Random selection for political office-or "sortition"-is increasingly seen as a promising tool for democratic renewal. Critics worry, however, that replacing elected and appointed officials with randomly selected citizens would only exacerbate elite manipulation of political processes. This article argues that sortition can contribute to democratic renewal, but that its genuine promise is obscured by the excessive ambition and misplaced focus of prevailing models. Casting random selection as a route to accurate representation of the popular will, most contemporary proposals require randomly selected citizens to perform legislative tasks, whose open-endedness grants substantial discretion to elite agenda setters and facilitators. The real democratic promise of sortition-based reforms, I argue, lies in obstructing elite capture at critical junctures: a narrower task of oversight that creates fewer opportunities for elite manipulation. In such contexts, the benefits of empowering ordinary people-resulting from their immunity to certain distorting influences on career officials-plausibly outweigh the risks.A mid declining faith in familiar democratic institutions, the premodern practice of random selection for public office-or "sortition"-has attracted growing attention from scholars and reformers, as a powerful corrective for elite domination of political processes (Gastil and Wright 2019; Guerrero 2014; Landemore 2020). Yet others doubt sortition-based reforms could effectively confront contemporary democratic crises (e.g., Lafont 2019), and some suspect they would make matters worse (Landa and Pevnick 2021; Umbers 2021). Weighing concerns on all sides, this article argues that sortition does have a unique and important role to play in enhancing democracy, but that its potential is obscured by the excessively ambitious models favored by contemporary proponents. It therefore advances a revisionist account of the real but limited promise of lotteries as a tool for democratic reform: sortition as anti-corruption.As advocates of sortition have long pointed out, conventional methods for choosing public officials exhibit several pervasive pathologies that could be mitigated with random selection. First, wealthy elites and other powerful private actors can often ensure that officials chosen through election or appointment are sympathetic to their interests. Second, career officials are susceptible to certain distinctive incentives, which wealthy and powerful interests can use to pressure them after they are selected. Third, the class of political elites from which officials are drawn also exhibits pervasive biases and blind spots. By insulating policy processes from all three forms of capture, random selection of officials can substantially advance democracy.Given the risks of entrusting key political decisions entirely to ordinary citizens with little experience or expertise, however, we must also be careful not to overstate the democratic potential of sortition. As I demonstrate, more specifically, politically inexperien...