Brownstein: Right. What we did was we looked at the exit polls for the past five presidential elections, working with Ruy Teixeira -- he's a public opinion analyst at the Brookings Institution -- and what we did was... The exit polls, Tammy, are so large, they interview so many more people than a conventional poll, they are able to look at much more precisely targeted groups in the electorate. And basically what we did is, we looked -- we took a more finely grained look at the electorate, I think, than ever before, and what we've done is look at how white men and white women, married white men, single white men, married white women, single white women, depending on their college education, Hispanic Catholics, Hispanic Protestants, men, women, college, non-college -- we basically x-rayed the electorate to an unprecedented degree of depth and looked at the real patterns of, which are the swing voters and which are the stable voters locked in one party or the other? It really does give you a fabulous baseline, I think, for understanding the coalitions that McCain and Obama are trying to build this year.

Q: And what did you find?

Brownstein: Well, I think, you know, that there are many interesting findings along many interesting directions, but to me, I think there are a few big ones that jumped out. First of all, if you look at white voters -- we'll start with white voters -- the movement over the last 20 years has been almost entirely among women. It's extraordinary when you look at the Democratic vote among white men. These last five elections, as you know, Michael Dukakis, Bill Clinton, Bill Clinton, Al Gore, John Kerry -- very different candidates, very different circumstances, very different times -- in all of that period, the Democratic vote among white men has varied between 36 percent of the vote and 38 percent of the vote. That's the total variation. They've never reached 40 percent of the vote in these last five elections with either white men without a college education -- the kind of blue-collar whites -- or white men with a college education.

The movement has been much greater among white women, and there is often a substantial divergence between the blue-collar, non-college white women, what we've called waitress moms in the past, and those more upscale college-educated white women. And in fact, the Democrats in '04 -- big divergence. Bush dominated among those downscale white women -- that was one of the reasons why he won. The college-educated white women are much more in play. This year Obama is improving on both fronts. He is gaining ground on both fronts and even some ground, in some of the polls, to a lesser extent, among the college-educated men. The blue-collar men are still really tough for him.

Q: Before you go on, let's take it to Hillary Clinton. So is this part of the reason why Obama didn't pick Hillary Clinton, because he didn't really need her -- he was getting those women anyway?

Brownstein: Well, he wasn't getting them at that point.

Q: Oh, I see.

Brownstein: And I think -- you know, I think on balance he'd be doing better if he did pick her. I mean, because I do think that if you look at the structure of the white vote, men are very tough for -- white men are very tough for Democrats in general, and Obama culturally, ideologically -- perhaps race as well, all of those reasons -- he faces a lot of barriers with those voters. He has much better potential to expand his vote among white women. And, you know, if you look at where he has gained ground relative to Kerry, it is I think more among the non-college white women right now than the college white women -- who split about evenly in '04 -- are leaning slightly toward Obama this time.

But as I said, those blue-collar women broke overwhelmingly for Bush on cultural and security issues in '04. McCain by picking Palin was obviously aiming at those same women. She has -- I mean, clearly she is a candidate who is targeted much more toward downscale white women than upscale white women who probably don't -- many of them don't like her views on abortion or creationism and that sort of thing. Yet because of the economic distress is such that it's really helping Obama gain ground among those blue-collar white women, those non-college white women, and that's allowing -- I think that is the key to him being competitive in places like Ohio, Indiana, even West Virginia, which are, you know, very much of a kind of kitchen-table states without a lot of upscale voters.

Q: Wow. And what about the men?

Brownstein: Well, with the men I think we are seeing some improvement among the better educated men for Obama relative to Kerry, but not so much, at least in the polling that I have seen, among the non-college, blue-collar men -- Joe the Plumber and his, you know, counterparts -- Obama was still getting only about 35 percent. Kerry won only 35 percent of white men without a college education.

You know, the larger point here, Tammy, is that we are living through a class inversion in terms of the coalition to the parties. If you go back historically, Adlai Stevenson, John Kennedy, Lyndon Johnson, Jimmy Carter, Hubert Humphrey, all of those Democratic nominees in the '50s, '60s, and '70s ran substantially better among blue-collar whites, those without a college education, than they did among the more upscale white-collar white voters with a college education.

