During the 2016 U.S. presidential elections, Twitter served as an important platform for the spread of news articles, which have significant influence on public opinion. Yet the sharing of stories is often based on innate emotional triggers, seldom rational. In our research, we seek to examine whether the emotional vocabulary of political news stories can lead to their popularity. To explore these questions, we construct a corpus of 2,650 articles from 12 different news publications over 5 months, connected with the 123,113 tweets by 20,964 Twitter users that share them. Using the Harvard Inquirer lexicons, we automatically code stories for emotionality and positivity. We then run regressions between the independent variables of story length, emotionality, and positivity and the dependent variable of number of shares across 7 different political divisions of Twitter users, as well as the collective dataset. On the whole, we find Twitter users to favor stories that are Hobbesian in nature: nasty (negative in positivity), brutish (high in emotionality), and short (low in word count). However, differences emerge when considering different levels of political engagement among users.