This is a rush transcript from "On the Record ," October 30, 2008. This copy may not be in its final form and may be updated.

GRETA VAN SUSTEREN, FOX NEWS HOST: First: It is close, even closer than it was last night. It is now a 3-point race for the White House. A new FOX News poll shows the race tightening, with Senator Obama at 47 percent and Senator McCain at 44 percent.

Earlier today, we spoke with Karl Rove.


VAN SUSTEREN: Karl, I've seen the new FOX polls, and it looks pretty close between Senator Obama and Senator McCain. Then I look at your map and it doesn't look close. So can you explain, you know, which one -- you know, why the disparity?

KARL ROVE, FORMER GEORGE W. BUSH ADVISER, FOX CONTRIBUTOR: Well, the map is drawn on state-level polls over the last two weeks, so it is slow here in the final weeks of the campaign to show dramatic movement. In most instances, you have only a couple of polls in the last couple of weeks. And so rather than just showing the most recent poll, we've taken the last two weeks' worth of polls. So it will fail to show movement.

Now, there are a lot of polls at the state level, but there are a relatively small number per individual state. For example, between Sunday night and Wednesday night, there were 66 individual state polls, but that's divided over, you know, some 20-some-odd battleground states.

VAN SUSTEREN: So if you're Senator McCain looking at your map, you're thinking what?

ROVE: Well, first of all, I hope Senator McCain is looking at individual state-level data that his own campaign is generating on a real- time basis. You know, look, we've got all these public polls, and there are a plethora of them, but a campaign spends a lot of time deciding on who its pollster is and then very carefully monitoring their work. So my hope is that he's looking at, and I'm sure Senator Obama is looking at the individual polls done by their campaign's pollsters in those battleground states.

But look, the national numbers are showing a movement towards a tighter race. I suspect we're going to see here by the end of the week that some of these individual state numbers are going to show it, as well.

VAN SUSTEREN: If you were Senator Obama looking at your state polls or your state map, then, are you taking a deep breath, or are you nervous because of that margin of error?

ROVE: Well, he ought to be nervous because of the margin of error, and he also ought to be nervous because, clearly, the national race is tightening. And also over the course of, you know, January through June, Senator Obama continually had surprises in the primaries. He was expected to win New Hampshire handily and instead was nipped at the closing line by Senator Clinton.

He's not a good closer. I mean, there is a resistance to him, a doubt about whether or not he's experienced and qualified enough to be president that causes people at the tail end of the race not to end up in his column, and I suspect we're seeing that again. Despite all the money he's spending, despite the, you know, six television networks showing his half- an-hour infomercial last night, despite the fact that he's outspending McCain by a margin of, you know, $100 million on television ads in the critical battleground states, this race has gotten tighter, not wider. And it's because of this persistent doubt that people have about Senator Obama.

VAN SUSTEREN: Karl, when you worked on the two President Bush campaigns and you looked at your internal campaign polling data, was it markedly different than what we were seeing in sort of, like, the Gallup and the FOX and all the other polls?

ROVE: Well, first of all, remember, we're in a much different year this year. There have been, through yesterday, through Wednesday, 728 national surveys. By the end of Thursday, by the end of today, there will be in excess of 735 national polls that match the candidates head to head. By comparison, in all of 2004, there were 239, and there were just about half that number in 2000 -- in fact, less than half that number in 2000.

So we've seen a veritable explosion of the polls -- of the polling. This month alone, there have been 222. As of Wednesday, there were 215. As of Thursday, 222 national polls. There were only 239 national polls in all of 2004. So our races in '00 and '04 were a little bit different because there was some public data, but we were looking at our own data and then occasionally seeing other public polls that we could use to triangulate.

But look, you can make yourself crazy. I mean, think about this. On Monday, we had, I think, like, 9 or 10 polls. One showed the race as close as 3, and one poll showed the race as close -- as far away as 10 points difference between the two candidates. On Tuesday, the margin went from 2 points to 15 points. On Wednesday, it was 3 to 7. Now, we are seeing a tightening over those days among most polls, but nonetheless, there's a wide variation among these polls.

We shouldn't endow them. We give them such scientific precision that they simply do not have. We've got the law of probability, the quality of the pollster, the quality of the sample size, any adjustments that they make internally to it, their methodology. All these things come into play in making these things less precise and less reliable than we like to think they are.

VAN SUSTEREN: So you think most of these -- you think these two presidential campaigns -- and maybe relying on your own experience -- do you interview your pollsters, your internal pollsters ahead of time? Do you have more than one? Do you sort of cross-play one against the other in terms of the numbers? How is it done behind the scenes?

ROVE: Yes. Well, I can't speak to the experience of both of these campaigns. My sense is they've each got one major or principal pollster.

We used several different pollsters, one principal pollster and then some additional people who might have had special expertise in some states. But look, you spend a lot of time coming to understand your pollster and his or her record, understand their methodology, understand how they structure their sample, how they structure their questionnaire. You measure them and test them against other available numbers. But you eventually have to have confidence in your pollster as somebody who's doing the right thing and is speaking candidly about the results, not telling you what you want to hear.

