Some point to the fact that Iowa voters vote in the general elections quite similarly to the national results. Mark Shields had a column in today's paper pointing out how closely the two parallel each other.
However, this completely misses the point. The relevant inquiry is whether the voters of each party in Iowa mirror their national counterparts accurately. On this score, it appears they don't.
Here is a list of times the Iowa voters have favored somebody other than the eventual nominee:
1992 Dems -- Tom Harkin 1st with 76%, Bill Clinton a sorry 4th with only 3%.
1988 Dems -- Gebhardt 1st with 31%, Dukakis 3rd with 22%.
1976 & 1972 Dems -- "Uncommitted" finishing ahead of eventual winners Carter & McGovern.
2008 Reps -- Huckabee 1st with 34%, McCain 4th with 13%.
1988 Reps -- Dole 1st with 37%, Bush 3rd with 19%, behind even Pat Robertson!!
1980 Reps -- Bush beats Reagan, though barely.
Iowa does not seem all that relevant, upon careful analysis.
9 hours ago