ADDED: Following the GOP primary, there has been another development. In most recent polls, Trump is clearly ahead in New Hampshire, with Marco Rubio a moderately strong second or third. In various polls he is second in most polls (by a few points) and tied in one. Kasich is generally right behind Rubio, with Cruz in third place in a few polls.
Rubio crashed and burned in last night's debate, according to most observers. And he really did. So, this may be reflected in New Hamsphire with Rubio moving down quite a bit. He crashed in part because Christie skillfully skewered him. I suspect this could bring Cristie's numbers up a bit. We ight be looking at an order something like this: Trump, Kasich, Cruz, Christie, then Rubio and Bush coming in fourth and fifth. Carson will not do well, and this may be his last primary.
Who will win the GOP primary?
First, let us dispense with the Republicans. (If only it were so easy!)
Trump is so far ahead in the polling that it is impossible to imagine him not winning. He is so far ahead, that if he doesn't win, the we can expect most of his financial backers to back away and his candidacy to be severely damaged.
Of course, since he is probably his own main financial backer, that will mean that a damaged candidacy will continue to lead the Republican pack for a while. But, really, that is not likely to happen. He is going to win the primary.
The more important question is who will come in second and third. There are actually three candidates that have a good chance of coming in second: Rubio, Cruz, Kasic
h and Bush. (In that order according to FiveThirtyEight's Polls-Plus forecast). This turns out to be a fairly complicated matter, then, when tying to interpret the meaning of New Hampshire going forward. So, I made a chart:
Who will win the Democratic primary?
This is more interesting at this point. We can see from polling data that Sanders is likely to with the NH primary. But the amount he wins by is going to determine a partial answer to that question of viability for him. Meanwhile, if Clinton does better than expectations, she will win kudos for organization and appeal. If Sanders and Clinton come in about as expected, meaning they both show well but Sanders wins, then New Hampshire will be sending roughly the same message as did Iowa: Dear Democrats, you have two viable candidates. Continue with the primary process.
But what is the number and how far off do the final results have to be before we can say someone did better or worse than expectations?
Looking at just the last ten non-partisan polls (ignoring likely voters vs. not likely, because that is part of the ground game) with all these polls overlapping January 20th or later, the Sanders-Clinton breakdown is 56.3-35.6. There is some O'Malley and undecided in there, so the ratio is more important than the number. So, the expectation for Sanders would be about 60%.
This conforms to the most recent polls, so any recent change (to date) is probably captured here. The total range is close to about 10 points.
So, I would argue, using gut instincts and nothing fancy, that Sanders will meet expectations with a percentage anywhere from 50% up. In other words, any level of win by Sanders meets expectations. If he gets more than 65% that may be meaningful, but since he is a) expected to do well and b) the state (within the party) matches him fairly well, I'm not sure how many points he gets.
Conversely, since we are so often asking the question in terms of insurgent Sanders' viability, if he loses by only a few points, a signal of concern will be sent to his campaign.
Looking at it from Clinton's point of view, every percentage point below 40% that she achieves will be a mark against her, showing weakness against the insurgent.
One thing is almost certain. New Hampshire will not be splitting hairs. This will not be close. Most likely the New Hampshire results will conform to the current polling, and the result will be that the hypothesis that Sanders can't be a viable candidate will not be falsified. I'm wording that in a fairly negative way, i.e., a good win in New Hampshire does not push Sanders viability estimate much at all. That sort of outcome is more likely to happen in relation to South Carolina and Nevada.
Not looking at specific numbers yet, if Sanders does not lose by too much in South Carolina, the hypotheses that he will do poorly among African Americans is not supported. If he wins in South Carolina, that hypothesis is in serious trouble.
In Nevada, if I'm reading things correctly, the outcome is likely to be stark, one or the other candidates winning handily, it can be either one or the other, and it will be a signal as to which candidate labor and unions is breaking for. To me, Nevada may be the most important of the first four races. (Aside from the unlikely scenario of the insurgent losing badly in Iowa or New Hampshire, in terms of meaning.)
The reason I say that Nevada will likely break either one way or the other is that I expect the unions to make a relatively unified decision I just don't know what that decision will be.
Greg are you taking into account the way Latinos are voting (no idea how many there are in NH)? - In Iowa they turned up in record numbers to register as Republicans so that they could vote against Trump. That may very well have hurt Bernie as otherwise they probably would have voted Democratic and he's the most progressive of the two.
That could matter a lot in Nevada, and I have left Nevada wide open, partly because of that. I assume there is a large overlap between Democrat/Latino/Union-labor in Nevada. I don't know what the unions are doing there. Sometimes unions do things rather quietly and you don't know what there plan is until it was.
The "latino/hispanic" population in New Hampshire is about 2%.
NH, Maine, and West Virginia are tied for being the whitest states. Vermont is the whitest state. (By the way, Nevada is 28% latino, 50% white)
Douglas -- Latino voters are 3% of ALL the voters in Iowa (the population itself is larger but not by much, Iowa is ~91% white and ~3% AfAm and ~2% Asian, so Latinos would be at most ~4% or so).
So while Latino voters could sway a general election if it is close, unless almost the whole group registered GOP I can't see it making much difference in the primary. Some difference, but not a lot.
1) I know very little about the Northeast and have never lived there, but I've continually heard that Sanders will do well in NH because of the "neighboring State" effect. However, I've also heard talk from people that lived there, that NH is more "establishment" than VT. Hillary won NH in 2008. This would leave me to believe, indeed, that if Bernie does better than 60% that it will be a big win for him.
2) Without a lot of research, information which might not even be available, I'm not convinced the "union" factor supports Hillary. Internet gossip is that unions that let the workers vote... mostly went with Bernie. Unions in which only management got to vote, and not the workers themselves, endorsed Hillary.
@Randy -- I know a bit about the Labor movement, and I can tell you that there is some division on this point. The SEIU leadership backed HIllary Clinton but the rank and file (the locals) were divided, or example, AFSCME was the same -- the Wisconsin chapters seemed to want Bernie Sanders and the leadership went with Clinton. Communications Workers of America went for Sanders whole hog. IBEW in Nevada went Sanders.
Generally, the big national unions went with Hillary, largely because the bigger locals did. So you could have a situation where some locals wanted Bernie and some wanted Hillary and the national leadership went with the majority (or at least the biggest and wealthiest locals). I don't know the details of every union and they all have different structures.
So call it a split decision. The unions are ambivalent about Clinton because her lukewarm policy support hasn't always helped them out. They're ambivalent about Sanders because he says the right things but has never been in a position to do a ton and even if he were president right now he still might not be, if you see what I mean.
In Iowa, snow held off until after the caucuses closed and did not put a damper on participation. Checking whether weather might affect the turnout in New Hampshire, I find that temperatures are expected to be in the 20s with little wind and only scattered snow flurries. This is fairly typical for NH in February and is unlikely to deter many voters.
Martin O'Malley never polled higher than single digits in New Hampshire, according to the New York Times. With his exit, I expect the majority of his supporters to switch to Sanders. It will be interesting to see how much difference this makes.