VAN SUSTEREN: In terms of the numbers, we've seen that the numbers are probably not as good for Governor Palin as she would like. Yet she draws giant crowds. And my inbox and my Gretawire blog it was, like, incredible enthusiasm for her. So it's, like, you know, I don't know what to measure it by.

ROVE: Well, look, again, this is one of the things that comes into play in a political campaign. We're talking about the vice presidential running mate, which is rarely dispositive in people's vote decisions. Generally, no matter -- even her numbers, which are upside-down -- that is to say, more people have got a negative feeling about her than a positive - - well, the people who've got a negative feeling about her are basically people who aren't voting for McCain and never had a desire to vote for McCain.

She clearly has had a very positive impact on energizing the grass- roots activists within the Republican Party, who McCain needs to have making the phone calls, knocking on the doors, putting the stickers on cars, you know, revving up the local community, distributing the literature, manning the headquarters. All of those things, clearly, she's had a big impact on.

Mike Duncan, the chairman of the Republican National Committee, told me that the day after she was -- the day she was announced, the Republican National Committee got 40 percent more money on line than it got the day before. I've talked to a dozen state party chairmen around the country who told me that the week before she was named as a running mate, their headquarters were getting the volunteers, but it was sort of tough. The week after she was named, the headquarters were fully manned and they had to be looking for cell phones and additional ways to make use of the volunteers they had. So I think she's been a plus for the ticket. I know there's been criticism of her, and the media has been relentlessly negative about her, but I think she's been a big plus for the ticket.

And look, Biden -- there's the example. I mean, his favorables are better than his negatives, but who thinks that he's played a significant role in influencing either the direction of the race or the enthusiasm of the Democrat activists? I mean, he's been a non-player. It's a sign that the people end up voting for the president, and the vice president only matters to the degree that they help fill a political need for the presidential candidate, or more importantly, that they do something in -- for the act of governing afterwards.

VAN SUSTEREN: Well, it's interesting with Governor Palin. Whether you agree with her or not, you have to admit that she's very intriguing and has sort of electrified a lot of people, even the media outlets that don't particularly like her.

ROVE: That's right.

VAN SUSTEREN: So it seems to me that her value is there's so much media attention on her that she can get her message out, which -- free, inexpensively on the networks that carry her news. And so they don't have to do this huge ad buy to sort of compensate for the tremendous difference in fund-raising.

ROVE: Well, it's always better to have a constant flow of television ads. You know, I do think television ads play less of a role in a presidential campaign than they do in most any other kind of a race, but they still have a pretty big impact, particularly when you're able, in a state like a Florida or an Indiana or a Montana or a Colorado or a Nevada to be outspending your opponent 3 or 4 to 1, as Barack Obama was able to do for most of the fall.

Now, here at the end, they're shouting with roughly equal volume. But for weeks in September and early October, Barack Obama in these battleground states was having a much bigger impact through the television ads. And remember, the object with those television ads is not to sway 20 percent of the vote, it's to move 5 percent or 10 percent of the voters because in these battleground states, that means a lot. In Florida, George W. Bush won in 2004 by 5 points. If you can move 6 or 7 points towards the Obama column by spending $39 million on campaign activity and advertisements in Florida, as the Obama campaign is doing, that wins you Florida and maybe wins you the election as a result.

So I don't want to diminish the importance of television ads. I don't want to overstate it, either. But in this kind of a campaign, the kind of throw-weight advantage that Barack Obama has presented a real challenge to McCain, that even having an outstanding running mate like Sarah Palin and getting out a message and getting on the news can't fully overcome.


Watch Greta's interview with Karl Rove: Pt. 1 | Pt. 2

VAN SUSTEREN: Coming up, much more of Karl Rove. Does Karl think Senator Obama's infomercial was a smart move? Was it a good value for the enormous price tag? Plus, the behind-the-scenes story tonight, the stuff we all want to know. What are the last few days and hours of a presidential campaign like on the inside? Now, Karl's been there twice.

And later: Is someone getting really dirty, snooping around and trying to dig up dirt on Joe the plumber? What's with that? We're going to tell you what we know tonight, and it sure sounds suspicious.


VAN SUSTEREN: More of our interview with Karl Rove.


VAN SUSTEREN: Any idea -- what would be your best estimate how much that 30 minutes cost last night for the Obama campaign?

ROVE: Well, I wouldn't be surprised if it didn't cost him, you know, $6 million to $8 million to get on those seven networks. I wouldn't be shocked at all.

VAN SUSTEREN: And now that we've had the benefit of 24 hours of it and 20/20 hindsight, if -- $6 million to $8 million -- if you were running a campaign, is -- you know, is that something that you thought was a smart or shrewd move? I guess we have to wait until Tuesday to know for sure, but useful?

