Blog / Believe It or Not, Social Media Data Predicted Trump’s Victory Topics  

Believe It or Not, Social Media Data Predicted Trump’s Victory

Why are you surprised?

Unlike any race before, this election was won and lost on social media. Trump and his mastery of social media was the game-changer. He mobilized his masses and pissed off his detractors. While most pundits predicted it was over before election day started, the masses flocked to social media and spoke out loud.

For the past 18 months we’ve analyzed and benchmarked each candidate, tracking their performance versus a four-week moving average, letting social data tell the story. We accurately predicted Iowa and a majority of the Super Tuesday states.

For election day, we created a sentiment tracker for Reuters, who has been using Zoomph to better understand America’s perspective on the race. As you can see in the timeline, Donald won against Hillary.

 

Elections social media sentiment

 

At 3PM, I joined Vincenzo Landino’s Aftermarq FaceBook Live show. They broadcasted throughout the entire day, and wanted my perspective on how the race was unfolding.

I pulled social media data from 6AM through 3PM for the nation, then Florida, Pennsylvania, and Ohio. comparing the ratio of positive/negative sentiment for Donald versus Hillary. Despite the New York Times suggesting an 85% chance that Hillary would win Florida, I contended Trump was ahead. Social data saw Trump’s ratio of positive/negative activity double that of Hillary’s.

In this chart, you can see Trump earning six positive posts per minute in Florida, versus 3 per minute for Hillary. As for negative sentiment, each candidate received 3 posts per minute. Trump’s ratio was far stronger than Clinton’s.
 

Trump Hillary Social Media Sentiment

 
We saw similar trends for Pennsylvania, Ohio, and all the battleground states. For the nation, the ratio was 1.4 versus 1.2 in Donald’s favor.

Some other interesting stats we saw in Florida (and echoed across the nation):
 

Ethnicity

Trump earned more positive activity from Hispanics and African Americans compared to Hillary.

Elections Florida Sentiment by Ethnicity
 

Gender

Comparing gender, 51% of the positive conversations for Hillary were from females compared to 48% for Donald.

Elections Florida Sentiment by Gender
 

Generation

From a generational perspective, it was a relatively even distribution talking about both candidates.

Elections Florida Sentiment by Generation

 
We started this journey with a hypothesis that predictive social media analytics can be more accurate and powerful than traditional polling.

With a solid methodology in place, we believe media outlets and politicians need to embrace social analytics to keep a finger on the pulse of their constituents.

With this historic election behind us, I am reminded of a famous quote from Winston Churchill:

“Now this is not the end. It’s not even the beginning of the end. But it is, perhaps, the end of the beginning.”

I’ll let you interpret what that means, and will leave by congratulating Mr. President Elect on a game well played. Go forth and unite this divided country.
 

Topics:
WP_Term Object ( [term_id] => 728 [name] => Social Media Monitoring Resources [slug] => social-media-monitoring-resources [term_group] => 0 [term_taxonomy_id] => 729 [taxonomy] => capabilities [description] => A collection of advice on how social monitoring can help you spot trends and keep tabs on your competition. [parent] => 0 [count] => 44 [filter] => raw [object_id] => 14925 )
Monitoring & Tracking
WP_Term Object ( [term_id] => 733 [name] => Social Media News and Trends [slug] => social-media-news-trends [term_group] => 0 [term_taxonomy_id] => 734 [taxonomy] => updates [description] => Get the scoop on social news and trends, plus expert analyses, advice, and hacks for joining the most important conversations. [parent] => 0 [count] => 134 [filter] => raw [object_id] => 14925 )
News & Trends

Subscribe to Zoomph's Newsletter

Top 15 Influencers on Twitter by Industry—February Recap

March 07, 2017

On Twitter alone, your brand now competes with 7,534 tweets each second, shrinking your opportunity to be discovered on a day-to-day basis. But a select number of brands have managed to carve out strong identities,… Read More

ZIndex: A Social Media Index for Staying Ahead of Industry Trends

March 04, 2017

It's impossible to win without knowing what you're up against. And in today's always-on, on-demand age, the race to get in front of our customers is especially vicious, making it ever-more important for us to… Read More

What Is Social Identity Data?

February 27, 2017

For starters, it’s a term that our team uses a lot. It’s what we’ve identified as the wave of the future, not by our own creativity, but by the witness of our fellow data geeks… Read More

How to Bring the “Custom” Back to Customer Experience Using Social Media

February 10, 2017

Across every industry, and in every space, there are countless companies jockeying for the same consumers. When it comes to researching products and services, buyers will often come across a number of companies with similar,… Read More

All Posts