The Iowa polls:Second week of January

The Dems: Last week, the data set used to discuss the situation in January was only two polls, which for me is the absolute bare minimum. It is now up to five polls, a more robust sampling, and the story has changed very little. Hillary leads in three of the five polls, Bernie leads in the other two. Every poll has both candidates getting 40% or more, none has anyone getting 50% yet. For O'Malley, one outlier poll has him at 8%, the rest have him between 3% to 5%.

So the average of the five polls now has Hillary with a 0.6 point lead. Since this is effectively a two candidate race, I can use my Confidence of Victory method and give Hillary about a 70% chance to come out ahead using the median poll, where her lead is 42% to 40% in a poll of 504 likely voters. As more polls come in, these numbers are very likely to change, but in what direction I cannot say.

The GOP: We could say Iowa effectively looks like a two person race, but I am reluctant to use the Confidence of Victory method when the top two candidates account for barely over half the responses, unlike the Democratic side where the top two candidates get about 90%. The January numbers are now based on six polls, and the respective positions of the four top candidates have not changed in any appreciable way, with Ted 'The Snake' Cruz still holding a slim half a point lead over Donald Trump. A story on the fivethirtyeight.com website rates the Cruz ground game as much better than Trump's, but for a news organization that prides itself on using math well, the method used for this story is slipshod.

A number that looks unimportant that I still find interesting is the slowing rate of decline for The Completely Sane and Honest Ben Carson in Iowa. In New Hampshire, his numbers are down among the no-hopers, but in Iowa he is still dominating establishment guys like Christie, Kasich and Jeb:(, all of whom are in the race for second in the Granite State. Once again, the ground game can make a big difference in Iowa, where turnout is tiny compared to New Hampshire, so I would actually be surprised if the polls are a perfect predictor of the candidate's relative positions when The Voters of the Corn actually come out on February 1st.

Tomorrow, the numbers from New Hampshire.