Prosecutors under common‐law tradition exercise a significant amount of discretion in the criminal justice system. In the United States, the dominant form of accountability is that prosecutors must be reelected by the voters. Recent empirical work illustrates that election concerns open up the potential for distortion in the decision making of prosecutors. Specifically, it has been shown that prosecutors take more cases to trial and plea bargain less when running for reelection. Does this hawkish behavior of prosecutors lead to inaccuracies in the criminal justice system? A panel data set of appellate decisions in western New York is analyzed. It is shown that if the initial felony conviction takes place in the six months prior to a reelection and is appealed, the probability that the appellate court upholds the lower court's decision decreases by 5.1–7.1 percentage points. Additional investigation into the types of mistakes made is done. Thus, the popular election of prosecutors results in inaccurate sentences, wrongful convictions, and, consequently, successful appeals.