In the present paper, I reexamine the sample-survey methodology for estimating war deaths and consider the future of the field. Section 2 gives a critical overview of this approach. In section 3 I work through case studies from Kosovo, Darfur, the DRC and Iraq. I draw some conclusions and look to the future in section. The present paper has a number of limitations due largely to space constraints. First, the only human cost I consider is death, leaving out such important costs as injury, displacement and rape. The existing literature focuses strongly on death, perhaps because this is the most dramatic human cost of war. However, the conflict field should allocate more effort into measuring other human costs in the future.