The first votes of election 2016 will be cast in a matter of weeks and on Thursday the GOP conducted the sixth debate of its presidential candidates. As the competition intensifies, so too does the social media race–where the data provided by data analysis tells a different narrative than pundits and mainstream media in terms of who’s gaining momentum.

Throughout the course of Thursday’s debate, momentum shifted between candidates on stage–and on Twitter. The contest on social media suggested a number of interesting trends.

During the 2016 presidential campaign, TechRepublic is reporting on how technology, big data and social media are shaping the outcome of the race. Around the sixth GOP debate, we performed analysis of Twitter data to see how each presidential candidate Twitter account performed prior to, and immediately following the first Republican debate of 2016.

However, without context numbers and charts can often be confusing and uninformative. We are a news outlet, and not a data firm. We strive to use data responsibly to identify trends and narratives. While our conclusions are not polls, predictions or forecasts, they can help understand how debate performance echoes on and influences the social web. This is the same kind of analysis that companies like Ford use to predict consumer demand and customer behavior, and that more companies should take advantage of.

Our charts were produced with assistance from political scientist William P. Stodden, using Stata, a common data analysis tool. There was no magic involved with how we acquired our data. We pulled snapshots at 5:30 pm Eastern Standard Time on January 14, and at 8:45 am on January 15, and included follower count, listed count, followed by, to following ratio, and other publicly available data available on candidate accounts.

Here are the three most significant Twitter data takeaways from January 14’s GOP debate.

Takeaway 1: @RealDonaldTrump won big

Trump’s Twitter numbers were big. Really big. @ RealDonaldTrump already led the Republican field in followers, and over the course of the debate he tacked on an additional 10,568 followers. @TedCruz, Trump’s closest competitor, trailed significantly, earning only 4,611 new followers. @MarcoRubio remained competitive on Twitter, and tallied 3,854 new followers. The remaining contenders–@RealBenCarson, @ChrisChristie, and @JebBush–each gained a little over 1,000 new followers each.

Takeaway 2: @ChrisChristie surged

In terms of relative growth–how much an account experienced momentum compared to itself over time–by a significant margin Christie’s account appeared to experience the most momentum during and after the debate, with a relative growth of 1.64%, nearly triple that of Ted Cruz, whose account grew 0.645%, and ten times that of Donald Trump’s relative growth of 0.185%. This indicates that, relative to his previous influence on Twitter, Christie’s voice was significantly amplified. To be clear, while our data suggests that lots of people are checking him out for the first time, this is different than actually demonstrating proof of a true momentum shift.

Takeaway 3: @HillaryClinton and @BernieSanders also had a good night

The Democrats were not debating, and their data was included in our sample as a control. In this context, they can be seen as “people following politicians on Twitter.”

As expected, during the debate Republicans gained more followers overall. Average Republican Twitter follower growth was twice that of Democratic follower growth.

However, Democratic candidates also experienced a (mostly) positive night on Twitter. Frontrunners @ HillaryClinton and @BernieSanders each outperformed most of the Republican field. Clinton’s account added approximately 6,288 new followers, and Sanders gained 5,791 new followers. Former Maryland Governor @MartinOMalley lagged the pack, and gained only 130 followers.

Over the course of the evening Democratic candidates were added to user lists 0.015% more frequently than Republicans. The Sanders account experienced a small surge of 0.505% relative growth, a significantly faster pace than Clinton’s relative growth of 0.122%.

TechRepublic has a long history of reporting on big data. Through the course of the campaign, TechRepublic will use social media data to help create a variety of stories. If you are a data scientist, media or technology professional, or a loyal TechRepublic reader, we’d love your ideas for using data to tell the story of the 2016 election.

See Also: