Deliberative Polling is a promising model for fostering public participation in political deliberative processes. However, its specific design involves tradeoffs among various objectives and values. Thus, some of its intentionally created “features” can simultaneously constitute a shortcoming or a “bug.” My critical assessment of Deliberative Polling begins by highlighting some limitations of this kind—those that arise from the goals of the model itself. Insofar as this is the case, these limitations should not be seen as “net negative” features that ought to be removed; indeed, eliminating them would jeopardize much of the value found within this model of public participation. This leads me to focus on what I take to be the most serious shortcoming of Deliberative polling, namely, the disconnect between its participants and the nonparticipating public. In contrast to the limitations analyzed in the first section, I argue that this specific shortcoming needs to be squarely addressed and eliminated because it threatens to undermine the legitimacy of using Deliberative Polling in political decision-making processes. In a last step, I discuss some institutional features and mechanisms that could be combined with Deliberative Polling to strengthen its connection with the broader public and to thereby improve its legitimacy.