Journalistic crisis responses have been object of critical research, based on the analyses of wartime and terror reporting, as well as the coverage of natural disasters. This research notes that news media consistently fail to adequately cover political action during such crises, focusing too much on a limited set of political elite actors. Following this line of thought, the current analysis is interested in identifying general patterns of political reporting in the first nine months of the coronavirus crisis. In a large-scale computational content analysis of news media’s Facebook messages in Germany, we apply named entity recognition and network analysis in order to identify political actors: how they were connected to specific topics in the coverage, and how this has changed during the various phases of the pandemic. The analysis reveals a focus on governmental elite actors and a limited set of experts, while the parliamentary opposition did not receive much attention. In contrast, conspiracy theorists and some foreign actors were covered prominently. However, this focus was not uniform throughout the year, and in a later phase of the pandemic, the analysis reveals a “normalization” with a less reduced set of individualized political actors.