The laws of nature have a special relation to counterfactuals that distinguishes them from non-laws. For example, If the Earth’s axis had been perpendicular to its orbital plane, many things would be different (for example, there would have been no seasons). But we think that the laws of nature still would be the same. Explaining this characteristic counterfactual invariance is challenging for Humean theories of laws. Humeans hold that laws of nature are mere regularities in the distribution of the non-modal properties. However, if the laws are mere regularities, why should they have any special modal status with regard to counterfactuals? Dorst (2022), and Loew and Jaag (2020) have recently offered pragmatic explanations for the counterfactual invariance of laws: We use counterfactuals for prediction and confirmation, and they serve these evidential purposes best if we hold the laws of nature fixed. Lange (forthcoming) criticizes these accounts by arguing that counterfactuals do not in fact play the relevant evidential roles. In this talk, we argue that Lange’s criticism leaves the broader Humean strategy intact by clarifying the evidential role of counterfactuals. Moreover, we argue that Humeans can consistently hold that laws are invariant under counterfactual variations despite lacking inherent necessity.