Voter turnout is a key measure of the legitimacy of any given democracy. A government elected in a high-turnout contest has a strong mandate from the state's citizens to make key decisions on their behalf. Low turnout raises questions about why citizens have chosen not to participate in the decision-making process, and about the representativeness of the elected government. Often, changes in levels of engagement and in the geography of participation are relatively small, but they may be decisive in key competitive elections. Using the state of Michigan as a case study, this brief commentary considers the changing geographies of participation in U.S. Presidential elections from 2012 to 2020. In Michigan, a notable voter-turnout upsurge in what traditionally had been low-turnout rural areas had a significant bearing on the victory of Donald Trump in 2016. This upsurge in rural support for Trump continued in the 2020 contest but was counteracted by a more positive (though not overwhelming) level of turnout in urban centres. After explaining these trends, this commentary briefly compares the case of Michigan with the Republic of Ireland in general elections from 2007 to 2016 to highlight the importance of urban-rural distinctions in competitive electoral settings.