Since I first came to New Hampshire 19 days ago, the polls have been fairly steady. Donald Trump has held a two digit lead, in some surveys as much as 20 points. Ted Cruz has often been in second, and the four candidates in the so-called “establishment” lane—Bush, Kasich, Rubio, and Christie—have bunched together at between 8 and 14 percent. There are two outlier results: In the ARG survey about two weeks ago John Kasich surged to a clear second place, with about 20 percent, and fell back to 17 last week. In the Emerson poll released last Wednesday, Jeb Bush rose to second, at 18 percent. Kasich has certainly had a good couple of weeks, scoring a slew of local newspaper endorsements, plus those of the Boston Globe and New York Times, which could serve him well if he is still competitive when the campaigns go more national. Expectations for Jeb Bush had gotten so low that his rise in one poll had a bit of a surprise factor: I would say, having attended two of his events, that people tend to like him better after hearing him.

But none of the five Trump challengers could develop any sustainable momentum, and Cruz and Christie both seem on a downward trajectory in New Hampshire. One day of the news cycle last week was dominated by audio of Christie sarcastically berating a young woman at a town hall who was worried about storm flooding in New Jersey, where she had gone to college. It can’t have helped him. Cruz meanwhile has been taking heavy flak from everyone, and there simply are not that many hard-right people New Hampshire. His strategy was based on a win in Iowa giving him momentum for a strong second place here. But he’s now dropping in Iowa, and could conceivably finish third there behind Rubio and Trump. That would doom his chances for a successful New Hampshire finish, I think.

The thing to remember is that the Iowa results will probably shake up New Hampshire dramatically, and that far more change is likely in the next ten days than in the previous sixty. In that sense, politics is a little like NBA basketball: not that much is decided before the last seven minutes. For an historical illustration, here are snippets and polls quoted directly from Josh Marshall, covering the Democratic primary struggle in 2004:

Jan. 14th: One of the campaigns has tracking numbers out of Iowa which shows each of the top three — Dean, Gephardt and Kerry — clumped to within a point or two. And today’s Zogby poll seems to point in the same direction Dean 24%, Gephardt 21%, Kerry 21%. Meanwhile, the ARG tracking poll in New Hampshire shows some more movement after several days when everyone seemed to stay in place. Dean 32%, Clark 22%, Kerry 13%.

Jan. 15th: Surprising numbers out today. Zogby has Kerry not simply surging but actually in the lead in Iowa: Kerry 22%, Dean 21%, Gephardt 21%. Especially in a daily tracking poll a point difference means less than nothing (note to statisticians: don’t even say it). But this does square with what a lot of people have been saying about Kerry’s move in the state. Meanwhile ARG’s New Hampshire tracking poll has Dean 29%, Clark 24%, Kerry 15%.

Jan. 16th: Today Zogby has John Kerry opening up a five point lead over Howard Dean and Richard Gephardt in Iowa. The numbers: Kerry 24%, Dean 19%, Gephardt 19%, Edwards 17%. Even more interesting however are the ARG numbers out of New Hampshire. Today’s numbers there are Dean 28%, Clark 23%, Kerry 16%.

Jan. 17th: The Zogby poll out of Iowa continues to have Dean, Gephardt and Kerry grouped in pretty much a tie (Numbers: Kerry 23%, Dean 22%, Gephardt 19%, Edwards 18%). But the bigger news is out of ARG’s New Hampshire poll (Numbers: Dean 28%, Clark 22%, Kerry 18%) Clark remains a half dozen points behind Dean. But look at Kerry — back at 18%. A week ago he was at 10%.

Jan. 18th The new ARG New Hampshire poll out late this evening has Clark and Kerry in a virtual tie (Numbers: Dean 28%, Clark 20%, Kerry 19%).

Jan. 20th Iowa result: Kerry ends up getting more than twice Dean’s number, according to the late numbers. And of course Edwards too with more than Dean and Gephardt combined (Tonight’s ARG poll has Kerry ever so slightly in second place again (Numbers: Dean 28%, Kerry 20%, Clark 19%.)

Then after the famous “Dean scream” and Dean’s unexpectedly weak third place Iowa showing:

Jan. 21st: Not unexpectedly, John Kerry has made a big jump in this morning’s ARG tracking poll. The Numbers: Dean 26%, Kerry 24%, Clark 18%.

Jan. 22nd: ARG’s Dick Bennett seems to think that Dean is falling quickly into third place in New Hampshire. Today’s numbers are Kerry 27%, Dean 22%, Clark 19%. But a further breakdown of the numbers, provided on ARG’s site, shows steeper deterioration for Dean and some uptick for Clark.

Jan. 23rd: ARG’s latest daily tracking poll has Kerry 31%, Clark 20%, Dean 18%. That means that from January 19th to January 22nd Dean fell from 28% to 18%. In fact, from yesterday to today he fell 4 percentage points.

