This research examines the key characteristics of hyperpartisan news pages on Facebook and how audiences interact with politically polarized content through the visual-emotional shorthand of Facebook Reactions. Through a quantitative content analysis of 4,236 posts shared by the most popular hyperpartisan U.S. Facebook pages before, during, and after the 2016 U.S. Presidential Election, the researchers introduce the concept of affective affordances to analyse emotional reactions elicited through Facebook Reactions in response to right- and left-leaning Facebook news posts, as well as the political topics, rhetorical devices, stylistic devices and emotionally charged content that are most likely to elicit emotional responses and inspire shares and comments from audiences in reaction to liberal and conservative content. The results are interpreted in light of the theory of affective intelligence.