On the heels of a new poll from Gravis showing Tillis up 5 points, and reports generally of a surge in Tillis’s support, Tom Jensen of PPP is out with a survey showing Hagan still up 3, meaning the race is more or less the same as it was a month ago, and that the incumbent is still the favorite.

My take? I don’t buy the Gravis poll – or the PPP poll. I think the real state of the race is probably Hagan up by maybe a point or so, reflecting the RCP average, which shows her up by 1.2%. And there has been a momentum shift toward Tillis in the last two weeks. That cannot be denied. The question is whether or not Tillis’s “comeback” will be enough for him to win. Be highly skeptical of individual polls. Hopefully we’ll get a lot more of them this week. I think it will confirm that if Hagan leads, it’s only by a very small amount.

Some more info to keep in mind: PPP’s final poll in 2012 had North Carolina tied (Romney won it by 2). Gravis had Romney winning by 4. I would trust PPP over Gravis (I’m disappointed that the latter didn’t include Haugh), but I still think this poll is too friendly to Hagan and overstates her lead. This is a true toss-up contest, maybe tilting toward Hagan, but some Republicans think Tillis could be up by a point or two. We’ll know a lot more next week.

OK, now that we’ve got that out of the way, let’s look at the poll and see if we can detect any interesting trends.

Senate race
46% Hagan (+2)
43% Tillis (+3)
5% Haugh (no change)

There are 6% undecided. Most people in the know are saying most everyone has made up their minds and it’s going to be all about ground game. Assuming PPP is correct in where the candidates stand, Tillis needs more than 2/3rds of undecideds to win. That’s doable, but it’s not where a challenger wants to be at this point in time.

What would be really helpful to Tillis is if Haugh lost some of his support. Or would it? PPP finds Haugh taking equally from both Tillis and Hagan. This is in contrast to the SurveyUSA/Time Warner Cable poll, which finds Haugh taking disproportionately from Tillis. Without Haugh, SurveyUSA says Tillis is up by 1 point.

Interestingly, it’s now October and Haugh still maintains 5% of the vote. That could be due to his sole debate appearance, or due to the high unpopularity of both candidates. Either way, it looks like Haugh will take more of the vote than Libertarians are used to taking in North Carolina. If Haugh keeps 5% on Election Day, Tillis will probably lose.

Obama Approval
42% Approve (no change)
52% Disapprove (-1)

Obama approval too high. He’s more like a -15.

Favorable/Unfavorable
First number: net favorability. Number in parentheses is change in support from last month.)
Hagan 41/50 (-9) -4
Tillis 37/49 (-12) +2
Haugh 10/17 (-7) +2

Hagan is at 50% job disapproval, not a good place for an incumbent to be. But she’s winning because Tillis is still disliked by the electorate, and only slightly less so than last month. Only 27% of voters know who Sean Haugh is and can give an opinion.

McCrory Approval
40% Approve (-1)
43% Disapprove (-3)

A lot of political-minded people in NC keep writing about McCrory like he’s some kind of pariah who is certain to lose in 2016. Kind of like how a lot of people keep saying Hagan is the clear favorite when she’s clearly not. They’re living in September, and the people who maintain that McCrory is wickedly unpopular are living in September 2013, when he actually was a pariah. To be sure, McCrory will face a tough reelection and he’s hardly the popular figure he was two years ago, but Roy Cooper is not going to waltz into the governor’s mansion two years from now either.

Burr Approval
31% Approve (+1)
31% Disapprove (-1)

Nothing to see here.

General Assembly Approval
25% Approve (+7)
52% Disapprove (-6)

Now this is interesting. The General Assembly has seen a significant rise in approval. My guess is that Republican partisans are feeling a little more loyalty to the institution now that Democrats are running hard against it. Still unpopular and still a millstone for Tillis, but interesting nonetheless. I also wonder, now that the General Assembly has become such an issue in the Senate campaign, if these numbers are more ‘real’ than in the past, when a lot of voters would simply rate the General Assembly poorly simply because it is a legislative body, and legislative bodies are not popular these days.

Legislative Parties Favorability
NC Democrats 38/48 (-10) +3
NC Republicans 40/45 (-5) +13

Well, this is interesting. Both NC legislative Democrats and NC legislative Republicans are unpopular, but NC Democrats are twice as unpopular as NC Republicans. I really can’t explain this. The +13 swing in Republican favorability is probably due to the “rally around Tillis” effect. As for the NC Democrats, usually a party in the minority and against a majority that’s very unpopular experiences an upswing in their numbers, but not here. What’s going on here? I don’t know. But it certainly doesn’t bode well for Democrats taking back the majority or making big gains any time soon.

Legislative Generic Ballot
46% Republican (+2)
44% Democrat (+1)

Interestingly, Republicans lead on the generic ballot. My guess is that this is due to two things: (1) the legislature is not nearly as unpopular as people have been led to believe (still unpopular, of course, but not toxic) and (2) ‘generic Republican’ might poll better than individual legislators, something we’ve seen in the Civitas polls. That Republicans shouldn’t be too comfortable with their generic ballot lead, as the results on Election Day might defy this statistic. It does mean, though, that being a generic Republican, even after the last two sessions of the legislature, is still not a bad thing in North Carolina. At the very least, it’s still better than being a generic Democrat.

More on the Senate Race
Hagan is up 12 with women. Tillis is up 7 with men. Women are 2/3rds of undecideds. That’s good news for Hagan.

Tillis is winning only 10% of Democrats. Hagan is taking 8% of Republicans. My guess is that these ‘Republicans’ are big RINOs, but usually there aren’t that many Republicans who cross over to vote Democrat. Tillis also needs to win around 15% of Democrats to be competitive, an indication that conservative Democrats who voted for Romney are unusually lukewarm about Tillis. (Tillis is concentrating heavily on Eastern North Carolina in the final weeks of the campaign.)

Tillis is +21 with whites. Romney was +36. I imagine whites are more favorable to Tillis than the poll indicates, but he still has work to do with white voters. Hagan was always going to win more whites than Obama, fortunately for Tillis the electorate is more white in midterms, so he can afford some loss of ground with these voters. But not that much.

By region – Hagan dominates in the 252 area code (Northeastern NC) and the Triangle. Tillis wins everywhere else. He’s ahead in Greensboro, Charlotte, the Mountains, and Southeastern NC. High margin of error on the regional breakdowns.

The bottom line.
I think Tillis is in a more favorable position than this poll indicates. Don’t trust any individual polls, just put them in the average and you’ll get the result that’s closest to reality. I think Hagan is probably up by a point, maybe less. I really do think in the end this will come down to ground game. If you think that’s where Democrats have an advantage, advantage Hagan. Besides that, the only question is what happens with Haugh supporters. Do they stick with Haugh? If not, do they go disproportionately for Tillis? Or not?

Maybe, in their elections reform bill, the Republicans should have required a runoff in the case no one gets a majority in the first round, like Georgia and Louisiana. And come on, who wouldn’t want to have another month of Hagan and Tillis going at each other?

By the way, if Hagan wins, it won’t be on Election Day. It’ll be during the one stop early voting period. She’ll need a surge in turnout from the young, of African Americans, of women – those who supported Barack Obama but didn’t turn out in 2010. The early voting period starts this Thursday.