Jam. 24th: Some of the late polls show Dean stopping his slide going into the weekend, but having lost a lot of support since Iowa. This morning’€™s ARG poll says the deterioration of his support has ended. But they also have him down at 15% support.

Jan. 26th: All the tracking polls yesterday showed Dean, at a minimum, stopping his decline and in most cases making up some lost ground. Now the movement seems clear. Tonight’s ARG poll has Kerry 38%, Dean 20%, Edwards 16%, Clark 15%.

Following his Iowa win, Kerry won New Hampshire’s January 28th primary decisively: Kerry 38 percent, Dean 26 percent, with Clark and Edwards each at 12 percent

How could the GOP race turn within the next nine days? As mentioned above, Kasich and Bush have each made upward feints in the polls in New Hampshire, and Rubio has been trending upward in Iowa. The latest Des Moines Register poll puts him in third with 15 percent, but if he outperforms that even slightly he comes back to New Hampshire with a bounce that Kasich and Bush certainly won’t have.

Rubio’s appeal is lost on me, except that he is of Latino heritage and could supposedly solve the GOP’s problems with the nation’s changing demographics. At his events, he speaks of immigration movingly, from an immigrant’s perspective, and that is of course valuable. It appeals especially to members of the GOP donor class, who personally experience no downside from mass immigration at all. In his TV ads, he touts himself as a “national security expert”—though its not clear what that means, besides that he can recite with uncanny fluency neoconservative talking points. I doubt it’s accidental that his campaign’s slogan “New American Century” evokes the name of Bill Kristols’s PNAC group which agitated for war against Iraq from the mid 1990s onward.

McKay Coppins’s useful book on the Republican field contains nearly a chapter on the life of various Rubio “rumors” that began circulating in Florida political circles well before his 2012 Senate campaign. Some rumors involve financial malfeasance—many of which have been reported—while others involved reports of a “love child” or of various mistresses. None of the latter have been substantiated. Coppins describes the efforts by Rubio’s PAC to hire a private investigator in 2012 to see if there were any real skeletons that needed to be known about before his campaign begun. Nothing apparently was found. Thus far in this race, no one has gone negative on Rubio for his unusual use of government credit cards and the like. But that could come.

A possible wrinkle in the rumor field was raised by a Washington Post story, published last month. It recounts Rubio being arrested in a Miami public park when he was 18. It seems a banal incident, and the actual records of the arrest have been destroyed. One first wonders how the event—not recounted in Rubio’s biography—ever came to the Post‘s attention: it’s not like newsrooms these days are overstaffed with reporters eager to chase down minor peccadillos. I would conjecture that someone near Jeb’s campaign thought it worthwhile to inject the details of the arrest into the news cycle. The Rubio campaign responded, almost instantly, with an ad saying in so many words “Marco Rubio also once got in the express line with too many items” and “once tore the tag off a mattress.”

But the details of the arrest—because of the park, because of the subsequent career of the friend he was arrested with—have instigated glimmers of speculation in the Miami press, and also the gay press. My one twittersphere friend from Miami, a lawyer, Cuban-American, and Trump supporter, says she is certain that the speculation is groundless, and that the arrest was a random event in the life of a teenager, signifying nothing at all. And, I would add, even if it did signify something, its political meaning would probably be next to nothing. We live in a new era from the 1950s, or even the 1980s, and in most realms (several of my children have told me) sexual experimentation is normal and commonplace. It’s unlikely that the park arrest could be the silver bullet that could allow Jeb to surge in the polls past his ungrateful protege. But of course, if the two are close coming out of New Hampshire, one doesn’t really know how the South Carolina electorate feels about such matters.

In any case, a significant Rubio bump out of Iowa is possible, and Kasich and Bush are clearly hoping that does not happen. One thing that will probably happen after Iowa is that several candidates will drop out if they don’t do well. Santorum, Huckabee, and Rand Paul all are likely Iowa casualties, and it is worth exploring to whom their voters will go. My sense is that the sizeable Ron Paul vote from previous elections has gone to Cruz, who has scooped up many libertarians and small government activists, and to Trump, for his anti-establishment and non-interventionist positions. Rand Paul now polls at roughly 3 percent in New Hampshire (Huckabee and Santorum are lower) so there isn’t much to chew over. But if that 3 percent percent were to go mostly to one candidate, it would be significant. Trump, it should be noted, does poorly in response to “who’s your second choice?” questions, coming in sixth (!) with 6 percent in one recent New Hampshire poll. (Rubio and Cruz are tied at 17 percent.) That’s the kind of statistic which gives all his rivals ample reasons for hope, if only they could narrow the field.

Scott McConnell is a founding editor of The American Conservative.