Election 2016 Intro: Who am I? Why am I analyzing the Election?
I came into this completely backwards, am not a traditional political analyst, and am not even registered with any political party. In a ‘normal’ election year, this opening statement would disqualify me from credibly analyzing the election but in the current contentious, divided, and biased environment, such a description is an advantage.
Though light on politics, I do have a rather long investment background having worked as a hedge fund portfolio manager and as an investment analyst. More recently, I founded a company, ZettaCap, which uses social media and on-line activity patterns to help to analyze stocks and investments. Applying this background to analyzing the 2016 election provides unique insights into the race as it appears so different than the backgrounds of others analyzing the race.
In the summer of 2015, I began to apply some of the techniques that were used by ZettaCap to help forecast stock performance on the US Presidential Election. At the time, pundits, polls and betting markets were pointing towards Jeb Bush and Hillary Clinton winning their races without much fanfare. Our Social Media Influence index (SMI) somewhat surprisingly called for a fairly easy win by Trump in the Republican race and for a very tight race between Clinton and Sanders with Clinton having the advantage on the Democratic side. At that time, these calls were almost laughable considering the stance of pundits.
Our work continued to show that social media and on-line activity should work on elections just as it worked for product releases and stock performance forecasts. Sure enough, the SMI continued to make strong out-of-consensus calls in front of pundits. In an election year when so many pundits called both races terribly wrong, it was a huge boost to us to have called them correctly.
Around the beginning of 2016, I started to look more deeply into polls and analysis of the race itself. Frankly, this began due to the success of the SMI in making forecasts and then having the polls chase after the SMI. It seemed like social media and on-line activity were leading indicators to polls, and I wanted to understand the dynamics better.
This lasted until the early-Spring when it became more apparent that the general election would be Trump versus Clinton. Our SMI and our other social media and on-line activity measures were showing that Trump would win the election. However, polls, pundits, and betting markets consistently showed that Clinton would fairly easily win the general election.
I was perplexed. How could our indicators that were so successful prior to and during the primaries show Trump winning in the general election and pretty much every traditional measure show an almost landslide for Clinton?
This is when I dug in and began to look at the election as a fundamental analyst. I read every poll I could get my hands on, read almost every political analysis that was released and started to follow the betting markets – all in order to see how there could be such a divergence between the social media / on-line based analysis and traditional polls / pundit analysis.
The result is this series of posts. They are a mixture of some publications from ZettaCap and my own analysis of the election. For the most part social media and on-line activity analysis is provided by ZettaCap and the election analysis is my own.
You should know that I do have a bias. I am naturally biased towards social media and on-line activity analysis. These datasets are larger, more unfiltered, faster and more dynamic than traditional data such as polls. Having said that, in the base case scenario, the various datasets such as the ‘newer’ ones, like those based on social media, and the ‘traditional’ ones, like polls, should confirm one another. When they actually come to very different conclusions it is a true learning experience.
These posts are an exercise in leaning how these methods could produce such divergent conclusions regarding the 2016 election.
With this as a backdrop, please enjoy the posts. For comments or specific questions regarding the social media and on-line activity analysis, please use kevin at zettacap.com. For all other inquiries, feel free to contact me via kjcoogan at yahoo.com or on twitter at @kjcoogan.