The mismatch between bony endplates (BEPs) and grafted bone (GB) triggers several complications biomechanically. However, no published study has identified whether this factor increases the risk of screw loosening by deteriorating the local stress levels. This study aimed to illustrate the biomechanical effects of the mismatch between BEP and GB and the related risk of screw loosening. In this study, radiographic and demographic data of 56 patients treated by single segment oblique lumbar interbody fusion (OLIF) with anterior lateral single rod (ALSR) fixation were collected retrospectively, and the match sufficiency between BEP and GB was measured and presented as the grafted bony occupancy rate (GBOR). Data in patients with and without screw loosening were compared; regression analyses identified independent risk factors. OLIF with different GBORs was simulated in a previously constructed and validated lumbosacral model, and biomechanical indicators related to screw loosening were computed in surgical models. The radiographic review and numerical simulations showed that the coronal plane’s GBOR was significantly lower in screw loosening patients both in the cranial and caudal vertebral bodies; the decrease in the coronal plane’s GBOR has been proven to be an independent risk factor for screw loosening. In addition, numerical mechanical simulations showed that the poor match between BEP and GB will lead to stress concentration on both screws and bone-screw interfaces. Therefore, we can conclude that the mismatch between the BEP and GB will increase the risk of screw loosening by deteriorating local stress levels, and the increase in the GBOR by modifying the OLIF cage’s design may be an effective method to optimize the patient’s prognosis.