I don't know if you have been following Chris Bower's Nomination At a Glance series at OpenLeft, but he has his final projections in. And he's calling for double wins in Iowa and New Hampshire for Mitt Romney and Barack Obama. His methodology is solid, but it doesn't take into account the very short time between the Iowa and New Hampshire contests and it is, by necessity, a victim of the tightness and unpredictability of the Iowa polls.

I'll be honest...Iowa is too close to call on both sides. I've followed the polls, the national media, the local media, and the history. But my projections are more from the gut than a result of any systematic analysis.

I am comfortable saying right now that Hillary Clinton winning tomorrow is the least likely outcome in the Democratic caucuses. What I'm looking for is hidden support. Obama's hidden support is big turnout, particularly among independents and young people, and Kucinich's endorsement on second choices. A big turnout will favor Obama (and it could favor Edwards vis-a-vis Clinton). Edwards has several potential sources of hidden support. First, he is polling stronger in rural areas, which are weighted more heavily than urban and suburban areas in the caucus system. In other words, Edwards will get more than one vote for every vote he receives, while Obama and Clinton will get less.

Second, Edwards is doing well as a second choice. In each caucus a candidate must pull 15% to get a delegate. If they get 30%, they get two delegates, etc. Some people are focusing on candidates that can't reach 15%, but it is more complicated than that. Here's an example.

In a hypothetical caucus, let's say there are 100 participants. Using the current RCP average let's say that the original tally is as follows: Obama (29 votes), Clinton (29 votes), Edwards (26 votes), Richardson (5 votes), Biden (4 votes), Kucinich (2 votes), Dodd (2 votes), and Undecided (3 votes). In this scenario, no one has 30 votes. Obama, Edwards, and Clinton will all be awarded one delegate (33.3%). It then becomes a contest between the three frontrunners to get to 30 votes and get that second delegate. In this case, there are 16 votes available. If Obama or Clinton can get 15 of them they will be awarded a third delegate. If Edwards can get a four votes, he will get a second delegate and tie Obama and Clinton.

As you can see, second choices can be critical. With the polls this close, any marginal advantage can be critical. Simply put, Clinton is going to lose hundreds of little battles over single delegates. Let's use another example to make this clear.

Let's say that we have a different caucus location (still 100 participants) with the following initial results: Obama (35), Edwards (22), Clinton (22), Other (21). If the 'others' split evenly (seven a piece) in second choices, Obama would have 42, Edwards 29, and Clinton 29...and Edwards and Clinton would get only one delegate to Obama's 2. But if Edwards can get one more undecided than Hillary, he will get 2 delegates and tie Obama.

If all that makes your head hurt, it gets even more complicated. In the above example, Obama has 42 votes. Forty-two votes aren't any better than thirty votes. Obama can send twelve of his votes to help another candidate reach a 15% plateau. If he wants to make sure Hillary loses, he may send those votes to Edwards. Or, he may send them to Kucinich if they will get him a delegate.

Clinton might get some second choices votes from Richardson and Biden supporters and she might do well among women, but she is going to lose the overall battle. I predict that Edwards will make up any narrow initial preferences deficit with a combination of rural weighting and his second choices advantage. However, I think he will end up essentially tied with Obama. Clinton will come in a close third.

In that scenario, I believe Obama will win New Hampshire, followed by Edwards, and Clinton again finishing third.

On the Republican side, I predict that Mike Huckabee will win in Iowa quite comfortably. John McCain will win New Hampshire in a landslide. And I don't think McCain will really have to break a sweat after that...he'll be the nominee.