Q: Take us down to these last days, and what should people be looking out for?

Brownstein: Well, first of all, you know, we are already in the middle of the election. It is an incredible total -- Michael McDonald of George Mason University does the best job of keeping track of this -- on his Web site this morning we are past 20 million people who have already voted in this election, and of course that's just his keeping track of the stats; it is probably higher than that.

Look, at this point in a campaign you do not have to read the polls to know where the trends are. You really just have to look at the calendar and the schedule of the candidates. The fact that John McCain is spending so much time in Pennsylvania, a state where he has been trailing by double digits in almost all polls in the month of October, pretty much tells you what you need to know -- is that he has a very tough pathway to 270, the 270 Electoral College votes he needs to win.

Barack Obama is barraging and besieging an array of red states, including some like Virginia, North Carolina, Indiana that have voted for a Democrat at most once or none -- no times since 1964. That pretty much tells you the story now. Obama is pushing at the doorway in a lot of different red states. He has a lot of different pathways to 270. McCain pretty much is acknowledging that he is going to lose some of those Bush red states from '04 and the only way he can get there is by taking Pennsylvania -- a state where he has been trailing substantially and which Democrats have won four consecutive times.

Q: Undecided voters -- there is a new AP/Yahoo News poll out today that says that one in seven are undecided. Is that true?

Brownstein: Well, it's probably true that there are one in seven who are willing to say they are not entirely committed to one or the other. But to be truly undecided -- no, I think that is much too high. I think it is probably in most polling about half that or even less when you consider that some of the people who are saying they are undecided today are undecided between voting or not, in effect. I mean, look, we have been living through this campaign for what, almost two years -- unprecedented viewership of the debates, of the conventions, of the general election debates, of Obama's infomercial, advertising at unprecedented levels, incredible amounts of voter contact -- some people who are still undecided by now simply may not make their way into the ballot box.

So you know, I don't think it is that large. I mean, I think obviously there is some play at the end, and we are always surprised, Tammy, in the exit poll when -- how large a percentage of people say they made up their mind in the last three days. But I don't believe that one-seventh of the electorate is truly undecided. That includes, I think, people who pretty much know where they're ending up but are not yet willing to make that unequivocal commitment.

Q: Let's go back to the white male voters. Non-college whites are still the largest voting bloc, right? You have written about this.

Brownstein: Yes they are. And in fact... I mean, look, if you kind of look across the board, one way to kind of -- there are a lot of different ways you can cut the electorate, but one simple way to kind of look at it, if we think about it, four, I guess five major groups -- African-Americans, Hispanics, college-educated whites, non-college whites and everybody else, Asians, etc. -- well, at least on four of those groups the likelihood is that Obama is going to improve on [John] Kerry's performance in 2004.

Kerry won about 88 percent of black voters, about what Democrats usually win; Obama could be in the mid-90's, and the turnout could be up as a share of the overall electorate. Kerry won, depending on how we look at it, somewhere around 60 or a little bit over 60 percent of the vote among Hispanics; Obama seems on track to get between 70 and 75 percent. Forty four percent of whites with a college education or more, who have been trending Democratic -- these upscale socially liberal white-collar whites, as we have talked about -- Kerry won 44 percent of them; it seems entirely likely from the polling that Obama is going to come in higher than that. He might even in fact win them, which is something that no Democratic presidential nominee has ever done in the history of modern polling going back to 1952, not even Lyndon Johnson.

So the one piece that is out there that we don't know is what is going to happen with these blue-collar white voters. Joe the Plumber and Mrs. Joe, really, are John McCain's kind of last hope in this election, because given the likelihood that Obama is going to improve among all these other segments, the only way McCain can make the math add up is if he drives down Obama's vote, I think, even further among those non-college whites. 2004, those whites without a college education voted 61-38 for Bush. They were predominantly for Bush in 2000 again.

The economic crisis has given an opportunity for Obama to sort of improve on that, but culturally, tax issues, national security -- those are all issues on which these voters, especially the men, tend to favor the Republicans. So that I think is where McCain's last hope is the last question for Obama, kind of the last piece in the puzzle that he is trying to put together, and it has been, as you know, the problem for him from the outset. In the Democratic primary those non-college white voters voted 2 to 1 cumulatively overall for Hillary Clinton.

Q: So you are saying that the McCain campaign, which has had a barrage of criticism at a level I have never seen, actually is doing the right thing, because they have Joe Plumber out there, they have got the Joe the Plumber tour, and your numbers seem to say they are going in the right direction.

Brownstein: Well, I think, you know, they made a kind of fateful choice when they picked Sarah Palin. When they picked Sarah Palin, who is, you know, very culturally conservative, opposes abortion even in cases of rape and incest, has praised the teaching of creationism -- they basically were, I think, saying they did not expect to significantly move those socially moderate to liberal upper, better-educated voters, especially the college-educated white women, who are the most Democratic of white voters.

And I think at that point they kind of put their chips on mobilizing the same kind of downscale coalition that Bush did. Blue-collar men have been very resistant to Obama all the way through, and I think, you know, will be on Election Day. John Kerry only won 35 percent of white men without a college education. Joe the Plumber is a Republican, and many, you know, white working-class men like him are Republicans.

The moveable group here are the non-college white women, and they have been more of a -- their preferences have bounced around more over the years and even in this campaign. And one reason why Obama is in such a strong position is because he is running generally -- for example, in our Diageo/Hotline poll -- he is running ahead with those women, and they voted overwhelmingly for Bush in 2004; Bush won 59 percent of those non-college white women. Obama is now leading among them, so I do not know how much more advantage McCain can squeeze out of the non-college white men, Joe the Plumber himself. I mean, as I said, only 35 percent of them voted for Kerry in '04; it is not clear how much lower it can go.

