Oct. 25: Romney Ripple Recedes To Margin Of Error, White Share Mediocre Mid-Fifties—With One Starting Exception

I've decided not to wait for this evening's WaPo/ ABC tracking poll, but will file again if it deigns to mention the racial breakdown, so here goes:


The post-debate Romney ripple can still faintly be discerned today, but (with one startling exception) the presidential race had slipped back into margin of error territory. Romney's white share (with that one exception) continues to be blah.


The MSM rarely pays attention to the white vote, but when it does it has a funny habit of exaggeration. Thus The New Republic just posted a confused article Romney Has Historic Lead Among White Voters, by Nate Cohn (October 23). Actually, as we have repeatedly shown, Romney doesn't have a historic lead: it is ignominously stuck in the mediocre mid-50s range achieved by post-Reagan GOP candidates, far below Reagan's 64% in 1984. (More comparisons here—scroll down). 


TNR's headline, like that on Aaron Blake's Oct. 18 WaPo article, seems to pander to some atavistic fear among liberal readers, and editors.


The most recent crop of polls, beginning with the startling exception:


  • Something called QStarNews has just (Oct. 25) put out a national poll showing Romney ahead 53.99-45.12.

Romney's white share: 62.01, for a 25-point lead over Obama.


I have no idea how accurate this poll is (discussion here). But it's internally consistent—if Romney were to be that far ahead, this is what his white share would have to look like.


Otherwise, the race is floundering along much as before.


  • IBD/TIPP, recently the outlier in predicting a statistically significant Obama lead, regressed even further to the mean today (Oct. 25), showing Romney trailing Obama just 44.7-46.7.

Romney's white share: still just 53%, as it has been for some time.


  • PPP last night (Oct. 24) showed Romney trailing Obama 48-49. Romney's white share: 58%, for a nineteen-point lead over Obama.

  • AP-GfK this morning shows Romney ahead of Obama 47-45, within the margin of error.

Typical of MSM preoccupations, this poll has attracted attention because AP reports the gender gap has been eliminated. (Both ways—Romney has supposedly lost his edge among men, a very unusual result). Weirdly, this datum does not appear in AP-GfK's detailed cross-tabs. Nor, of course, does the racial breakdown.


  • Reuters/ Ipsos (Oct. 25) has just shown Romney ahead of Obama 47-46.

Reuters/ Ipsos does not release racial breakdowns, but on Oct. 7, when Romney trailed 45-47, his share of white Registered Voters—as opposed to Likely Voters, which all polls now focus on—was reportedly a very low 50%.


  • Gallup (Oct 25) shows Romney's lead holding since yesterday at three points, 50-47, within the margin of error.

Once again, Gallup did not publish its racial breakdown. On Oct. 16, it showed Romney getting a record 61% white share, which at that time translated into a four-point national lead over Obama.


  • Rasmussen Reports (Oct. 25), shows Romney now ahead of Obama 50-47, with Obama up a point since yesterday.

Rasmussen Report's Premium Platinum subscribers show that Romney's white share remains only 57%, for an eighteen-point lead over Obama, who gained a point among whites yesterday.


Two state races polled by Rasmussen today continue the story of Romney's relative weakness aong whites in the industrial North.


In Pennsylvania, Romney trails 46-51. Rasmussen's Platinum subscribers learn that Romney's white share among likely voters is just 55%, for a thirteen-point lead over Obama. Not enough.


In contrast, in Virginia, Romney is now ahead 50-48—and he's getting 64% of the white vote.