Evaluativism about desire, the view that desires just are, or necessarily involve, positive evaluations of their objects, currently enjoys widespread popularity in many philosophical circles. This essay argues that evaluativism, in both its doxastic and its perceptual versions, overstates and mischaracterizes the connection between desires and evaluations. Whereas doxastic evaluativism implausibly rules out cases where someone has a desire, despite evaluating its object negatively, being uncertain about its value, or having no doxastic attitude whatsoever toward its evaluative status at all, perceptual evaluativism cannot even properly apply to the large class of standing desires. It is also argued that evaluativism about desire is not even well-motivated in the first place: the theory is supposed to solve a particular puzzle about the role desires play in the explanation of action, yet in fact it does not offer any help whatsoever in dealing with the relevant puzzle.