US Elections 2016: A Final Big Data Perspective

Peter Hafez - Chief Data Scientist - RavenPack | November 08, 2016

This has truly been a roller coaster ride of an election campaign. Read our final big data perspective.

It's Polling Day!

We’re finally at the end of what has truly been a roller coaster ride of an election campaign. Since our first update on the 26th of October , FBI Director James Comey has announced the reopening of the investigation into Hillary Clinton’s use of a private email server and subsequently confirmed that there are no grounds for charges against Mrs. Clinton.

It is doubtful that we’ll see the effect of Comey’s recommendation not to pursue charges against Mrs. Clinton, but the news sentiment for Mrs. Clinton spiked following Director Comey’s announcement, and the spread moved back in her favor after having moved in favor of Donald Trump for a short time.

As we did in the first article, we’re going to look at the 14-day rolling average of the sentiment spread vs. the poll spread. Unlike our first article, we’ll also take a look at the 3-day rolling sentiment and poll spread numbers to focus on the short term impact of media sentiment from Comey’s announcement. We’ll leave the final judgement to you on whether or not the sentiment reversal in favor of Mrs. Clinton will translate into a better margin at the polls today.

Sentiment Spike in Favor of Clinton

In the 14-day rolling average graphic for both the spread in sentiment and poll numbers between Mrs. Clinton and Mr. Trump, we can easily see how the October 28th announcement by FBI Director James Comey reversed the momentum Mrs. Clinton had built up in her favor.

alt text

The polls, which had already been narrowing since mid-October, narrowed further on the back of the announcement. Using a 14-day rolling average, however, doesn’t tell the full story in this case. In order to fully account for the magnitude of Comey’s announcement on November 6th that the FBI has found nothing on Mrs. Clinton that warrants charges, we need to decrease the size of our rolling-window. With the objective of isolating the post-announcement effect, we changed the rolling window to a three-day moving average.

alt text

There is far more variation in the data, but if we focus on the very end of the time-series, we can clearly see the effect on media sentiment of Comey’s announcements. Before Director Comey reopened the investigation, the sentiment spread trended significantly in favour of Mrs. Clinton. Off the back of that news, it quickly reversed direction towards the end of October. On the 6th of November, when Director Comey announced that no further action would be taken against Mrs. Clinton, we see the sentiment spread move dramatically in her favor again.

The chaotic nature of the short-term movements in both the polls and sentiment measures make predicting off the back of any one spike a perilous exercise. That said, we do know that there is a high correlation between the sentiment spread and the polling spread when we use a larger rolling-window so it certainly won’t hurt Mrs. Clinton’s chances. The main question is whether the last two days has been enough time to recover with undecided voters? Hopefully, we’ll get our answer before the day is out. Happy voting!

By providing your personal information and submitting your details, you acknowledge that you have read, understood, and agreed to our Privacy Statement and you accept our Terms and Conditions. We will handle your personal information in compliance with our Privacy Statement. You can exercise your rights of access, rectification, erasure, restriction of processing, data portability, and objection by emailing us at in accordance with the GDPRs. You also are agreeing to receive occasional updates and communications from RavenPack about resources, events, products, or services that may be of interest to you.

Data Insights

Read More