Lovering et al. (2016) present data on the overnight costs of more than half of nuclear reactors built worldwide since the beginning of the nuclear age. The authors claim that this consolidated data set offers more accurate insights than previous country-level assessments. Unfortunately, the authors make analytical choices that mask nuclear power's real construction costs, cherry pick data, and include misleading data on early experimental and demonstration reactors. For those reasons, serious students of such issues should look elsewhere for guidance about understanding the true costs of nuclear power. (C) 2016 Elsevier Ltd. All rights reserved.