The NBC/Marist poll out today has some useful data to explain why Romney may win Iowa. Simply put, the state’s Republicans are increasingly pragmatic. When Marist asked likely caucus-goers back in November whether they preferred a candidate who shared their values/was closest to them on the issues, or a candidate who could beat Obama, values/issues won 60-21. The poll out today shows a 53-28 split on this question. If Romney wins, that change will more than account for his victory.
That said, Iowa Republicans still seem to prefer a true-believer over a tactical choice. In addition to the aforementioned question, the Marist poll also asked whether voters prefer a candidate who is a “true conservative” or one who can “best debate President Obama.” "True conservative" won by a 54-39 margin. That’s good news for Santorum, whom a plurality of Republicans saw as the true conservative in the field—23 percent, versus 7 percent for Romney. (Ron Paul was second at 21 percent.) Also good news for Santorum is that 16 percent of voters saw Bachmann as the “true conservative.” If the Minnesota congresswoman continues to flame out (she's at 6 percent in the same poll, so she could shed a few more points), I suspect Santorum will be the beneficiary of the conservatives suddenly searching for truth.
Finally, it's worth pointing out that, given Iowa Republicans' preference for a true believer, and their skepticism that Romney fits the profile, there's no way he could be close to winning there without so many rivals tempting conservatives. In the Marist poll, there are four candidates who receive double-digit support as the "true conservative."
[Anyone interested in hashing these posts out with me after-the-fact, or helping me work my way into them beforehand, can follow me on @noamscheiber.]