Malaria presents a tremendous public health burden, and new therapies are needed. Massive compound libraries screened against Plasmodium falciparum have yielded thousands of lead compounds, resulting in an acute need for rational criteria to select the best candidates for development. We reasoned that, akin to antibacterials, antimalarials might have an essential pharmacokinetic requirement for efficacy: action governed either by total exposure or peak concentration (AUC/CMAX), or by duration above a defined minimum concentration [time above minimum inhibitory concentration (TMIC)]. We devised an in vitro system for P. falciparum, capable of mimicking the dynamic fluctuations of a drug in vivo. Using this apparatus, we find that chloroquine is TMIC-dependent, whereas the efficacy of artemisinin is driven by CMAX. The latter was confirmed in a mouse model of malaria. These characteristics can explain the clinical success of two antimalarial drugs with widely different kinetics in humans. Chloroquine, which persists for weeks, is ideally suited for its TMIC mechanism, whereas great efficacy despite short exposure (t1/2 in blood 3 hours or less) is attained by CMAX-driven artemisinins. This validated preclinical model system can be used to select those antimalarial lead compounds whose CMAX or TMIC requirement for efficacy matches pharmacokinetics obtained in vivo. The apparatus can also be used to explore the kinetic dependence of other pharmacodynamic endpoints in parasites.