Just how important is historical precedent for the election in 2016?
For those of you that haven’t been following my Twitter account, I’ve been pretty intrigued with calculating possible scenarios and victories for the presidential primaries. I spent a couple weeks spamming predictions and results, but in order to stop a political overflow on my site (I went about a solid month without posting any history, psychology, or sociology because of it), I started doing independent research and county-by-county predictions with a group of friends of mine.
I, so far, have about a 90% accuracy rate. It’s really not that hard to predict these sorts of things, so I have no idea why people praise FiveThirtyEight as the gods of political (and sports) predictions. No, I don’t have a fancy formula. No, I don’t plug in concepts to some statistical equation. No, I don’t blindly point at candidates in the dark. I pay attention to polling data, people. I look at demographics, living conditions, and what’s going on today. That’s what matters most.
Now, granted, some polling data is horrendously skewed — either by missing important variables and factors in the average voting scenario or through your typical media bias and slander. Michigan, for example, is a state that everyone saw going to Hillary Clinton. However, after Bernie Sanders’ wonderful victories at the Flint Debate on CNN and the Fox News Town Hall (yes, Clinton fans, I’m sorry — not a single poll had Clinton beating Sanders in either of these two events), polling data wasn’t able to refresh itself to show possible connections or a correlation in a possible Sanders uprising. It’s being called the “biggest upset in modern political primary history” for a reason.
But let me say something about Michigan. A progressive movement is building quickly in the state due to the Flint Water Crisis and the abhorrent leadership from Governor Rick Snyder. This isn’t looked at in demographics, for the most part. That last minute push for Bernie Sanders (especially after the Flint Debate) might have allowed that victory to take place.
Paying attention to polls, whether they be county by county (like in states such as New York, Michigan, Florida) or statewide (states like Hawaii and other caucuses tend to be harder to verify through county polling), can basically show us lines of prediction for both the Democratic and Republican nomination.
But, that brings me to a good point: whether the state has a primary or a caucus drastically changes the principle of the election. It’s much harder to predict who will win caucuses, because caucus turnout is oftentimes lower comparative to primaries.
I’m not a fan of caucuses. They seem to be a 19th century styled format of “if you’re here, you count.” They don’t tend to last all day, making it impossible for people who have to go to work or can’t get through traffic stuck in a position where their democratic vote is lost to the tests of literal time. Primaries last all day, seem to be more official in terms of privacy and voter rights, and give us larger polling opportunities and a general idea of what’s going on. So, in states that have caucuses, our overall conception of “who’s going to win” is skewed with a simple factor: only a fraction of the fraction that were going to vote in a primary are going to show up for a caucus.
I’ve gotten a primary or two wrong as well, of course. I pay a little too much attention to historical precedent in a race that has obviously been swayed away from every factor of precedent imaginable. We’ve ushered in a new age of politics, one where Bernie Sanders and Donald Trump have chances. New Hampshire remains to be my biggest upset (in both terms of numbers and personally) for my prediction on the Republican side.
New Hampshire Republicans like establishment candidates. Let me say this again: New Hampshire Republicans like establishment candidates. Romney won in 2012, Bush won in 2008, McCain won in 2000, Buchanan won in 1996, Bush won in 1992 and 1988, Reagan won in 1984 and 1980, Ford won in 1976, Nixon won in 1972 and 1968. Do I need to go further? New Hampshire Republicans like establishment candidates. […] and that really leaves Kasich as a favorable establishment candidate. — February 9th, 2016 “My Predictions: New Hampshire Primaries”
Well, unfortunately, the pattern of historical precedent has been thrown to the wind in this case. Yes, yes. Establishment candidates up the wazoo throughout elections’ past…yet the establishment candidates didn’t succeed in New Hampshire in 2016.
Thus, I make my case for caution when claiming “it hasn’t ever happened in history before!” A Trump has never made it this far in a presidential race before, either. A Bernie Sanders hasn’t been able to get $45 million in grass roots funding in one month before, either. A Jeb Bush or a Marco Rubio seem to always do well in polls.
So, for all of you all so high and mighty victors of “precedent”, this election has thrown all previous statistics out the window when it comes to polling data. Of course, history still reigns. Trump’s statistics rely heavily on specific demographics and voting bases.
Historical Precedent for Trump’s Polls…
For example, take a look at how he does with white voters that do not have a high school diploma…
Now take a look at how Trump does with people who, when asked about their heritage, simply call themselves “American”:
Now look at, my personal favorite, how Trump does with areas of the United States that voted for segregationist and blatant racist George Wallace in 1968:
For copyright and sourcing purposes, these three charts come from Nate Cohn over at The New York Times.
So, of course, in an election that has thrown all historical precedent to the wind, there still remains some historical correlation when it comes to ideas, stances on the issues, and the demographics of the voting base.
Individual states still will lean either left or right. Influence from Canada in the northern peninsula of Michigan and in Maine will still benefit Democratic candidates. Idiots in Texas will still always vote for idiotic Texans (Ted Cruz).
Previous elections and candidates, such as Wallace in 1968, will still benefit candidates that mirror or relate to that specific type of voting base. My point isn’t that we can blatantly ignore those factors. But we sure has hell can lessen the impact of such comparisons.
When it comes to depictions of establishment? To some extent specific race (with Latino votes critical in this election)? Or even when it comes to what’s actually gone down in the past when our modern election seems to want nothing to do with the past? This election is changing historical precedent. This election is changing how we think about polling. FiveThirtyEight exited 2012 with a 100% passing rate with polls…they’re nowhere near that in the first quarter of 2016.
Oh…while I got you here.
Ohio? I see it going to Bernie Sanders, but not by much. Sorry, John Kasich, you won’t be able to take your political home-state away from Donald Trump. I’m taking Michigan not with a grain of salt, but with the entire salt shaker. The mid-west has been shaken up.
Illinois? Clinton, but not by much. I’m taking that upheaval in Michigan as a factor for the mid-west, now. Donald Trump will wipe the floors with everyone else.
Florida? Going to Clinton and Trump with numbers almost as high as what we saw in South Carolina and Mississippi. I’m enjoying the insane back and forth.
2016 is the year that will possibly start a new historical trend. With progressivism and white angst populism on the rise, America — regardless of who wins and stays in the White House for four full years — is going to be interesting.
For the Democratic nomination, I still believe that if you take out super-delegates, the race isn’t exactly easy to call yet. I still hold onto that faith that something could spark up a change. It’s a super small and tiny chance, but it still remains. Precedent be damned. As for the Republican nomination? Hope you like Donald Trump.