FiveThirtyEight has created a fun tool for projecting estimating whether Donald Trump will obtain the GOP nomination on the first ballot. Here is my best guess, as of this week:
Trump will end up with 1,215 delegates, 22 short of a majority https://t.co/4Ii17u1V7M #losewithcruz
— Blue Ball Skeptics (@BlueBallSkeptic) April 14, 2016
I’m probably being overly generous on the uncommitted delegates, but I figure Trump will pull out all the stops to woo at least a third of them over. (There are relatively few legal restrictions on what can be done in this arena, as reported by WaPo.) This will leave him short of the goal, but the margin is probably going to be razor thin. If so, this brings up the deeply disturbing possibility of Cruz supporters winning and Trump supporters rioting, in that order. Go play with the 538 widget, see if you can come up with a less depressing but equally realistic scenario.
On the Democratic side, it’s much easier to project since the electorate has been voting somewhat predictably based on underlying demographics. The following table is from a simple least-squares bivariate model which weights Blacks about three times as heavily as Latinos in determining the proportion of pro-Clinton voting:
The model may well miss on Connecticut and Pennsylvania, but the delegates from those states will likely be fairly evenly split regardless of who finishes first. Superdelegates aside, this would mean around 1,889 pledged delegates for Sanders and 2,162 pledged delegates for Clinton, following the close of Democratic voting on June 14th, which means that Clinton will most likely prevail. (Barring some sort of intervening black swan event, such as her name showing up in the D.C. Madam’s little black book.) Political prediction is a mug’s game, though, and I’ve been way off before during this election cycle. Grains of salt to taste.
