In this paper, I will discuss comparativism, the view that comparative beliefs (e.g., being more confident that P than that Q) are more fundamental than partial beliefs (e.g., being x% confident that P). There are three parts to the paper. The first part is sympathetic: using an idea from Ramsey, I show how comparativists can overcome one major challenge for their view --- viz., how they can keep enough structure in the comparative beliefs to make sense of cardinal information without thereby presupposing an unrealistic degree of rationality. In the second part, I argue that even the most plausible versions of comparativism have severely limited resources with which to accommodate contemporary modes of descriptive and normative theorising about partial beliefs, and face several challenges on that front. And finally, in the third part, I argue that there are plausible non-comparativist accounts of what partial beliefs are and how they are measured which handle the same challenges much better.