A recent article in Nature examines the gender imbalance in scientific authorship, by building a probability model of how likely women are to fail being associated with the authorship of a scientific paper to which they contributed. I did not read the paper in detail (while on a train) and thus cannot comment about the scientific basis of the model but am surprised at the specific case of Mathematics, where the 50% share in the above graph seems to conflict with the strong (and sad) gender imbalance in the current faculty composition for the field. The analysis is based on US data from (a) a self-reporting survey, (b) administrative data, where records are available about “every payment that is made during each pay period from each grant and provide information on each employee’s job title”, which only covers US research work covered by grant money, and (c) percentage of women against paper citation index. For (b), it is unclear to me who would qualify as research staff in a theoretical math research project.