Replies

You’re wrong, Santorum’s support did collapse on the last day. You are right that Romney had a lead from the very earliest deciders but that switched to Santorum in February. You need to look at the exit poll:

http://www.cnn.com/election/2012/primaries/epolls/mi?hpt=hp_t1

Scroll down to the “Vote by Time of Decision” section.

Romney’s lead in absentees was probably from people who decided in 2011. A whopping 32% of them did and Romney won them 53% to 16%.

Romney just barely won people who decided in January.

Santorum won big from everybody who decided in February up until the last day. He was in the 50s, Romney in the low 30s. Santorum began losing steam in the last few days though.

Then suddenly Santorum collapsed on the last day. His support went from 50% to 31%, putting him way under even his January average of 41%. The votes that left him split evenly among Newt, Mitt and “Uncommitted,” whatever that means in an exit poll.

I do believe exit polls do incorporate polls from people who voted absentee.

I don't buy the exit polling in a winter election. The reason is snowbirds. They aren't home. If you've been to the Florida Gulf Coast, you'll see a lot of Michiganders. A lot of snowbirds keep their homesteads here (often in Northern lower Michigan) because of property tax caps. Non homesteads get taxed at a much higher rate than homesteads. My neighbors pay more for their small cottage up north than they do for their 4br house. With "prop a" the assessments are capped and the tax rate is limited its millage rate. The rate is often doubled on 2nd homes.

As for the collapse, people who made their mind up on the last day are likely not sold on anyone to begin with. They wern't the '55%' to start with.

I'd have to go through every county's election results to prove it, but I think this was a case of Santorum winning the official election day and losing the real election day that started when those ballots first arrived in the mail down in places like Bradenton, Sarasota, and St Pete.