Research on citizens’ reactions to incumbent opportunism demonstrates that citizens punish incumbents who time elections strategically. However, we lack evidence of how citizens react to early election announcements in real-world settings. We address this gap by studying the 2018 snap election announcement in Turkey. Since the election outcome was not known at that time, this setting allows us to investigate how the Turkish public responds to incumbent opportunism irrespective of belonging to the winners or losers of the electoral contest. We theorize that the announcement of early elections unleashes enthusiasm about the country’s state of democracy, as citizens expect to influence politics in the near future. We exploit a regression discontinuity design (RDD) using data from the 2018 World Values Survey (WVS) to test the causal effect of the early election announcement on citizens’ attitudes on elections and democracy. We thereby introduce a new focus on effect heterogeneity in the context of natural experiments to disentangle the effect between partisans. We find a positive effect on citizens’ evaluation of whether elections offer voters a genuine choice in the whole sample that is mainly driven by partisans of the incumbent parties. Surprisingly, despite the overall positive reactions of opposition leaders to the election announcement, we do not find similarly positive effects among opposition partisans. Our results imply that, even though strategically called, early elections enhance confidence in elections only among incumbent supporters, while opposition supports neither share nor oppose this enthusiasm. This calls into question the potential role of citizens to serve as a check on opportunistic incumbent behavior.