The Los Angeles Times released a new Iowa poll on Tuesday:
Clinton 28
Edwards 23
Obama 19
Richardson 10
The results have been discussed at MyDD and Open Left:
Jonathan Singer made the key observation at MyDD:
I do just want to add that the “likely” voter screen for the LA Times and Bloomberg is a bit loose. Doing some back of the napkin math, the likely voter model used here would include more than 870,000 Democratic caucus participants — which is a bit more than the 122,000 or so that participated in the last Democratic caucus. While there may be more excitement this time around than there was in 2004, there isn't that much more excitement. So what does that mean? A lot of the folks the Times and Bloomberg are deeming “likely” voters or caucus goers are in fact unlikely to participate come January. Most in fact. As such, while these numbers might be good gauges of the general sentiments of these states, I'm not certain how good of gauges they are of the sentiments of those who are actually going to play a large role in selecting the next Democratic nominee.
The problem with almost every Iowa poll I've seen lately is that the universe of likely voters sampled suggests a ridiculous number of caucus-goers.
If turnout exceeds 150,000 at the Iowa caucuses on the Democratic side, I'll be surprised. If it exceeds 200,000, I'll be shocked.
My sense is that a lot of Hillary supporters and leaners have never caucused before and won't caucus this year either–especially if we have to go the week after New Year's.