A ubiquitous experiment to characterize the formability of sheet metal is the standard uniaxial tension test. Past research [1-3] has shown that if the material is repeatedly bent and unbent during this test (termed Continuous-Bending-under-Tension, or CBT), the percent elongation at failure increases significantly (e.g., from 22% to 290% for an AISI 1006 steel ). However, past experiments have been conducted with a fixed stroke of the CBT device, which limits the formability improvements. This phenomenon has also been empirically observed in industry; the failure strains of a sheet which is passed through a drawbead (i.e., that has been bent and unbent three times before entering the die) are higher than those of the original sheet. Thus, the residual formability of the material after a specified number of CBT passes is of interest, to determine if multiple drawbeads would be beneficial in the process. Also of interest is the localization of the deformation during the process as this will provide a better physical understanding of the improved formability observed. In this paper, numerical simulations are presented to assess these effects. Results show that the formability during CBT is dictated by the uniaxial response of the material until the standard elongation at failure is exceeded. This limit can be exceeded by the CBT process. However, failure then occurs as soon as the CBT process is terminated. Also, the deformation is more uniformly distributed over the entire gauge length during the CBT process which leads to the increased elongations observed.