How do we effectively measure poor state performance (PSP)? Determining fragile and poor performing states is an important but difficult task. This paper discusses and evaluates definitions and indexes of PSP. The author suggests that most of these exhibit poor conceptual definition, vast conceptual dispersion and confusion between causes and definitions. Moreover, they contain an intrinsic ambiguity and a lack of awareness of the problem of order. By undertaking a detailed analysis of methods of coding and ranking countries according to poor state performance, the author concludes that such indexes can only provide a useful resource for researchers if their assumptions are carefully discussed and if they can express critical aspects of the concept which they seek to codify. This paper identifies the difficulties inherent in classifying and ranking 'good' and 'bad' state performance using PSP indexes and databases. PSP models should complement sound qualitative and historical analysis to inform policy and research but, this research argues, they can only fulfill this task once they are critically assessed. Of the foremost challenges affecting the usefulness and accuracy of PSP databases, the author highlights hidden assumptions (revealed through detailed analysis of the 'numeraire' - a standard by which values are measured), definitional inconsistencies and intrinsic ambiguity. In order to avoid the political and economic consequences presented by an over-reliance on inaccurate rankings and to develop better systems of classifying state performance, open acknowledgement of the assumptions and limits of these models is essential.