Transcription – Ken Mehlman Interview

Q: Mm-hmm. Was there a better strategy for Senator Kerry, in that election?

MEHLMAN: It’s so easy to look back and play Monday morning quarterback. I felt that Senator Kerry and his team ran a very strong campaign and a [45:00] very strong effort and a smart effort. So, you know, it’s easy to look back and say, “I would have done this differently,” or, “that differently,” but anybody can do th That’s amateur hour. (laughs)

Q: What di ? I mean, you’ve seen other reelect campaigns.

MEHLMAN: Yeah.

Q: And usually, a president gets reelected and his party loses seats in Congress.

MEHLMAN: Yeah.

Q: So could you talk about why that didn’t happen in 2004?

MEHLMAN: Well, I think there are a couple different factors. One is I do think the country, again — the increased polarization is one explanation for that. Second of all, [43:00] if you stop and you think about it, at least some… You’re electing House. You’re electing Senate. You’re electing the presidency. So the Senate election in 2004 to think about, is most relevant is 1998. That was the election before 2004. That was the six-year senator was up who had been elected in ’98. That was a good year for Democrats. So if six years before had been a good year for one party, that party has to worry, because six years later their tide which had come out to help them usually sweeps in a little bit. So that was one challenge, I think. So you’ve got, first, increased polarization, second, that unique attribute. In the House, you had the fact that there had been redistricting in ’02 that was generally pretty good for Republicans. That helped Republicans. That helps explain it. And then finally, I do think it matters that we tried very hard to avoid a lonely victory, to run a campaign that was very much focused on collaborating with Republicans around the country. And I think those factors all matter. [44:00]

Q: Did you have a sense or did the president have a sense that there’s a trade-off there, tha you might be able to win a 55% victory by running a purely pre and traditional —

MEHLMAN: Yeah.

Q: — selfish campaign but…?

MEHLMAN: Yeah, we thought about that. And we thought that it was more important to try to win an election in which we could get things done by electing allies than it was to try to say, “OK, let’s run up the numbers a little bit more.”
Election Day. The exit polls, which I think historically had been pretty reliable —

MEHLMAN: Yeah.

Q: — by about five o’clock —

MEHLMAN: Yeah.

Q: — are showing you guys losing —

MEHLMAN: Yeah. Yeah!

Q: — and losing pretty big.

MEHLMAN: Yeah.

Q: So what do you do with that information, when it comes in?

MEHLMAN: Well, we had developed our own internal tracking system. [45:00] We had identified key counties around the country and key areas around the country that we thought were a bellwether — they might have been bellwether, because they were Republican strongholds, where we wanted strong turnout, or, alternatively, they were swing areas. And we were tracking our own turnout and our own performance. And our data looked quite good. So when we first saw that, we thought, “This could be a repeat of what happened four years ago,” where the exit polls were inaccurate. And we, in fact had pulled the exit polls from four years ago to share with the press. And so that information, as we compiled the two together — you know, within about three or four hours, we realized that was the case. But for three hours, two hours, we didn’t know for sure that was the case. And I remember having conversations with both 41 and 43, as well as with Vice President Cheney, where I walked through, for example, why I thought the fact that there had been huge turnout north of the I 4 corridor all around Florida gave me confidence we would win Florida, whereas the exit polls had us losing Florida, [46:00] and where I looked at Lancaster County and York County, Pennsylvania, two red counties — Lancaster has become a bit bluer recently but, at the time, two red counties where there had been incredibly robust turnout. And so we looked at Ohio, where, in a lot of the collar counties, outside of the three big cities, again, very strong turnout — in the — and not just collar but the exurban and rural counties. And all those things said to me that we might lose but we would lose a very close election. And the exit polls showed something very different. So we became convinced that essentially the statistical model upon which the exit polls were premised had been flawed.

Q: Were you worried, as a campaign, that, if that became the story, by five o’clock Eastern Time, two o’clock —

MEHLMAN: Sure.

Q: — Pacific Time, if that became story, “It looks like a big Kerry win coming”?

MEHLMAN: We were but we conditioned people on our team and we conditioned the public to understand what had happened four years ago, the mistakes that had happened with the exit polls. And we thought that was important. I remember going on Fox News, at about 8:30 at night, nine o’clock at night. It was 8:30, I think. And I just walked through very specific data about these Florida counties — occasionally still get people coming up saying, “I remember election night in 2004, when you went on Fox News.” And so the difference was… They had a statistical model that had proved flawed before. So I came back with my own model, that was real. And I said, “Here’s our data. Here’s what it shows. Here’s what we see in these counties. Here’s the initial results in these counties. To me, the initial data indicates our model is a bit more accurate than the model in the exit polls.” And it turned out to be true.