The active use of metrics in everyday research activities suggests academics have accepted them as standards of evaluation. Yet when asked, many academics profess concern about the limitations of evaluative metrics and the extent of their use. Why is there such a discrepancy between principle and practices?