Wednesday, May 4, 2016

Why I saw Trump's victory coming before other political scientists

Excuse me while I preen for a bit.

When Donald Trump entered the Republican presidential contest last summer, I wrote him off as a joke, just like pretty much every other political scientist.  I began extracting my head from my rectum in mid-December of 2015, long before most of my fellow political scientists.  I know of at least one who saw Trump's victory before I did:  Jon Krasno.  A smart guy, and a mensch.  Most of the field, though, kept denying Trump's chances for far longer than I did.

In fact, that's why this little blog exists.  Trump denialism was so rampant, even in early 2016, that I couldn't get a guest post on a political science blog pointing out how badly our models were doing.  This blog is where I can say what nobody else is willing to say, and on February 16, I posted the first in my "Trump to Political Science: Drop Dead" series.

Why did I get it right before the rest of my field?  Preening time.

1)  I had no stake in it.  Plenty of political scientists had either developed, or become attached to a particular model of presidential nomination politics.  If you wrote a book claiming that party elites control the nominating process (The Party Decides), or have become attached to such a book, it isn't hard to see why you have trouble accepting the reality of Trump's impending victory.

For the last several years, though, I have had a sort of gap on my syllabus for Introduction to American Politics.  During the week that I cover presidential nominations, I have been telling students about how all of our models suck.  I didn't have a better one, but I didn't believe the ones we had.  So, for several years, it has been a week of throwing up my hands in confusion.

But, that left me more open to the possibility of a Trump victory than most other political scientists.  (Except Jon Krasno.  Damn him for being smarter than me.)

2)  I'm a contrarian.  Does anyone even get the reference in the title of the blog?  My default is to question consensus.  I'm more prone to errors in which I wrongly reject the consensus view than wrongly accepting consensus.  In a weird year like 2016, that gives me an advantage.  Better lucky than good, right?

3)  I payed attention to the polls.  This is the big one.  As I pointed out yesterday, the 2016 race stands out for just how dominant Trump's performance was.  There is plenty of political science literature explaining that early polls don't predict outcomes very well.  That's why I spent the summer and fall of 2015 predicting that Trump's polling lead would evaporate.  The problem is that once January rolls around, the polls start to have predictive power.  The main error other political scientists made was in rejecting the polls for too long.  Trump never really lost his lead in the polls.  Eventually, that became the most important empirical observation for me.

If you are reading this pretentious little blog, it's because you think I have something useful to say.  And the next time you hear people talk about how flummoxed political scientists were by Trump's victory, you can tell them about someone who got it right before the rest of the field.  And direct them to this blog.  Why?  Because Jon Krasno doesn't have one.  He got it right before I did.  But hey, I beat most political scientists.  That counts for something, right?

OK, enough preening.  For now.

No comments:

Post a Comment