Purpose: Cartilage damage around the acetabular rim is common in individuals with hip pathology. Redirecting hip loading away from commonly damaged areas of cartilage to non-degenerated areas of cartilage may be desirable for individuals with hip pathology. The deep hip muscles are a group of small muscles that may be best suited to make small changes to the direction of hip loading. These muscles have limited torque generating capacity, suggesting changes in the direction of hip loading could be achieved without substantially affecting hip motion. The aim of this study was to evaluate whether the deep hip muscles could redirect the orientation of hip loading towards the centre of the acetabulum - away from commonly damaged areas of cartilage. Methods: Three-dimensional marker trajectories, ground reaction forces, and electromyography (EMG) from 12 superficial and 4 deep (gluteus medius, piriformis, obturator internus, and quadratus femoris) muscles of the lower limb were recorded during walking in 12 participants with no history of hip pain (age 24±4 yrs, BMI 23 kg/m2, 33% female). A generic musculoskeletal model was linearly scaled to each participant’s dimension and used to perform inverse kinematics, inverse dynamics, and muscle analysis in OpenSim. Muscle-tendon parameters were calibrated to the individual before calculating muscle forces via an EMG-informed neuromusculoskeletal model. Muscle forces were calculated using three model configurations that had different deep hip muscle excitations, but identical excitations for all other lower limb muscles: (i) deep hip muscles (gemelli, obturator externus and internus, piriformis, quadratus femoris) informed by EMG measurements (assisted activation), and (ii) zero (zero activation) and (iii) maximal activation (maximal activation) imposed onto the deep hip muscles, respectively. Muscle forces were used within the joint reaction toolbox in OpenSim to calculate hip contact forces (HCF). The loading pattern in the acetabulum during the gait cycle was calculated as the intersection of the hip contact force and a sphere fitted to the acetabulum. Additionally, the orientation of hip loading in the acetabulum was quantified as the angle between the HCF and the vector from the centre of the femoral head to the centre of the acetabulum (HCF angle). The orientation of the acetabular rim was calculated as the angle between the vector from the centre of the femoral head to the acetabular rim and the vector from the centre of the femoral head to the centre of the acetabulum (AR angle). A smaller HCF angle equated to hip loading towards the centre of the acetabulum, whereas a HCF angle closer to the AR angle equated to loading closer to the acetabular rim. The HCF angle was compared over the gait cycle between models using a withinsubject analysis of variance with statistical parametric mapping methods (P<0.05). If main effects for model configurations were found, SPM paired sample t-tests with Bonferroni correction were performed. Results: Maximal activation imposed onto the deep hip muscles changed the location of acetabular loading to be more anterior and inferior during walking compared to both assisted and zero activation (Figure 1A). The HCF angle differed significantly over the gait cycle between models. Post-hoc analysis revealed HCF angle was smaller for maximal activation compared to both assisted and zero activation models over most of the gait cycle (Figure 1B). HCF angle was larger for zero activation compared to assisted activation during the early stance, and mid swing. Conclusions: Maximal activation of the deep hip muscles shifted the orientation of hip loading towards the centre of the acetabulum, away from commonly damaged areas of the cartilage. The deep hip muscles may be able to prevent edge loading and/or play a role in load distribution across the articular surfaces. Targeted training of these muscles may be relevant for people with hip pathology