More Ruction In Iowa And New Hampshire

Iowa Democrats. In the latest average, Barack Obama is now leading Hillary Clinton 27 to 26 percent, with a competitive 23 percent for John Edwards. The Obama lead is accounted for by a late-coming Strategic Vision poll showing Obama leading Clinton 32 to 25 percent, with 25 percent for Edwards. Strategic Vision is a Republican firm, and I'm not putting much weight behind its results. If you believe it, it's great news for Obama; for me, it just confirms that this is a close three-way race. And Iowa polls, as I've written before, are problematic, since you're trying to isolate the 5 percent of the total electorate that will turn out on caucus night--at a time when some 2008 caucusgoers are not aware that they're going to participate, while others who are sure that they will participate will turn out not to do so. I agree with the speculation that the Clinton campaign has decided to attack Obama on the grounds that even if it hurts Clinton it will hurt Obama also and in effect will help Edwards, whose chances in southern-phobic New Hampshire aren't good and who doesn't have the money (he has gone on public financing) to compete successfully up through and including the big-state February 5 primaries.

Iowa Republicans. The average has Mike Huckabee leading Mitt Romney 27 to 25 percent, with 12 for Rudy Giuliani and 11 for Fred Thompson. As I've pointed out in my U.S. News column this week, Huckabee gets most of his support in Iowa from evangelical Christians, and they account for a much larger share of Iowa caucusgoers (about 38 percent) than New Hampshire primary voters (about 18 percent). Huckabee, even if he wins in Iowa, has to expand his base to be close to winning in New Hampshire. Obviously, Giuliani and John McCain would like to see a Huckabee win in Iowa, which they hope will deflate Romney's current lead in New Hampshire (below) to their benefit.

New Hampshire Republicans. Even so, Romney's lead in New Hampshire has, so far, pretty well held up. He's averaging a lead of 34 percent to 18 percent over Giuliani and 16 for McCain. New Hampshire historically has had an aversion to southerners, and the figures for them are discouraging: 10 percent for Huckabee, 6 for Ron Paul, 3 for Thompson. (In retrospect, it wasn't very wise for Thompson to skip the New Hampshire debate to announce on Jay Leno.) Romney's position here is the only serious lead anyone has in the Iowa caucus or New Hampshire primary polling. But one can't regard it as etched in stone. Republican voters, to judge from polling, are less firmly committed than Democrats, who aren't all that firmly committed themselves; and one can imagine these numbers shifting significantly should Romney lose in Iowa. A window of opportunity, maybe, for McCain and, maybe, for Giuliani.

New Hampshire Democrats. Overall numbers are fairly comforting for Hillary Clinton: Clinton 34, Obama 24, Edwards 16. But look at these four results, with Clinton and Obama numbers: ABC/Washington Post, 35-29; Marist, 37-24; Rasmussen, 33-26; Fox, 30-23. The averages of these four are 34 for Clinton and 26 for Obama--pretty close, actually. Is New Hampshire Hillary's firewall? Not necessarily. No wonder Clinton's going after Obama and not after Edwards, who in 2004 got 12 percent in New Hampshire after taking second place in Iowa with 32 percent eight days before. Actual vote numbers are, after all, even if four years old, harder than poll numbers from last week.

South Carolina. Pew polling suggests that South Carolina voters, especially on the Republican side, are far less focused on the contest and (therefore, presumably) far less strongly committed than Iowa caucusgoers or New Hampshire primary voters. Therefore, I'll eschew any analysis of the South Carolina polling today.

Finally, a note on national Republicans. cott Rasmussen's national poll now shows Huckabee leading Giuliani 20 to 17 percent. This represents a big loss for Giuliani over six days--the November 29 figures show Giuliani leading Huckabee 27 to 12 percent. How to account for this and for the fact that these numbers are out of line with other national polls-- Rasmussen employs a tighter screen, which tends to exclude independents from the Republican primary polls more than other pollsters do. Earlier in the year, Rasmussen was the one pollster to show Fred Thompson leading Giuliani. This doesn't mean that Rasmussen is right and other pollsters are wrong, or vice versa. It just means that they're looking at different universes.

Bottom line. I've written often that we're in a period of open-field politics. The foregoing is a good example of what it looks like: scrambled eggs.

By Michael Barone