We develop a new method to quantify media content based on people and their associated Wikipedia topics (e.g. political affiliation gender race etc). Our method is objective precisely measured across a wide array of source material and admits both cross-sectional and time-series analysis on topic intensity and sentiment. We apply our method to a novel dataset the closed captions of cable TV news. We find that Fox News covers Democrats more extensively and more negatively than other news networks while MSNBC focuses airtime on impugning Republicans. Outside of politics we find that Fox News is more negative about females African-Americans and LGBTQ people relative to CNN and MSNBC. Across business news networks we find starker differences in content: CNBC and Bloomberg appear to be focused more on economics and market related content relative to Fox Business which maintains the political tilt of its news only affiliate.