But my standard deviation of reviewing time is high. I review a fair number of papers that fail the "Can the methods possibly answer the research questions?" test. Those I write short reviews, offering ideas for what questions their data can answer, or how to design a study to answer their questions