Humean reductionism about laws of nature holds that the laws reduce to fundamental nonmodal
facts. The preferred Humean account of this reduction, the best systems account
(BSA), identifies laws with axioms of the best systematizations of actual facts (cf. Beebee
2000; Lewis 1973, 1983; Loewer 1996). A recent challenge for the BSA is that it arguably
cannot motivate why science aims for laws that apply to a wide range of counterfactual
situations (Hall 2012, 2015). If laws are part of the best systematizations of fundamentally
non-modal facts, as proponents of the BSA hold, why are the actual laws geared toward
describing what would happen in counterfactual situations? In this paper, we defend a new
version of the BSA that meets this challenge. We argue that best systems need to be best for
us, that is, useful for creatures with limited epistemic capacities. This version of the best
system is both motivated on Humean grounds and accords with scientific practice. In
particular, we show that the modal content of the laws is a byproduct of their practical
usefulness for creatures like us