It is a bit of pablum we all got drilled into our heads from our parents as a kid, but it is also quite appropriate for this particular polling wrap: if it sounds too good to be true, it probably is.
For the first time in a long time, the bigger presidential polling story might be on the general election front, rather than in the Republican primary. Lots of political heads were turned over the weekend with the NBC/Marist polls, which claimed a fairly wide lead for Barack Obama over his leading GOP challengers in Florida and even a small lead over the GOP hopefuls in South Carolina.
South freaking Carolina. As in, hasn't gone Democratic since neighbor Jimmy Carter won it in 1976. As in, lost by Barack Obama in 2008 by a nine-point margin.
That sounds really good, if you are a Democrat. Too good to be true, even.
We'll explore the likelihood of that right after the jump.
But, first, the data, starting with the new numbers for the Republican primary:
NATIONAL (Gallup Tracking): Gingrich 33, Romney 23, Paul 8, Bachmann 6, Perry 6, Santorum 3, Huntsman 2
FLORIDA (NBC/Marist): Gingrich 44, Romney 29, Paul 8, Perry 4, Bachmann 3, Huntsman 3, Santorum 2
IOWA (ARG): Gingrich 22, Paul 17, Romney 17, Perry 13, Bachmann 7, Santorum 7, Huntsman 5
IOWA (Univ. of Iowa): Gingrich 30, Romney 20, Paul 11, Bachmann 9, Perry 8, Santorum 5, Cain 4, Huntsman 2
SOUTH CAROLINA (NBC/Marist): Gingrich 42, Romney 23, Paul 9, Bachmann 7, Perry 7, Huntsman 3, Santorum 2
And, now, for those rather eye-popping general election nums:
NATIONAL (Rasmussen): Romney d. Obama (45-42)
FLORIDA (NBC/Marist): Obama d. Romney (48-41); Obama d. Gingrich (51-39); Obama d. Paul (49-36)
SOUTH CAROLINA (NBC/Marist): Obama d. Romney (45-42); Obama d. Gingrich (46-42); Obama d. Paul (47-37)
Of course, the question most likely to be asked based on these poll numbers is: can we now consider South Carolina to be a swing state, given this most recent offering from Marist, or is this poll simply too optimistic an assessment of the president's prospects in the Palmetto State? A dispassionate analysis of the numbers tends to support the latter, though in the long-term it is not difficult to make a claim that this state will trend Democratic over time.
2008 represented a bit of a high water mark at the presidential level for Democrats. Barack Obama's vote share (44.9%) was the best performance for a Democrat since incumbent president Jimmy Carter came within an eyelash of holding the state in 1980. And even though Democrats did suffer a disappointing loss in 2010 with the defeat over veteran Rep. John Spratt here, they also had a surprisingly strong statewide effort from Democrat Vincent Sheheen, who came within a few points of the upset against Republican Nikki Haley.
Furthermore, South Carolina is still growing, and a lot of that population growth over the last decade has come from state-to-state migration. Given the states that have high out-migration, it is not unreasonable to conclude that South Carolina's new residents are coming from states that have political traditions more amenable to the Democrats.
Having said all that, however, there is ample reason to believe that the poll by Marist showing the president leading the trio of leading GOP contenders is more than a tad optimistic.
First of all, there is the matter of contrary data. Tom Jensen of PPP, for example, was quick to point out that their August poll in South Carolina had Mitt Romney leading Barack Obama handily (53-38). Even NBC/Marist's earlier poll in the race (in October) had a markedly different picture of the race, showing Romney ahead 46-40. National numbers have gotten a bit worse for Romney since October, but nine points worse?
Second, it looks like both Marist's Florida and South Carolina polls have samples that run counter to some recent evidence. For example, South Carolina's survey has a sample where Democrats outnumber Republicans by a couple of points. That is several points different from the 2008 exit polls, which showed a 41-38 advantage for the GOP. And it is miles apart from the 2010 exit polls, where the GOP advantage was eleven percent (44-33).
Now, a reasonable explanation for that disparity is that the Marist poll surveys registered voters, and not likely voters. Fair enough, except that the Marist poll in Florida had a sample that was 44 percent Democratic and 33 percent Republican, which is quite different from the most recent voter registration stats in Florida, which had a considerably smaller Democratic edge (41-36).
Of course, there could still be a way that this poll is a legitimate reflection of the state of the electorate in these two pivotal southern states. When many on the left (including, admittedly, myself) wished to discredit ugly data in the 2010 cycle, we'd point to how different the samples were from the most recent exit polls. What we underestimated, until late in the game, was how big the enthusiasm gap for Democrats had become that year. 2010 was not 2008, nor was it all that close to 2006, either.
And, perhaps 2012 is not like its predecessors, either. Perhaps Democratic fear of a "brave Newt world" is getting the left amped for 2012. Or perhaps the GOP is not enthused about next year, rending their garments as they look at their presidential prospects, wondering when it all went so terribly wrong. That would also explain the outsized voter ID numbers, when compared to previous years.
If that's the case, though, there will be more evidence of it in future data. Which is something to keep an eye on as we close the books on 2011 and head into 2012.