The real issue is kind of the Mrs. Joe, and whether the tax arguments and some of the cultural arguments and Palin's kind of cultural connection can help him there. Right now the answer is no, and that is probably the biggest thing that's changed, I think, from '04 to '08. To some extent upscale men have changed, Hispanics and blacks have changed, but these non-college white women are really key to Obama's strong position right now.

Q: But I also do not think it is hard to figure out that Joe the Plumber's wife does what he says.

Brownstein: Well, there you go. Actually, you know, they vote -- you know, there is a difference. As I say -- I mean, you go back and you look at these last few elections, Democrats -- since 1988, Tammy, no Democratic nominee has exceeded 38 percent of the vote among the non-college men, among the Joe the Plumber-type voters. Among the non-college women they have gotten as high as 48 percent for [Bill] Clinton in '96 and 45 percent for [Al] Gore in 2000, so they kind of go their own way.

If that vote collapsed -- I mean, remember we are talking about the waitress moms, security moms in '04, it's now coming back -- it is entirely possible that Obama will win those voters, who are, by the way, of course the foundation of the Hillary Clinton coalition in the primary; I mean, she got about 67 percent of the vote among those non-college women. And if Obama can perform credibly among those women, continue his strength with the upscale women, the college-educated women, and he is doing better than Democrats have done somewhat among the college-educated men; you add to that Hispanics and blacks, there just isn't enough place for McCain to go.

And that is why I think those kind of working-class waitress mom, white women are really critical if McCain has any hope of even significantly narrowing this gap in the end. And there is some evidence that the heavy emphasis on the tax argument, the size-of-government argument is helping McCain more than the earlier attempt to kind of paint Obama as kind of un-American or unpatriotic, which I think most voters found kind of over the top. You know, he is sort of consolidating those right-of-center, culturally conservative independents and so forth. Whether that is enough to get him over the top is another question, but it is kind of putting him in at least a more competitive position that might help some Republicans downballot at least a little.

Q: Is there a traditional tightening of the polls for each election? I would like, for your final questions, if you can just tell our listeners the things to look out for election night, and I want to begin with the tightening of the polls, which we are hearing everyone say.

Brownstein: Well, you know, I mean we've been living in a 50/50 country, and in particular Democrats have very rarely had blow-out elections. Only one Democratic presidential candidate since World War II has gotten past 50.1 percent of the vote -- Lyndon Johnson. Now Barack Obama has a real chance to do that, which would be remarkable -- the first African-American candidate becomes the first Democrat other than Johnson since World War II to exceed 50.1. But the idea, if a Democrat -- for him to get up to 53 or 54 percent, which would be required for a seven or eight-point win that we are seeing in some of these polls, would be remarkable by historic standards. I'm not saying he won't do it, but just think about the magnitude of the achievement. John Kennedy couldn't do it, Bill Clinton couldn't do it. No Democrat has been able to do it since World War II.

Now if you look at the remaining undecided, they are overwhelmingly white, three-quarters white, and the white voters are more non-college than college, and that would seem to imply that there would be a tightening of the race, especially in some of these red states. So if you are a white, older or non-college voter and you have not committed to Obama yet, it is hard to imagine that most of them are going to end up doing so in the last 96 hours.

The countervailing force here is that I tend to think that the number of new voters, young people and African-Americans and Hispanics, is going to be at the high end of everybody's estimate. Whatever the polls are estimating, I think it is going to be, you know, a point or two higher than that, and Obama is winning those new voters overwhelmingly. That will push the result back in the opposite direction.

In terms of what to watch, I think it is pretty simple. If Obama wins Virginia or North Carolina and Pennsylvania, the election is over. If he wins Virginia, North Carolina, or Florida, actually, and Pennsylvania, the election at that point is pretty much over. It is almost impossible mathematically for McCain.

Q: So, I believe -- I am not looking at -- Virginia's polls close at seven. So we will know by seven?

Brownstein: Well, I think -- we won't know by seven, but I think we will know once we have Virginia and Pennsylvania. I mean, I think unless McCain can take back... if McCain cannot win Pennsylvania he in essence cannot lose any one of the red states that they're still contesting, because Kerry... If Obama holds all the states that Kerry won -- that is 252 Electoral College votes -- Iowa and New Mexico, which Bush won, are pretty much conceded to Obama; that's 264, and that means any one out of all the others we are talking about, Tammy, any one out of Virginia, Colorado, Florida, Ohio, Missouri, Indiana, North Carolina -- any one of those makes him president. So that's why taking away Pennsylvania is the only way that McCain gets any breathing room. So if Obama wins Virginia in the seven o'clock hour, takes Pennsylvania, I believe in the eight o'clock hour, we are functionally done.

Q: So you think that if there is no -- if the networks, if everyone doesn't announce at seven o'clock when the polls close in Virginia -- in other words if it is tighter than we think, and all the votes are in.

Brownstein: Right.

Q: That that's an indication that Obama is not doing as well as these polls say.

Brownstein: Right. Absolutely right. Absolutely right.

Q: OK, so look for that. All right, my friend. Ron Brownstein, you have been incredible all the way through. We will talk to you when we have a new president.

Brownstein: All right. Absolutely right.

Q: Thanks, Ron.

Brownstein: Thank you, Tammy.