Wow. Let's think about this: 232 years since the Declaration of Independance, 143 years after the Civil War, 80 years after Women's sufferage, and 40 years after the Civil Rights movement, and Ambinder is writing "The Gap -- And There Really Is Only One Gap... It's white males."

Really? Now that's a shocker.

Now sure, white males have so rarely had a voice in Presidential Politics. I mean, it's entirely possible that one of our own may not be president this time next year! I feel so discriminated against. Thank god we'll still have an important say in this upcoming election, regardless of who wins. I just hope I get some ads directed at me, maybe some good old fashioned white male pandering. I hope McCain shows up at a White Male advocacy conference. I hear there is one later this August in Minneapolis.

I'm so glad that in a time of recession that Mark Ambinder continues to draw a salary. Maybe next week he'll write about how the candidate who's supporters are alive will have the best chance of securing their vote.

McCain is winning white voters overall by 7 points, not white men.

Presumably, I'm guessing he's up by about 12 amongst white men and 1-3 points amongst white women.

Actually, McCain doesn't have a male problem.  Steve Lombardo writing over at Pollster.com says

Given his struggles to woo older voters away from Hillary Clinton, it is somewhat surprising that Obama is in a statistical dead heat with McCain among voters 65 and older (he actually leads among those ages 55 and older). With Obama continuing to carry all voters under 35 by the wide margin that propelled him to his primary victory, it's natural to wonder where McCain's support comes from.

The answer is middle-aged and older men. The only age/gender categories where McCain leads? Men aged 35-54 (McCain + 10), 55-64 (McCain +7) and 65 and older (McCain +17). Of course in the past these cohorts have been the most likely to make it to the polls on Election Day.

Let's not forget that come election day, a lot of non-blacks (male and especially female) just aren't going to show for Obama. It won't be reflected in any poll, because it's become commonplace for racists to lie when they're asked. They've had forty years to practice lying and they're good at it. It's the new variable and there's almost no way to measure it.

That's why it's become conventional wisdom that Obama needs more than his (almost static) ~4% margin going into election day. It's probably correct. Less than that and any last minute onslaught of negative advertising from Republicans could swiftly lose the election for Obama. In 2004 Kerry should have been further ahead in the polls before he was swiftboated. If he was, he could have won come election day. Same will likely hold true this year. Obama needs to be more than 4% ahead in October or he will lose in the closing week. It happened in 2000. It happened in 2004. It will happen again.

In 2006 Harold Ford and a couple of other blacks running in races where the majority of the voters were white only preformed 1 or 2 point below where the last polls before the election had generally had them. In contrast, Bradley finished close to ten points below where his final polls indicated. The Bradley effect isn't what it used to be for a four point Obama lead in the final polls should now lead to a two or three point win instead of about a five point loss.