Stem design is a crucial element for the success of shoulder prostheses. Various components of stem design have been investigated;​ however, little research has been conducted on the effects of (1) implant girth and (2) an implant collar on load transfer. A generic implant model was designed and employed in a (FE) model to determine 3 outcome measures:​ changes in the degree of bone to implant contact (BIC), changes in cortical and trabecular bone stresses from the intact state, and changes in cortical and trabecular strain energy densities (SED). The variables examined were (1) implant girth (small, medium [generic base model], and large sizes), and (2) the implant collar and collarless. The small implant produced the overall greatest amount of BIC when compared to the other two sizes. The small implant also produced the lowest change in stress from the intact state in both cortical and trabecular bone, as well as the lowest amount of bone volume expected to resorb. Removing the implant collar caused an increase in the degree of BIC, when compared to the collared state. In terms of the changes in stress, removing the implant collar resulted in an increase in both the change in cortical and trabecular bone stresses, and resulted in an increased risk in the amount of cortical bone expected to resorb. Collectively, these findings suggest that a smaller sized implant may be beneficial, while the collar may be beneficial if less stress changes in bone relative to the native state are desired.