Which Republican presidential candidate will have the most delegates after March 15?

Donald Trump seems to have something close to a 100% chance of having the most delegates after March 15. How did I arrive at this crazy result? The place to start, of course, is with the current state of affairs (morning of March 10):

  • 231 Ted Cruz
  • 25 John Kasich
  • 110 Marco Rubio
  • 329 Donald Trump

The total number of delegates available between now and the end of March 15 is 723[1]. By my count 471 of these delegates are awarded (more-or-less) proportionally, while the remaining 252 are winner-take-all. (RealClearPolitics counts Missouri as a winner-take-all state, but I’m not counting as one since it’s only a winner-take-all contest if a candidate receives more than 50% of the state-wide votes, and there’s fairly good reason to think that won’t happen according to the Kansas City Star[2]).

The winner-take-all states are easier to deal with, so let’s start with them.

  • Florida: 99 delegates
    • Probability of a Trump victory = 0.68
    • Expected Trump delegates: 68
    • Probability of a Rubio victory = 0.31
    • Expected Rubio delegates: 30
    • Probability of a Cruz victory = less than 0.01
    • Expected Cruz Delegates = 0
    • Probability of a Kasich victory = less than 0.01
    • Expected Kasich Delegates = 0[3]
  • Illinois: 69 delegates
    • Probability of a Trump victory = 0.67
    • Expected Trump delegates: 46
    • Probability of a Rubio victory = 0.28
    • Expected Rubio delegates: 19
    • Probability of a Cruz victory = less than 0.04
    • Expected Cruz Delegates = 3
    • Probability of a Kasich victory = less than 0.01
    • Expected Kasich Delegates = 0[4]
  • Ohio: 66 delegates
    • Probability of a Trump victory = 0.37
    • Expected Trump delegates: 24.5
    • Probability of a Rubio victory = 0.13
    • Expected Rubio delegates: 8.5
    • Probability of a Cruz victory = less than 0.08
    • Expected Cruz Delegates = 5
    • Probability of a Kasich victory = less than 0.41
    • Expected Kasich Delegates = 27[5]
  • Northern Marianas: 9 delegates (There’s not much information available, so I’m substituting the national averages. It’s not ideal, I realize.)
    • Probability of a Trump victory = 0.43
    • Expected Trump delegates: 4
    • Probability of a Rubio victory = 0.19
    • Expected Rubio delegates: 2
    • Probability of a Cruz victory = less than 0.17
    • Expected Cruz Delegates = 1.5
    • Probability of a Kasich victory = less than 0.07
    • Expected Kasich Delegates = 0.5[6]

S0 adding the existing delegate totals to the expected delegate totals as a result of the winner-take-all contests, we get the following:

  • 240.5 (=231+9.5) Ted Cruz
  • 52.5 (=27.5+25) John Kasich
  • 169.5 (=59.5+110) Marco Rubio
  • 471.5 (=142.5+329) Donald Trump

And now my life gets a little easier since I’m writing this on March 10. There are 155 delegates up for grabs in Kansas, Kentucky, Louisiana, and Maine. Here’s how things shook out while I had my head in the clouds:

  • Cruz 265 (=204.5+23+12+13+13)
  • Trump 513 (=471.5+9+9+13+11)
  • Rubio 180 (=169.5+5+0+0+6)
  • Kasich 59 (=52.5+0+2+0+5)

(I’m dropping the spurious 0.5s from each of the scores as they were just an artifact of being unwilling to round up or down.) So that leaves 273 delegates to be dealt with. And Cruz would have to win 249 (=514-265) of the remaining delegates to pass Trump. That’s slightly less than 90% of the available delegates, without the benefit of any winner-take-all states, since I’ve already figured them in above. Can Cruz do this? Well, anything is possible, but he’s been averaging about 30% (mean = 0.3107, median = 0.3044, with standard deviation = 0.1739). So the probability of this happening is 0.0003. Or zero, as I like to call it. Even more so (if that makes any sense), since in order to beat Trump with 514 delegates, Trump would have to resist – somehow – picking up even 1 delegate from the next 13 contests. Of course, the picture would like quite different if Cruz had a reasonable shot in the Miami, Ohio, and Illinois. Right now, the polls suggest he’s miles behind Trump, but that could change, and if it does, I’ll adjust my forecast.

But all of this leaves me with a problem, because at the moment it seems like my model (if you can even call it that) is telling me that Trump has a 100% chance of having the most delegates after March 15. That can’t be right, can it?

[1] Thought they could have been taken from almost any other good source, I gathered these numbers from http://www.realclearpolitics.com/epolls/2016/president/republican_delegate_count.html

[2] http://www.kansascity.com/news/local/news-columns-blogs/the-buzz/article50890980.html

[3] http://projects.fivethirtyeight.com/election-2016/primary-forecast/florida-republican/

[4] http://projects.fivethirtyeight.com/election-2016/primary-forecast/illinois-republican/

[5] http://projects.fivethirtyeight.com/election-2016/primary-forecast/ohio-republican/

[6] http://elections.huffingtonpost.com/pollster/2016-national-gop-primary

 

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s