This paper argues that the voting power approach is much more general than is portrayed by Albert (“The Voting Power Approach: Measurement without Theory”, European Union Politics, 4:3, 2003) and is therefore capable of generating important insights about voting systems, such as qualified majority voting in the EU Council. The voting power approach focuses on understanding the properties of voting systems by analysing outcomes and thereby is able to generate empirical facts that are not otherwise obvious. That the approach is so general has not previously been pointed out in the relevant literature; it has usually been taken as coinciding with power indices. Albert’s criticism is directed at one aspect of the theory of voting power indices: the assumption of probabilistic voting that underlies conventional power indices. It is argued that he fails to take account of the different uses of power indices and that the probabilistic voting assumptions he derides may or may not be useful depending on this. It is necessary to emphasise the key distinction between a priori power indices and measure of empirical voting power. Albert misrepresents the voting power approach and is not willing to allow that it encompasses a diversity of methods as well as a research agenda.