This seems to be confusing proportions with absolute numbers. Newspapers do this all the time about health statistics, going from a very small number of additio

This seems to be confusing proportions with absolute numbers. Newspapers do this all the time about health statistics, going from a very small number of additional deaths to scaremonger about a '300% increase in cancer'. Terrorism in the last year is a significant proportion of the total number of homicides in France, but the percentage increase doesn't mean the absolute number is relevant. If you didn't worry about being murdered in France before 2015, there's no reason to worry about dying in a terrorist attack after 2015. Moreover, we might expect terrorist attacks to regress to the mean, so 2015 would be unusually bad under that additional assumption. If the number of homicides increased by 14-20% due to some other cause, it likely wouldn't be seen as an urgent policy issue. > Those 600,000 people are going to die either way In fact, many will die from avoidable causes, like medical errors.