Bored piles with constrained grouting expansion (BPCGE) are an innovative approach for piling that integrates isotropic high-pressure grouting with constrained expansion techniques. This method is recognized for its robust compressive load-bearing capabilities and superior pile quality. However, the presence of an expansion body complicates the understanding of vertical compressive pile-soil interactions, particularly those associated with the load-bearing capacity and failure mechanisms. Hence, a visual semi-model pile testing schema is developed in this study. Experimental and finite-element analyses are conducted to investigate the bearing capacity and failure mechanisms of BPCGE. Initially, a visual static load test is conducted on a semi-modeled pile based on the principle of similarity to discern the load-bearing characteristics of the pile. Digital image correlation is used to measure the displacement and deformation patterns of soil around the pile. An expansion correction coefficient for the bottom resistance is introduced following the analysis. Subsequently, numerical simulations are performed using the ABAQUS software, where the modified Cam-Clay model is utilized to simulate the compressive static-load test of BPCGE under various size parameters. The effects of these parameters on the ultimate bearing capacity and the pile’s end-bearing capacity are investigated, which are then considered to further refine the expansion correction coefficient for the bottom resistance. Finally, a formula for calculating the bearing is established, and the calculated values are compared with data from field static-load tests. The findings indicate that BPCGE can be classified as end-bearing friction piles, with the shear failure of the pile tip soil being the primary cause of failure. Meanwhile, the expansion correction coefficient for the bottom resistance should be set from 1.92 to 2.30, based on practical project considerations. By selecting an appropriate coefficient based on the size parameters of BPCGE, the pile end resistance can be adjusted to account for combined end resistance. The ultimate bearing capacity calculated using the proposed formula is consistent with actual measurements, with an error margin of 9.3%, as compared with the field test results. This study provides theoretical insights and practical guidelines for the design and implementation of BPCGE.