Nanoporous beta-MnO2 can act as a host lattice for the insertion and deinsertion of Li with application in rechargeable lithium batteries. We predict that, to maximize its electrochemical properties, the beta-MnO2 host should be symmetrically porous and heavily twinned. In addition, we predict that there exists a "critical (wall) thickness" for MnO2 nanomaterials above which the strain associated with Li insertion is accommodated via a plastic, rather than elastic, deformation of the host lattice leading to property fading upon cycling. We predict that this critical thickness lies between 10 and 100 nm for beta-MnO2 and is greater than 100 nm for alpha-MnO2: the latter accommodates 2 x 2 tunnels compared with the smaller 1 x 1 tunnels found in beta-MnO2. This prediction may help explain why certain (nano)forms of MnO2 are electrochemically active, while others are not. Our predictions are based upon atomistic models of beta-MnO2 nanomaterials. In particular, a systematic strategy, analogous to methods widely and routinely used to model crystal structure, was used to generate the nanostructures. Specifically, the (space) symmetry associated with the nanostructure coupled with basis nanoparticles was used to prescribe full atomistic models of nanoparticles (0D), nanorods (1D), nanosheets (2D), and nanoporous (3D) architectures. For the latter, under MID simulation, the amorphous nanoparticles agglomerate together with their periodic neighbors to formulate the walls of the nanomaterial; the particular polymorphic structure was evolved using simulated amorphization and crystallization. We show that our atomistic models are in accord with experiment. Our models reveal that the periodic framework architecture, together with microtwinning, enables insertion of Li anywhere on the (internal) surface and facilitates Li transport in all three spatial directions within the host lattice. Accordingly, the symmetrically porous MnO2 can expand and contract linearly and crucially elastically under charge/discharge, We also suggest tentatively that our predictions for MnO2 are more general in that similar arguments may apply to other nanomaterials, which might expand and contract elastically upon charging/discharging.