@Wen @RustyBertrand If you go into any journal of applied statistics, look up metascience papers and regression to the mean. It's amazing how many papers in some fields prove later to be false. Big data mining fields like genomics are by far the worst.