March 5, 2016
Not happy with this one. For better or worse, I’m feeling confident about our Kansas, Louisiana, Maine, and Nebraska predictions, despite lacking any data in two of them, and not having enough in a third.
The GOP has Kentucky all to themselves today, courtesy of the Rand Paul presidential campaign, which needed a separate date in order to allow him to run for president and Senate re-election in the same year, while complying with state law.
Western Kentucky University has the polling universe all to themselves, having provided us with the single available survey. It was taken between 2/22 and 2/26, and features 22% of voters who were undecided or supported Ben Carson.
Lots of room for interpretation. Rubio was ahead of Cruz, 22/15. At that time he was ahead of Ted other places where he wound up finishing behind when Super Tuesday votes came in.
In states with polls before and after Super Tuesday, Cruz has improved his position compared to Rubio. We should probably assume they’re either tied, or Ted is ahead. I’m not sure which though.
Then there’s the problem of Kasich. He’s the governor of a neighboring state. He polled at 6%. A few days ago, this looked like another reason why he was a boutique candidate at best. New (at the time) surveys in Michigan were showing him as low as 8%.
The simplest explanation was his appeal didn’t cross state lines. We remembered his crowning achievement was reaching the mid-teens in New Hampshire after becoming a temporary resident.
Then he got 30% in Vermont, did well in the debate, and has seen his numbers go up in Louisiana, Michigan and Kansas. Rubio appears far less inevitable as a final Trump challenger.
Kasich will surely do better than 6% now, but how much better?
Then there’s Trump. Can anyone catch him. He had 35% in the single poll, 13 points ahead of Rubio, and a full 45% of the support given to candidates still in the race. It’s a Trump-friendly place, with a dose of populism.
A few days ago, I thought Rubio could grab enough undecided voters with a strong finish to make this potentially interesting, though The Donald was still favored. That seems completely off the table right now.
Cruz trailed him by a full 20 points. While I fully expect the gap to close during the real vote, and think the caucus format is to Ted’s advantage, that’s too much to make up.
Given all of the above information and speculation, I’ve reached the following conclusion/wild-ass guess about today’s outcome:
Donald Trump 38.8%
Ted Cruz 28.6%
Marco Rubio 18.6%
John Kasich 13.8%
Let’s see how close this estimate turns out.