How well do voters hold politicians accountable? Although a long-standing research tradition claims that elections are effective tools for the sanctioning and selection of leaders, a more recent literature argues that voters often reward and punish incumbents for "irrelevant events." The empirical literature on this topic is characterized by conflicting findings. Drawing on ideas from the open science movement and showing how they can advance the transparency of observational research, we replicated three prominent studies on irrelevant events and voting behavior: (1) Christopher Achen and Larry Bartels's 2016 study of droughts and floods; (2) Andrew Healy, Neil Malhotra, and Cecilia Hyunjung Mo's 2010 study of college football; and (3) Andrew Healy and Neil Malhotra's 2010 study of tornadoes. Each study replicates well in some areas and poorly in others. Had we sought to debunk any of the three with ex post specification searches, we could have done so. However, our approach required us to see the full, complicated picture.
