The cardiac hERG K channel constitutes a long-standing and expensive antitarget for the drug industry. From a study of the flexibility of hERG around its internal binding cavity, we have developed a new structural model of drug binding to hERG, which involves binding orthogonal to the pore channel and therefore can exploit the up to 4-fold symmetry of the tetrameric channel. This binding site has a base formed by four tyrosine side chains that complement reported ligand-based pharmacophores. The model is able to rationalize reduced hERG potency in matched molecular pair studies and suggests design guidelines to optimize against hERG not relying simply on lipophilicity reduction. The binding model also suggests a molecular mechanism for the link between high-affinity hERG binding and C-type inactivation.