Use of metal-selective ligands in solvent extraction is instrumental in extraction of critical materials and recycling, yet, diluent effects on extraction performance are not well understood. Experimental and empirical solvent parameters have been proposed to correlate with extraction performance, but are often inadequate predictors. We follow the hypothesis that the diluents' primary influence on extraction efficiency is whether or not it hinders assembly of the bulky extracting ligands into a geometry necessary for metal complexation. This behavior is readily accessible with molecular dynamics (MD), where the atomistic description of molecules can be applied to arbitrary extractant-solvent molecules and their mixtures. Several simulated quantities are considered, from both pairwise and graph theoretical analyses, and compared to experimental distribution ratio data for americium extraction by TODGA in a series of inert, non-interacting diluents. These simple properties, especially the formation of closed triplets corresponding to the 3 : 1 ligand : metal stoichiometric solvate, suggest a potential predictive power of this approach. This methodology provides a path forward to comprehensively understand and predict diluent effects in more complex systems involving different extracting ligands and multi-component diluent mixtures.