Arguments that we have too much litigation (overclaiming) or too little (underclaiming) cannot be valid without estimating how many of the undecided claims that are brought (actual claims) or not brought (potential claims) have or lack legal merit. We identify the basic conceptual structure of such underclaiming and overclaiming arguments, which entails inferences about the distribution of actual or potential claims by their probability of success on the merits within a claims-processing institution. We then survey the available methods for estimating claim merit.