Global, three-dimensional, ideal MHD simulations of Earth's bow shock are reported for low Alfven Mach numbers M(A) and quasi-perpendicular magnetic field orientations. The simulations use a hard, infinitely conducting magnetopause obstacle, with axisymmetric three-dimensional location given by a scaled standard model, to directly address previous gasdynamic (GD) and field-aligned MHD (FA-MHD) work. Tests of the simulated shocks' density jumps X for 1.4 less than or similar to M(A) less than or similar to 10 and the high M(A) shock location, and reproduction of the GD relation between magnetosheath thickness and X for quasi-gasdynamic MHD runs with M(A) his, confirm that the MHD code is working correctly. The MHD simulations show the standoff distance a(s) increasing monotonically with decreasing M(A). Significantly larger a, are found at low MA than predicted by GD and phenomenological MHD models and FA-MHD simulations, as required qualitatively by observations. The GD and FA-MHD predictions err qualitatively, predicting either constant or decreasing a(s) with decreasing M(A). This qualitative difference between quasiperpendicular MHD and FA-MHD simulations is direct evidence for a(s) depending on the magnetic field orientation theta. The enhancement factor over the phenomenological MHD predictions at M(A) similar to 2.4 agrees quantitatively with one observational estimate. A linear relationship is found between the magnetosheath thickness and X, modified both quantitatively and intrinsically by MHD effects from the GD result. The MHD and GD results agree in the high M(A) limit. An MHD theory is developed for a(s), restricted to sufficiently perpendicular theta and high sonic Mach numbers M(S). It explains the simulation results with excellent accuracy. Observational and further simulation testing of this MHD theory, and of its predicted M(A), theta, and M(S) effects, is desirable.