40R/21I (R+1).

Crazy, eh? When re-weighting to the electorates of the last two presidential elections, North Carolina swings as wildly as 50-45% for Obama (based on Marist poll crosstabs), or 51-45% for Romney. Honestly, NC needs to fall closer to the 51-45%-for-Romney end of the spectrum if he expects to win nationally. Obama barely carried NC in a year that he won nationally by over 7 points. If he is anywhere near leading the state on election day, he has most likely been reelected.

And again, just for fun, what would NC look like, using the Marist poll crosstabs, if the 2012 electorate was a cross between 2008 and 2004?
 When you average 2004 and 2008 NC turnouts, Romney manages a 48-47% lead. I think most could agree that that's not going to cut it for a national win.

1 comment:

  1. How do you personally look for info for your new articles and which exact search algorithms or techniques do you mostly turn to?


Note: Only a member of this blog may post a comment.