Reading the paper [1] they used the period 1950-2012. All Atlantic hurricanes had female names until 1979, however, which means that if older hurricanes were m

Reading the paper [1] they used the period 1950-2012. All Atlantic hurricanes had female names until 1979, however, which means that if older hurricanes were more deadly then it will look like female-named hurricanes were more deadly. And it does look likely that older hurricanes were more deadly. Looking at the deadliest 1950-2012 North Atlantic hurricanes in decreasing order of (international) death toll [2], we have Orlene (1974), Flora (1963), David (1979), Gordon (1994), and Janet (1955). You can also look at the paper's dataset [3] and see that three hurricanes killed >100 Americans before 1979, and only one after. Restricting their dataset to 1979+ (the period when both male and female names were used) I count 420 female-hurricane deaths and 452 male-hurricane deaths. While this decrease in deaths may be related to their naming, I would guess it has more to do with worse forecasting and lower building standards. And will likely reverse over time as storms become more intense and more people live in hurricane-prone areas. [1] https://www.pnas.org/doi/10.1073/pnas.1402786111 https://www.pnas.org/doi/10.1073/pnas.1402786111 [2] https://en.wikipedia.org/wiki/List_of_the_deadliest_tropical_cyclones https://en.wikipedia.org/wiki/List_of_the_deadliest_tropical... [3] https://www.pnas.org/doi/suppl/10.1073/pnas.1402786111/suppl_file/pnas.1402786111.sd01.xlsx https://www.pnas.org/doi/suppl/10.1073/pnas.1402786111/suppl...