Princeton Election Consortium

A first draft of electoral history. Since 2004

The Circular Firing Squad Re-Forms

April 25th, 2016, 5:00pm by Sam Wang


One day after the announcement of cooperation between Team Cruz and Team Kasich, John Kasich has already gone off script. I question whether this alliance will hold.

For the goal of stopping Trump, avoiding division is important, not just for Indiana’s 57 delegates next Tuesday, but also for California, where Cruz and Kasich are dividing the non-Trump support.

Even if their efforts stick, Cruz and Kasich may be too late. Today’s PEC delegate calculation is driven in large part by two polls in California showing leads of 18 and 27 percentage points for Trump over Cruz. Kasich’s support is unlikely to make up this difference. At best, the campaigns will need district-level information to coordinate their efforts. That seems difficult.

In addition, I also have a smattering of polls that confirm the missing-poll imputations I published over the weekend. N.‘s Google Correlate method gave 57% for Trump in Connecticut; the 3-poll median is 54%. Correlate gave 66% for Rhode Island, and the midpoint of two fresh polls is 58.5%.

Today’s GOP delegate calculation includes an assumption of Trump +1% for Indiana (a median poll margin of 7%, minus a 6-point bonus for Cruz that I’ve written about before). I am now also accounting for faithless Pennsylvania delegates with a multiplicative factor of 0.8 (i.e. one-fifth of delegates will renege). This is based on the Tribune-Review data (see the this post for an explanation).

Current MATLAB code is here and here and here.

Tags: 2016 Election · President

12 Comments so far ↓

  • allen

    The +1% Trump assumption for Indiana seems aggressive in light of the luke warm collusion attempt. The real world data point we have is one from Rubio in OH. I think Rubio’s 2% share shows that many of his supporters got the message and voted strategically. Given Kasich’s 19% support relative to Trump’s +6% lead….I would give the edge to Cruz unless Kasich completely recants. How would the numbers look if you ran this at Cruz +1% or more in IN?

    • Sam Wang

      That number is 6% more favorable to Cruz than the current polling median. Doing more would be, in my opinion, putting a finger on the scale. Maybe you can do what I do, and wait for data.

  • bks

    Nostradamus appeared to me this evening and told me that Cruz will underperform the polling in Indiana.

  • mediaglyphic

    a very cool (but i realize time consuming) graphic might be a graph showing Drumpfs EV delegate count vs. time. (a la the obama job approval chart)

    I realize this might be time consuming and wish i could write the code to contribute! (i am learning python and R so might be able to contribute at some point)

  • bks

    It should be noted that voting in Indiana started three weeks ago. 60,000 GOP absentee ballots have already been mailed:
    http://www.nbcnews.com/politics/2016-election/lid-cruz-kasich-alliance-rough-start-indiana-n562006

  • JayBoy2k

    Timothy,
    I was once new here. This blog is refreshingly different. Dr. Wang has a real day job, and maybe as a give back , he shares his insights with us. This is a fact based blog focused on statistics, probabilities. and what the polls actually say — very fact based.
    More so than at any other polling related site, opinion, bias, thumb-on-the scale discussions are discouraged. Most of us view Sam’s time as precious, and a new person needs to go to the upper left column, check previous postings, and educate yourself. If you have the skills and find something you think could fix it, you could offer your services to fix it.
    Welcome. I hope you are the type of individual who appreciates the unique characteristics of this site.

  • timothy

    for us novices who read Dr. Wang’s information please explain the upper right hand corner
    Trump (R): 1303 (IQR 1271-1326), 90%
    how does the uninitiated understand that? and why did it change recently?

    • Nick

      Timothy, IQR refers to the interquartile range of Donald Trump’s simulated delegate total (i.e. the middle fifty percent of the simulated outcomes). Ninety percent refers to the probability of Donald Trump finishing with >= 1237 delegates. I guess the recent change is attributable to the three factors mentioned above: (i) new polling, (ii) Cruz outperforming his Indiana median poll margin by 6 percentage points, and (iii) 20 percent of Trump’s PA delegates defecting.

      @Sam: Long-time reader, first-time commenter here. Thank you for your outstanding work.

    • timothy

      more specifically, why doesnt the percentage of chance of winning read like this…
      xxxx (IQR 1237-xxxx), xx%
      why doesnt the probability being asserted begin with 1237 since that is what everyone is looking for the probability of ?
      ( this assume an ignoramus like me has an intuitive understanding of IQR)?

    • Catherine

      I believe 1303 is the median delegates. The IQR means 25% of runs get less than 1271 delegates while only 25% of runs get more than 1326. 90% of runs get at least 1237.

    • Sam Wang

      That is correct.

      Do not look at the probability…or I might leave it off! :-O Because there are lots of uncommitted delegates, the nomination process’s most important outcome is number of delegates, not the probability of getting to precisely 1237.

    • Jay Sheckley

      Speaking of little stuff in PEC’s upper right, I love the smiley afloat high in the right margin. Is she new? Who’s her author, Sam? What’s her name? :)