The steady maturation of computational biomechanics is providing the musculoskeletal health community with exciting avenues for enhancing orthopedic practice and rehabilitation. Computational knee models deliver tools that may improve the efficiency and outcomes of orthopedic research and methods through analysis of virtual surgeries and devices. They also provide insight into the interaction of knee structures and can predict what cannot be directly measured such as loading on our cartilage and ligaments during movement. This project created subject-specific computational knee models of two young adult females using magnetic resonance imaging-derived knee geometries and passive leg motion measured by a motion capture system. The knee models produced passive ligament lengthening patterns similar to experimental measurements available in the literature. The models also predicted cruciate ligament forces during passive flexion with and without applying anterior-posterior tibia forces that were similar to experimental measurements available in the literature. The biomechanics of the posterior oblique ligament (POL) and the anterior cruciate ligament bundles during combined tibia internal-external rotation torque and anterior-posterior forces through deep flexion were then examined. The study showed that the central arm of the POL: (1) produces a maximum constraining force when the knee is at full extension, (2) constrains internal tibial rotation at extension, and (3) constrains posterior tibial translation at extension. The POL reinforces the constraint of the anterior cruciate ligament to internal rotation at extension and provides constraint for posterior tibial translation at extension, a position where the posterior cruciate ligament provides minimal posterior translation constraint.