According to the extant literature, citizens who voted for a governing party are systematically more satisfied with democracy after an election than opposition party supporters. Despite being one of the most robust findings in social sciences, this "winner-loser gap" lacks causal evidence. To re-assess this claim, we make use of rare electoral contexts where it was highly uncertain which parties would form the incoming coalition government. Our research design relies on post-electoral surveys, split by the announcement of the new coalition government. We compare respondents' levels of satisfaction with democracy just before and just after the announcement of the new government. In addition, we employ random permutation tests to assess whether satisfaction with democracy is affected by the announcement of the new government in a systematic and meaningful way. Our approach yields little support for a substantial effect of winning or losing on respondents' satisfaction with democracy. Hence, the findings suggest that winning/losing is a mere proxy for a more complex mechanism driving voters’ satisfaction with democracy around election time.