Protected: Two ways to estimate primary outcomes without polls (transcript)
This content is password protected. To view it please enter your password below: Password:...
Tonight on MSNBC with Lawrence O’Donnell
Tonight I’ll be on MSNBC’s The Last Word with Lawrence O’Donnell to talk about Super Tuesday – and beyond to the November election. We’ll start around 10:20. Other guests include David Corn and Tavis Smiley. Tune in! Postscript: clip here.
(The plot above shows the predicted relationship between the Republican front-runner’s level of support and the total fraction of delegates won. It holds under today’s conditions.)
Topics:
Nicely Done Sam. It would have been nice to have a less clear set of polls. Massachusetts is certainly a key indicator.
Here is some new data and a data driven toy from Sean Trende and David Byler: http://www.realclearpolitics.com/articles/2016/03/01/whats_going_on_with_the_republican_race.html
It really focuses on demographics of the Republican voters in the primary states. It also got me to an insight. When Trump says he is loved by “XYZ” demographic, he is really just saying that he is doing better in that demographic than any other Republican nominee. I saw a discussion on CNN last night about his claims on minority voters and how he would do compared to McCain or Romney in vote shares. I guess it might also be a reflection of Hillary not reaching Obama levels.
Given his most recent comments, it is hard to understand how nothing seems to “stick” and his polls if anything get better.
Thanks for the link!
This isn’t scientific in any way, but an anecdote that perhaps addresses in some why what you’re saying about Trump’s ‘Teflon’ quality: Bill James (of Sabermetric fame) had an interesting essay recently where he describes Trump’s platform/agenda as being essentially apolitical. Whereas typically a politician courts voters at least in part through a policy agenda, it’s almost impossible to pin down a coherent agenda for Trump because he contradicts himself constantly. And part of the reason other candidates have a difficult time cutting into Trump’s numbers is because they’re playing a different game–they’re running on some variation of a typical conservative platform.
When Trump says he is loved by “XYZ” demographic, he is really just saying that he is doing better in that demographic than any other Republican nominee.
Or spinning a pure fantasy. “I’m great with the blacks!”
There’s been this widespread assumption that minorities aren’t going to turn out for the Democrats this year like they did for Obama, but South Carolina suggests that might not be true. I think the mere presence of Trump in the race counts for a lot.
So, in short, what is the best estimate of each Republican candidate’s total delegate harvest on Super Tuesday?
Anyone care to venture a guess?
Is the y axis predicted % of *total* delegates or is it predicted % of delegates that will be pledged through end of SuperTuesday? (Is it % of 2340 or % of 728?)
Nevermind. It’s gotta be % of 728. It can’t be % of 2340 because 728 is already less than one-third of 2340.
Exactly. Sorry – it was ripped from its original context.
I recorded the program and just watched it. I don’t think unfavorables matter (people are always comparing candidates, not rating them on an absolute scale) and I don’t think that recent developments in the news cycle affect the polls much (e.g. many people vote absentee). The median voter is very, very low-information.
I’m not seeing how Sanders can dig himself out of the delegate hole he will likely be in after tonight. He’d need to start getting 55% or more in most states from now on. The party’s proportional delegate allocation rules make it very hard to come out of any significant hole.