Americans expect the president to lead Congress, but Congress’s partisan divide typically widens on presidential priorities. More often, presidents are reduced to leading their copartisans rather than Congress as a whole, but why? In this paper, I argue that competition for White House control creates incentives for parties to disagree on presidents’ policy agendas, regardless of the contents of those agendas. I use an original data set of members’ roll-call vote decisions on presidents’ agendas between 1971 and 2010 to show that partisan polarization is larger on presidents’ priorities and largest on their top priorities, above and beyond what we would expect from members’ ideologies and standard party effects. These findings persist over time and under a wide range of alternative model specifications, bringing us closer to understanding the partisan conflict so prevalent in today’s politics.