Tomorrow (February 1st) is the first time that voters will actually start making meaningful decisions as to who will be the nominee for President.
Obviously, this isn't new, but it has extra significance in my eyes for one reason, and that reason is Donald Trump.
For months, I've maintained that Trump has huge support in the polls because there is no consequence to supporting him. People might not like Trump as a candidate, but they like some of his positions (namely immigration) and they like that he says what he wants. They can handwave away all of the massive negative stuff about him because it doesn't matter--claiming to support Trump is an easy way to 'send a message' and then, when it comes times to actually vote, pick a different, more appropriate candidate.
As I said, I've been maintaining this opinion for months, and, sadly, I'm not sure if it still holds.
I still think there is an element to it. Iowa and New Hampshire are big, flashy spectacles--the sort of thing that Trump revels in. Once it starts grinding down into the rapid pace of weekly primary contests and Trump can no longer employ massive, long-term media coverage and has to rely on his bare-bones organizational skills, he will start to falter. (Trump has very little staff and an alarmingly undeveloped ground game.) I don't think he will weather Super Tuesday well at all, because his campaign staff is largely nonexistent, and that's not the sort of thing you can throw money at and whip together quickly.
The problem, though, is that aside from Ted Cruz, Trump is double digits above everyone else. Even if 5-8% of Trump's support drops off due to the above factors, he's still in first or maybe second place.
Anyway, tomorrow will tell a lot, but I don't think it will tell the whole story. Even if Trump wins both Iowa and New Hampshire, I don't think he'll last through the rest of the primaries. More worryingly for the Republicans, though, is that if not Trump, who? There are still too many candidates, and the anti-Trump (and even anti-Trump/Cruz) forces are split up amongst six or seven candidates. Even if Trump falters, if the other candidates each take turns winning primaries and no clear winner is found (a fear that happened for about a month in 2012) it may lead to issues. While, as a wayward political scientist, the chance that we may have a brokered convention would be fundamentally AWESOME, it might not bode well for the party overall.
In the end, many people (myself included) think that Trump will implode, and his commanding lead vs the sheer number of other candidates mean no one has any idea where those Trump voters are going to go, so even a candidate who is polling single digits today might still be a front runner. Most of the establishment candidates (namely, Bush, Christie, and Kasich) are biding their time; their low poll numbers won't matter when people abandon Trump and look for a more seasoned politician.
As for the Democrats, it's not quite as exciting, with only three candidates (and only two realistic ones). The Democrats caucus differently, in that non-viable candidate votes get reassigned (unlike the GOP, where people vote and then can leave). In practice, this means that Martin O'Malley voters might split in a few different ways (most polls show an even split, but you never know).
Still, the Democrat side could tell a lot: Iowa is one of Bernie Sanders' most promising states, so a loss here would make it problematic (although not difficult) to continue (his win in New Hampshire later in the month, neighboring his home in Vermont, is a shoe-in so doesn't mean much.) But even if it's a close call and Hillary Clinton still wins, he can probably spin that into a moral victory of some sort. So Iowa probably won't change much unless Sanders is completely blown out.
For both parties, the next voting is not for another two weeks, so whoever ends up winning (or has an unexpectedly strong showing) will dominate the media for those two weeks. However, it's always important to remember that winning Iowa doesn't always mean much: the winner of the last two GOP contests, Huckabee and Santorum, ended up going more or less nowhere (although Iowa gave Santorum a bit of a boost he probably wouldn't have otherwise had).
For the GOP: Trump has about a 6% lead as of this writing, and I suspect that will evaporate pretty quickly. I would put a Trump vs Cruz victory at about 50/50 for each. However, I'm more interested in the other candidates: I think Rand Paul has a chance to show off his organizational, youth-centered prowess and has a strong showing relative to his polling numbers; John Kasich will use his viability in New Hamsphire (where he's currently polling a surprising second) to gain some traction; and one of the other low-level candidates will have a surprising showing, possibly Christie. I think Carson, Bush, and Fiorina have bad nights, Huckabee and Santorum cancel each other out (and also have bad nights), and Rubio comes in either third or fourth.
For the Democrats: Hillary wins, but only by a small margin, so Sanders can claim strength as a victory. O'Malley won't be viable (you have to get at least 15%) but I think it's possible he punches above his weight and snags higher than what he is polling, adding a bit of strength to his strategy of being a decent alternative option.