Days of Change

Confirmation Bias | November 15, 2016

A trend is a trend until it isn’t.

Confirmation bias is the fancy term for giving more weight to statements that match your preexisting views on a subject. Media pundits and “journalists” are being beaten over the head over their conclusions that Donald Trump didn’t have a chance of winning. While polls had Trump behind the entire election season, he was within the margin of error for most of October, with polls tightening through those last weeks. He had a chance the whole time.

My confirmation bias tended toward a perception of Donald Trump as being a bad candidate who could not make up the difference in votes. In the last election, for example, it looked like there were 6 battleground states. Obama won 5 of them. I also have a belief that same day insta-voting and early voting make it easy for Democrats to collect big city dwellers to polls for weeks before an election and could target those states in play.

Here’s where the Trump supporters get roped in. Donald Trump won the election because he won the popular vote in a collection of states with the majority of Electoral votes. Trump’s chief pollster has taken credit for targeting states that mostly went Trump’s way in the final tally. Clinton’s campaign, in retrospect, extended itself too far, trying to win smaller, Republican-leaning states in an attempt to create a larger victory.

The Trump campaign interpretation is that he won this election, but the data looks more like Clinton lost the election. I suspect the confirmation bias among Trump supporters is that regular Americans didn’t feel like anyone was fighting for them until Donald Trump came along. Voters previously in the shadows came out for him. Going back to my confirmation bias, I was prepared to complain about fair weather conservatives who allow Obama to win elections and wait for things to go to hell before voting Trump. Then I looked at the numbers. Below is a visual representation.


As of this morning, Donald Trump is at about 60.5 million votes. This puts him below Mitt Romney’s 2012 total and a few hundred thousand votes ahead of John McCain in 2008. In the last 3 elections, the popular vote among Republicans has varied by less than 2%. On the other hand, Democrats went from a peak of 69 million to a low of 61 million, a drop that accelerated from -5% to -7% over three elections.

Even more interesting is that those voters who left the Democrats did not stay home. The total vote, when fully counted, will be less than 1% lower than in 2012. Clinton’s popular vote loss was the gain of third-party candidates. Again, more confirmation bias can be found here. 2016 saw the largest turnout of third-party votes since Ross Perot’s second run for president 20 years ago. However, both sides made the argument that it was a binary choice. Five percent of the electorate would say otherwise.

So, why interpret results if we could be wrong? The people in and out of power make decisions based on results, but also on the underlying factors. A Trump administration that believes America sided with them may not only eschew reaching across the aisle, but could also slap the hands of their detractors in the same party. In two years, there will be another elections where alternative choices are limited and Democrats will be hungry to reclaim power. Pride goes before a fall.

Or I could just be full of shit.


Posted in Uncategorized

%d bloggers like this: