Articles and Analysis


24% Leftovers

Topics: 2008 , The 2008 Race

Clinton pollster Mark Penn claims, at a reporter breakfast, "I think you’re going to see as much as 24% of Republican women defect and make a major difference nationwide in terms of, I think, the emotional element of potentially having the first woman nominee."**

Obama pollster Joel Benenson (once a principal in Penn's firm) quickly responds with a memo calling Penn's claim "entirely baseless and refuted by a number of public polls."

The Washington Post's Jon Cohen provides some context, including the finding that a 24 percent showing among Republican women "would significantly outperform any Democratic candidate since 1972."

Mark Penn then blogs back: An unspecified internal poll shows Clinton's support increasing "to 13 percent" among Republican women, while undecided "surged to 11 percent, so a total of 24 percent would either vote for her or consider voting for her."

Ben Smith, who followed the story all day, concludes: "if we can get Joel Benenson and [Edwards pollster] Harrison Hickman blogs, this could get fun." Amen brother. Or maybe just occasional "Guest Pollster" gigs at a certain poll obsessed web site?

First Read reports a memo by Biden campaign manager Luis Navarro that cites recent polling to claim his candidate is sneaking up on Bill Richardson.

Marc Ambinder shares a memo from McCain advisor Rick Davis that "pours water on Giuliani's electability argument…or tries to anyway."

And in other news…

Frank Newport reviews "where things stand" in campaign 2008.

Gary Langer looks at the President's ratings and tells us "where he's at."

Kathy Frankovic considers Al Gore and the way perceptions can change as former presidents and vice presidents shift from political figures to "elder statesment" and back again.

Mark Mellman shares results showing Americans "overwhelmingly oppose key elements of the Bush administration's surveillance agenda."

David Hill argues that, despite "growing evidence that the Democratic Party will expand its majorities in both the House and Senate," voters will resist one-party control by the Democrats.

Evan Tracey reports that two thirds of the television ads of the Democratic presidential candidates have focused on health care and Iraq.

As for the boy from Cleveland, to quote Ebby Calvin LaLoosh in Bull Durham, "Sometimes you win, sometimes you lose, sometimes it rains."

**The quotation comes from Ben Smith's blog and now reflects his "slightly corrected" version that is "slightly more hedged." The original version read: "You're going to see 24 percent of Republican women voters defect because of the emotional element of having a woman nominee." Thanks to reader rilkefan for the edit.



Mark Penn needs to stop spouting off imo. The only thing saying you expect women to vote for Hillary because shes a woman is going to do is turn off women voters. Thats the kind of thing you want to let happen without suggesting that you expect it to. I think there is a chance that some republican women will vote for Hillary, but only if you work for their votes and not take them for granted.



Nobody mentioned the inverse possibility that some centrist democratic voters may vote republican because they have a negative emotional response to a woman candidate.


Gary Kilbride:

Mark Penn has it wrong. There are no last minute surprises in presidential politics. The dynamics of the race are evident early and history indicates the poll leader several months in advance is most likely to prevail. It's similar nonsense to what we heard throughout 2004, the claims that John Kerry was a closer.

Everything tends to drift back to the beginning. You'll never go wrong embracing that truism. But it's classic faulty handicapping to assume just the opposite, that the aberrational drift will continue. That's what Penn is doing, noticing the very real movement toward Hillary among right leaning women and magically concluding the tilt will become more substantial, to the point it shocks us in November 2008.

Allow me to wager otherwise. It reminds me of every year in the NFL. Teams that improve their standing by 3 wins or more from one year to the next, or flop by 3 games are more, are overwhelmingly more likely the following season to drift toward their previous level. I could bore with plenty of specific numbers from year to year, but let's just say it's more than 75% likelihood. Of course, that doesn't stop Sports Illustrated from being remarkably ignorant, projecting the lame '06 Dolphins to the Super Bowl based on an irrelevant and misleading improvement from 4 wins in '04 to 9 in '05. In fact, it's natural tendency for fans in general to project movement in the same direction and not a reversal.

I realize that's a bizarre detour. But it reminds me of Hillary's situation. Even if she gains among independent and right leaning voters throughout the next year, the base reality has huge input in the end. Rubber band time, naturally yanking back some of her gains. You can't reject that she is a polarizing figure, and in a country that may not yet be eager to elect a female. Political conventional wisdom doesn't show up on its own, without merit.

If Hillary wins a general election it will be tight, and based on a generic advantage that she does not entirely forfeit.

The Clintons always run in favorable situations. You have to give them major kudos for that. Brilliant handicapping, which I always appreciate. Bill ran in '92 when the nation was weary of 3 straight GOP terms, and now Hillary properly avoids a certain loss to an war time incumbent in '04 to run in a terrific anti-Iraq war, pro-Democratic climate in '08.



I'm surprised that nobody has noted the following things:

1. The more important issue may be whether Senator Clinton can outperform past democratic candidates among independent and democratic women.

2. Data to examine the relative gender advantage of Clinton, Obama, and Edwards exist in the detailed cross-tabs of the SurveyUSA head-to-head match-ups from September through October polls. When not doing my day job, I've run the gender numbers for Guiliani versus Clinton/Edwards/Obama for about 8 states (not yet getting to California and Florida, which show some very interesting gender patterns), and without getting into the details, they show Clinton at about a +13 advantage among women; Edwards +11; and Obama +8. [These are averages for a number of SUSA state samples --not a national sample.]

Keep in mind that Gore had a +11 gender advantage in the 2000 race; Kerry much less so in 2004.

3. As I said, however, the gender numbers for Florida and California in the SUSA head-to-heads show a much more dramatic gender advantage for Clinton. There are also some other important states where this is true (New Mexico).


The link to the reporter breakfast report doesn't [any longer] say what you say Penn said.


Alan Abramowitz:

To put this in some perspective, we can look at the results of two key U.S. Senate races in 2006 that featured female candidates on the Democratic side: Claire McCaskill in Missouri and Amy Klobuchar in Minnesota. Both of these candidates won their races, Klobuchar by a landslide margin. According to the exit polls in these states, McCaskill received 8.5% of the vote from Republican women in Missouri while Klobuchar received 13.9% of the vote from Republican women in Minnesota. Based on these results, it is highly unlikely that Hillary Clinton, who is clearly a much more divisive figure than either McCaskill or Klobuchar, would receive anything approaching 24% of the vote from Republican women in a presidential election.


Post a comment

Please be patient while your comment posts - sometimes it takes a minute or two. To check your comment, please wait 60 seconds and click your browser's refresh button. Note that comments with three or more hyperlinks will be held for approval.