In 2000 that line crossed. Al Gore ran better among white voters with a college education than those without it. In 2004 Kerry ran six points better among white voters with a college education than those without it. Obama may improve among both groups, but I am betting that class inversion persists and we continue to see a Democratic coalition in which all of the rhetoric is, you know, we're the party of the little guy, we're the party of the working guy, but increasingly they're depending upon these upper-middle-class professionals, not the truly rich -- over $200,000 a year, still pretty Republican -- but we're talking about the lawyers, the professors, the architects, these families who live in Fairfax County, Virginia, or Mecklenburg County, North Carolina, or Oakland County, Michigan -- white-collar suburbs, that historically had been Republican. They are all moving toward the Democrats. And if Obama wins, that will be, I think the absolute foundation of his victory.

Q: So what can McCain do to win, based on this information?

Brownstein: Well, I think for McCain... And I would add one other thing that's important here, which is that -- one last point about another group, Hispanics -- Hispanics vote the way white voters did a generation ago. There is no class inversion. Democrats do better among the less educated, less affluent Hispanic voters. They do better among Catholics than Protestants, by the way; there is a big gap between Hispanic Catholics and Protestants that people don't focus on. President Bush actually won Protestant Hispanics -- many of whom were evangelical -- in 2004.

Well, if you look at this I think it's pretty clear that the best hope for McCain is to continue the Republican dominance among these blue-collar white voters, especially men. I mean, I think that Obama is a very strong competitor for these better educated white voters, especially independents, who, by the way, again, when you look at white independents: enormous class inversion. Last five elections, Democrats, as we've found --and this is the beauty of this kind of analysis, you can look at this in a way we never had before -- in the last five elections Democrats have run better among college-educated white independents than non-college white independents. So it's basically the Bill Bradley independents, not the Lou Dobbs independents, who Democrats rely on.

So if you look at all of this, it basically says to you that McCain has to recapture -- I think his best chance is to try to regain ground among those blue-collar voters, who tend to be culturally conservative, hawkish on national security, but are feeling economically strained and have moved, at least tilted, especially the women, toward Obama since the financial crisis. I think it is conceivable that Obama will actually win among white voters without a college education, and I cannot imagine that any Democrat has done that since Lyndon Johnson.

I think McCain's best chance are those working white guys. I mean, Joe the Plumber is his target audience, but unfortunately for him Joe the Plumber isn't enough. He probably has to convert Joe the Plumber's wife, who is a big -- as I said, 60 percent of those blue-collar women voted for Bush in '04. It was a big reason why he won, certainly the reason why he won a state like Ohio, which actually decided the race. McCain's best hope is to go after those voters. And probably he's doing it in the best way.

Q: With Sarah Palin. I mean, based on what you're saying, that was a really good pick, because Joe the Plumber likes her. He thinks she's hot.

Brownstein: Right. It is a good but limiting pick, because it basically accepts the kind of drift away from the Republican Party among the upscale, more socially liberal voters, right? I mean, Sarah Palin is someone who helps where she helps but kind of draws a circle, you know? She circumscribes the appeal of the campaign. I think she is a very tough sell in places like Montgomery County, Pennsylvania, or Mecklenburg County, North Carolina, or Fairfax County, Virginia, where you have a lot of these socially moderate to liberal upscale voters, these college-educated white women who split evenly and are a swing vote -- split evenly in '04. I think what she does is she does give him another weapon, but right now it's not working. As I said, he is losing ground among those blue-collar white women who tend to be economically strained. But that's where he has to go.

And you know what else? I mean, you know, if you look at kind of the macro here, what McCain is retreating to is kind of a red-state fortress strategy. They pretty much, you know, pulled their ads out of Wisconsin. The only Kerry states that they're focusing on are Pennsylvania and New Hampshire; they're behind by double digits in both of them. And again, those are states with large numbers of these upper-middle-class white professionals who've been trending Democratic, the suburbs of Pennsylvania -- suburbs of Philadelphia -- the counties near the Massachusetts border of New Hampshire. Those are very tough states for McCain.

So what does that leave him with? If Obama wins all the Electoral College votes that Kerry won -- 252 -- and you add to that Iowa and New Hampshire where both sides give him the lead, that puts him at 264. And what you're seeing from McCain now is, they're basically saying we have to draw the line there. We have to prevent Obama from winning any out of the remaining states that he's contesting, which are Colorado, Ohio, Virginia, Indiana, North Carolina, Florida, and Nevada and Missouri. If Obama wins any of those plus what we just discussed, he wins.

