In many machine learning problems, the performance of algorithms is evaluated using the emph accuracy and emph callback measures. These two measures can have a very different significance depending on the context. In this paper we study the behavior of various performance indices based on the precision-recall trade-off. We propose a new performance visualization tool that is the space trade-off representing different indices based on the precisionrecall trade-off. We analyze the properties of this new space and highlight its advantages over the precision-recall space.