The accelerating search for new types of drugs and delivery strategies poses challenge to understanding the mechanism of delivery. To this end, a detailed atomistic picture of binding between the drug and carrier is quintessential. Although many studies focus on the electrostatics of drug-vector interactions, it has also been pointed out that entropic factors relating to water and counterions can play an important role. By carrying out extensive molecular dynamics simulations and subsequently validating with experiments, we shed light herein on the binding in aqueous solution between a protein drug and polymeric carrier. We examined the complexation between the polymer poly(ethylene glycol) methyl ether acrylate-b-poly(carboxyethyl acrylate (PEGMEA-b-PCEA) and the protein egg white lysozyme, a system that acts as a model for polymer-vector/protein-drug delivery systems. The complexation has been visualized and characterized using contact maps and hydrogen bonding analyses for five independent simulations of the complex, each running over 100 ns. Binding at physiological pH is, as expected, mediated by Coulombic attraction between the positively charged protein and negatively charged carboxylate groups on the polymer. However, we find that consideration of electrostatics alone is insufficient to explain the complexation behavior at low pH. Intracomplex hydrogen bonds, van der Waals interactions, as well as water-water interactions dictate that the polymer does not release the protein at pH 4.8 or indeed at pH 3.2 even though the Coulombic attractions are largely removed as carboxylate groups on the polymer become titrated. Experiments in aqueous solution carried out at pH 7.0, 4.5, and 3.0 confirm the veracity of the computed binding behavior. Overall, these combined simulation and experimental results illustrate that coulomb interactions need to be complemented with consideration of other entropic forces, mediated by van der Waals interactions and hydrogen bonding, to search for adequate descriptors to predict binding and release properties of polymer-protein complexes. Advances in computational power over the past decade make atomistic molecular dynamics simulations such as implemented here one of the few avenues currently available to elucidate the complexity of these interactions and provide insights toward finding adequate descriptors. Thus, there remains much room for improvement of design principles for efficient capture and release delivery systems.