Conventional wisdom suggests that voters are biased against women candidates for public office. Voters’ hesitation in supporting female candidates is thought to depress the number of votes women receive, causing women candidates to lose more elections than their male counterparts. Despite reasons to expect voter bias against female candidates, rarely is bias found in empirical analysis of elections. This paper moves beyond these findings of no bias and examines circumstances in which voters may be biased in favor of women candidates. Specifically, voters in low-information elections may rely on a candidate’s gender in determining their vote choice. Because of the dearth of information available about the candidates, we hypothesize that certain voters collapse their votes onto the women candidates. Thus, women candidates will do better than their male opponents in low-information elections. A test of this hypothesis, examining election returns for the California State Legislature from 2000 to 2006, finds some support for a positive bias.