ROVE: Well, look, if you have that kind of money and can spend that kind of money, it is useful. Think about this. Let's say it was $8 million. They got roughly 26 million people on the tube. That means they spent what, 30 cents per person to watch that program? That's a pretty good expenditure of money. You know, if you wanted to mail 26 million people something, a piece of mail, it would cost you probably, all in, slightly more than that. And instead, they got a 30-minute television infomercial that was very carefully focused-grouped and tested.

There's no doubt in my mind. I mean, you looked at virtually every word that Barack Obama said, and you could see that that was something that they had carefully tested and honed. I mean, things like, you know, We're going to stop tax breaks for companies moving jobs overseas. Well, you know, he's never spelled out in the campaign exactly what he means by that.

We saw, I'm going to give a tax cut to 95 percent of Americans, when in reality, half the people who are going to be getting a check from the federal government are people who pay no income tax. They're not getting a tax cut, they don't owe any taxes. What they're getting is, is -- is a check annually from the government under the Obama plan. I mean, it's very carefully tested and focus-grouped.

And you know, it's -- I'm going to ask people to pay what they were paying in the 1990s. No, you're not going to be asking people, you're going to be telling people to fork over 40 percent of what they earn to the federal government. Let's be honest about it.

VAN SUSTEREN: In terms of between now and Tuesday, people who work on the campaign at a high level -- any sleep?

ROVE: Well, hopefully, you get a couple hours of sleep because if you -- you know, five days is a long time, and you do need to get six or eight hours of sleep a night in order to make certain that rest of the day, when you're working 100 miles an hour, that you're able to make good decisions and you're able to do the duty that you've been assigned.

But I've got to tell you, you're running on sheer emotion and adrenaline. I don't care if you think you're miles ahead or miles behind, you're straining every nerve and your -- every bit of your body is working as hard as it can in order to get to the goal.

And so much of the campaign comes here to this moment, I mean, so much of the ground game, so much of the messaging, so much of the candidate's travel, so much of the -- you know, all the efforts of the campaign sort of get concentrated in this 72 to 96 hours before the election, and it's a pretty amazing thing to see it come together.

So I hope they do get some sleep. They may not be good meals, but get a little bit of food and stay away from the alcohol and stay away from the cigarettes. And don't drink too much coffee. Drives the volunteers nuts.

VAN SUSTEREN: (INAUDIBLE) great advice. All right, now, 2000, we all know what happened with the recount. But where were you that night, at least the election night, as everyone was sort of waiting what we thought would be a regular night? But where were you?

ROVE: I was in the headquarters in Austin. I had an office that had glass walls on two sides of it, so my office had a lot of people coming in and out, but also a lot of people walking by. And I was there with some of the other number-crunchers, trying to make heads or tails of the evening. And I was literally there all night.

I remember when the networks called it for Governor Bush, President- elect Bush, early in the morning, I sort of got propelled by the crowd out of the office and up Congress Avenue in Austin. It was raining. It was cold. It was dismal. But the whole time, I had a sick feeling in the pit of my stomach because -- I remember turning to Matthew Dowd and saying, What do they know that we don't know? Because we found the numbers still close to call Florida when Florida was awarded to us.

VAN SUSTEREN: So I guess...

ROVE: It was one wild night.

VAN SUSTEREN: I imagine it was. And there will be a book, right?

ROVE: Oh, yes. In fact, I'm working on the book now. And in fact, I've been writing on 2004, which was also another interesting evening because of the exit polls. As you recall, the exit polls leaked out on to the Internet in the middle of the afternoon. And after a huge problem with the exit polls in 2000, the networks spent tens of millions of dollars to get it right and didn't. And in 2004, in fact, they were even worse.

And they colored everybody's attitudes not only until the actual vote totals began coming in, but in some instances, all the way through to the next morning. Some of the networks would not call Iowa or Nevada for Bush, even though -- in the case of Iowa, for example, Bush had a 13,000- vote lead, and there were literally two precincts in two small counties that were overwhelmingly Republican counties, two precincts that were not counted, and some of the networks still refused to call Iowa for Bush. And why? Because the exit polls had convinced them that the race was too close to call and they couldn't trust that there wouldn't be somehow mysteriously out of these two small precincts in two rural counties enough votes for John Kerry to swamp Bush in Iowa.

VAN SUSTEREN: All right. Well, this is going to be exciting this Tuesday. And I look forward to the book. I always love the behind-the- scenes books because we sure see it a lot different way than from the outside. So Karl, thank you. And I'll see you Tuesday, no doubt.

ROVE: You bet.

VAN SUSTEREN: Thank you.

ROVE: See you in New York.


Content and Programming Copyright 2008 FOX News Network, LLC. ALL RIGHTS RESERVED. Transcription Copyright 2008 ASC LLC (www.ascllc.net), which takes sole responsibility for the accuracy of the transcription. ALL RIGHTS RESERVED. No license is granted to the user of this material except for the user's personal or internal use and, in such case, only one copy may be printed, nor shall user use any material for commercial purposes or in any fashion that may infringe upon FOX News Network, LLC'S and ASC LLC's copyrights or other proprietary rights or interests in the material. This is not a legal transcript for purposes of litigation.