Now the Iowa caucuses have happened we can do some vague analysis of how my predictions stacked up. Lets look at the democratic results table to begin with:
Obama took first place as I predicted, however Clinton did rather more poorly than I originally anticipated, only placing 3rd. This isn't that surprising if you read the polling data for that in Iowa. Some of her campaign people were on TV before the results were out saying that it wouldn't be too tragic if she was 3rd. She also managed to avoid doing a Howard Dean-esque post Iowa TV meltdown, which is a shame really because "The Clinton Scream" or "The Hillary Scream" is a term I would very much like to enter the 2008 lexicon. I wouldn't worry too much if I were in her position, its just given people more opportunity for "comeback kid" reference to abound after NH. (Though I may predict Obama there, which will doom her)
She certainly did a lot worse than me on the night, with a mere 0.57 average place difference between my predictions and the actual results within the democrat results. <insert gloat here> Probably the biggest issue is that I had forgotten that Mike Gravel was still in the race, but apparently so did the electorate, since he placed joint last with Kucinich.
The republicans, however, were a hard race to predict and I didn't come off quite as well, though I was still respectable. Have another table of results:
I think a lot of this was due to not emphasizing importance on last minute shifts, in the electorate. For example the Thompson bounce that took him above McCain happened (if one is to believe the polling data) within the 2 days before the election. In other words - after my predictions! I think for N.H. I shall be leaving it to the last minute, in an attempt to improve accuracy, I notice most of the polling people don't actually predict until the evening of the caucus.
My average deviation was 4/3 places from the actual result, so still not too tragic, but none of the candidates were accurately placed on my behalf. More of a surprise for me was not that I didn't do as well as for the democratic race, but that Huckabee took a lot more votes than Romney, 34% vs 25%. Thats an incredible swing compared with only a month ago.
Huckabee has been rising in the polls for only about 2 months and fairly solidly, but it looked only 2 weeks ago as though he had topped - unfortunately for Mr. Romney he topped at the top, which I guess is a good place to stop. Everyone expected Rudy Guiliani to do badly, he never really tried in Iowa, so in some sense this isn't a terrible loss for him. It was interesting to here old Pat Buchannan on the radio last night claiming his strategy was terrible. The issue for Guiliani is that he needs to pickup some momentum BEFORE Florida now, since Huckabee is now leading in the polls in Florida. Since he won't be doing well in New Hampshire either, this means he needs something from South Carolina to give him a boost going into Florida. He has a chance in South Carolina- but not if he doesn't campaign there, he also needs to do this in a way that doesn't sound like an about face in terms of strategy.
I also said that I suspected that some candidates would drop out after the race, Chris Dodd and
Duncan Hunter (Democrat radio stations refer to him as "Duck and Cover" - how witty are they?) have fulfilled that element of my prediction. Unfortunately Fred Thompson did better than expected and so is still in the race, the rumours that claimed he was going to concede and back McCain appear to have been people going all Mark Twain on him.
My only remaining comment is roll on New Hampshire. There will be two republican debates before then (one on Fox, so that should bring out the wacko crowd) that I shall attempt to watch. I'm sure we will be hearing a lot from the democratic candidates as well. I suspect that there will be N.H. predictions and breakdown as well.