There is an ongoing debate about bilateral investment treaties (BITs) – and investor-state arbitration, in particular – between those who maintain that BITs encourage investment in developing countries by providing enforceable rights and protections for investors, and those who suspect that these new rights and protections have a chilling effect on regulation for public and environmental welfare and actually hinder development. For years, both ?camps? have drawn heavily upon anecdotal evidence and observations to support their view, as no systematic, comprehensive study of empirical data on investment arbitrations had been undertaken. To fill this void, legal scholar Susan Franck has evaluated the criticisms of investment arbitration based on empirical studies of published or known disputes (Franck 2009; Franck 2007). These efforts produced helpful data and initiated a productive discussion of these issues. However, the results and conclusions that can be drawn from Franck’s work are more limited and warrant more nuance than Franck and others so far have taken into account. Franck’s work is now widely used to support the notion that developing countries do not disproportionately ?lose? under the investment arbitration regime. Such a conclusion does not appear to be supported by Franck’s data. This article analyzes Franck’s work to show where differing conclusions emerge. We show that: 1) there is a lack of adequate sample composition and size to conduct rigorous empirical work from which an analyst could draw such bold lessons; 2) discounting the fact that developing countries are subject to a disproportionate number of claims is not to be overlooked, especially when looking at claims by the United States; and 3) relative to government budgets and in per capita terms developing countries pay significantly more in damages than developed nations do.