The redshifted 21-cm line of distant neutral H atoms provides a probe of the cosmic "dark ages" and the epoch of reionization ("EOR") which ended them, within the first billion years of cosmic time. The radio continuum produced by this redshifted line can be seen in absorption or emission against the cosmic microwave background ("CMB") at meterwaves, yielding information about the thermal and ionization history of the universe and the primordial density perturbation spectrum that led to galaxy and large-scale structure formation. Observing this 21-cm background is a great challenge, as it is necessary to detect a diffuse signal at a brightness temperature that differs from that of the CMB at millikelvin levels and distinguish this from foreground continuum sources. A new generation of low-frequency radio arrays is currently under development to search for this background. Accurate theoretical predictions of the spectrum and anisotropy of this background, necessary to guide and interpret future observations, are also quite challenging. Toward this end, it is necessary to model the inhomogeneous reionization of the intergalactic medium and determine the spin temperature of the 21-cm transition and its variations in time and space as it decouples from the temperature of the CMB. In my talk, I summarized some of the theoretical progress in this area. Here, I will focus on just a few of the predictions for the 21-cm background from the EOR, based on our newest, large-scale simulations of patchy reionization. These simulations are the first with enough N-body particles (from 5 to 29 billion) and radiative transfer rays to resolve the formation of and trace the ionizing radiation from each of the millions of dwarf galaxies believed responsible for reionization, down to 108 M[sun], in a cubic volume large enough (90 and 163 comoving Mpc on a side) to make meaningful statistical predictions of the fluctuating 21-cm background.