So what was McCain's message in that debate? It was much more of a base-oriented message, not only in terms of small government, low taxes, less spending, but also as you saw, right, I think we were all surprised by how hard he hit the abortion card. It fits in with his kind of Electoral College situation now and his demographic situation. I think you were seeing a kind of circle-the-wagons, last stand at the Alamo, can we hold these red states by pulling back culturally conservative voters, particularly the downscale culturally conservative voters, and you know, kind of peter out Obama at 264. That's where we draw the line. We hold everything else. But it obviously leaves him with literally no margin for error. Any one of those other states -- Florida, Ohio, Colorado, Virginia -- and this is over, unless McCain can win a Pennsylvania or New Hampshire or Minnesota, all of which now seem unlikely.

Q: So if you're in the McCain campaign and you read this story already through plus all the polling that you have, what will they do in the next two weeks? Because people tend to think in Washington that they're now going to go ultra-negative. Do you think based on these numbers and where they need to go -- this is purely a numbers question -- that their strategy will change? Will they become much tougher, or will they pull back, try to appeal to women and look less angry?

Brownstein: I tend to think the latter. I mean, I think where he ended up in the debate is the best place he can be for the final stage of the campaign. If you recall, as you do, in 2000 on Labor Day George Bush introduced a new stump speech, the core of which was for the final two months of the campaign, I trust the people, he trusts the government. That was essentially -- talking about Al Gore -- that was essentially where McCain ended up at the final debate. I trust the people, he wants -- he's Senator Government as, you know, that kind of Freudian slip in the debate -- and that basically you make an ideological argument here.

The problem is that, you know, when people are feeling economically strained, especially these blue-collar voters, you know, the safety net doesn't look that scary to them right now. Taxes do. And certainly the culturally conservative argument is one that, you know, can have an appeal with those blue-collar women, talking about abortion again.

Like the pick of Sarah Palin, I thought the fact that McCain emphasized his opposition to Roe v. Wade so unabashedly at the debate was a kind of tacit admission that upscale is not going to be great, that, you know, Obama is going to run better among the socially liberal white-collar whites than we have seen other Democratic nominees do. He may not win them, but he could run about even, and that the growth potential in these final weeks is probably more downscale -- you know, Joe the Plumber, and Mrs. Joe the Plumber even more importantly, because John McCain's already getting a lot of Joe the Plumber votes. You know, 65 percent of white men without a college education voted for George W. Bush. Democrats have never reached 40 percent. They only average 36 percent among those blue-collar men in the last five elections.

Q: That's remarkable.

Brownstein: So you know, McCain is going to do very well there, but as Clinton showed, you can win, you know, with a relatively low number among those men; when the blue-collar white women also go the other way, though, when they go toward the Republicans, that's when the math gets much tougher for Democrats. That was always the group -- if you look at all the stuff that's been done this year -- Stan Greenberg's focus groups in Macomb County, my own experience talking to voters -- there is a big difference. The blue-collar women are much open to Obama. The men are focused on national security; they doubt that he has the experience to be commander in chief. The women are different. They are open. They have always been open to him primarily around economic distress. They probably would have been even more open to Hillary Clinton, but that's another story. And that, I think -- if McCain can't turn that around, he can't win.

I mean, I think that the other three -- he might also be able to reverse some of Obama's gains among upscale men, who are very tax-sensitive, these college-educated men. Many of them own their own business. They are very focused on taxes. That may be another group. So that's sort of like the... You know, if you kind of look at it, what you've got here is college-educated -- if you think about the white vote, and I think that we have to think about the white vote because the black vote is overwhelmingly for Obama, he's going to get more than Kerry did, he's going to go into the mid '90s, the Hispanic vote is going to be significantly better for him than it was for Kerry, particularly with this downscale strength that he has and in fact McCain has lost ground, I should point out, in the story, even among the better educated Hispanics where Bush ran competitively -- so what you're left with is really the white electorate.

And if you look... A final thought: If you look at the white electorate through these four quadrants, you know, women are more Democratic than men, upscale whites are now becoming more Democratic than downscale whites. OK, so if you kind of think of yourself as a grid, you've got, the place where the two Democratic advantages intersect are with the college-educated women, and that's going to be a very good group for Obama. The place where the two Republican advantages intersect are with the non-college white men, and that's going to be a very good group for McCain, and then you're left with the swing groups, right? I mean you have the non-college women who are pulled in different directions, and right now they're breaking for Obama -- tilting toward Obama after breaking decisively for Bush. And then you have the college-educated white men, which are very close right now after voting pretty decisively for Bush. So those two groups are probably the ones that we have to watch the closest in the last two weeks, because they are the most likely to move.

Q: I strongly suggest everyone listening to go to NationalJournal.com right now and download Ron's cover story, "The Hidden History of the American Electorate." Thank you, Ron.

Brownstein: Thank you, Tammy.