The outcome of the recent US Presidential Election of 2016 shocked and baffled many. Some claimed that social media may play a larger role in influencing the outcome that expected. This study examined Twitter messages containing political discussions with references to both Trump and Clinton to uncover insights about the role of social media sentiment in political elections. We adhere to the social media analytics (SMA) framework of Fan and Gordon (2014) and the sentiment analysis taxonomy of Abbasi, Chen, and Salem (2008) as a structure to extract positive and negative sentiment from the collected tweets during the pre-election period between Nov 3 and Nov 7. The first finding reveals that Trump has an overwhelmingly larger volumes of total, positive, and negative tweets over Clinton implying a higher volume of public discourse around Trump. Secondly, the propagation of negativism towards Clinton is much more than Trump although both candidates have increasingly more negative tweets days leading up to the Election Day of Nov 8. Finally, word clouds for both candidates reveal that the Twitter public are engrossed with more negative topics against Clinton than Trump. This study clarifies the role of social media sentiment, specifically in how Trump is able to use Twitter as a conduit to reach his intended audience over and above traditional media. In addition, the influence of negative tweets seem to have a toll on Clinton creating distrust and weakening her political position especially among the working and middleclass communities that made the difference leading to Trump’s eventual victory.