PURPOSE: To compare knee kinematics before and after anterior cruciate ligament ACL reconstruction (ACL-R) using hamstring graft (HG) and a double-stranded iliotibial tract graft attached to Gerdy's tubercle (providing an extra-articular anterolateral tenodesis) (named the modified iliotibial tract graft = MIT). METHOD: Eighteen cadaveric knees were tested in a 6 degree of freedom kinematics rig. An optical tracking system recorded kinematics of the knee from 0 to 80 degrees of flexion applying no load, internal/external rotation (IR/ER), valgus/varus rotation (VGR/VRR), simulated pivot shift (SPS), anterior translation (AT) and posterior translation loads. The knee was tested before and after resection of the ACL and the anterolateral ligament (ALL), respectively; then after HG-ACL-R and MIT-ACL-R. Grafts were fixed at 20° of flexion. Results were compared to the intact knee. RESULTS: ACL resection resulted in a significant increase in AT (p < 0.05) over the entire range of motion, peaking at 20° of flexion, mean difference 6.6 ± 2.25 mm (p = 0.0007). ACL-R with HG-ACL and MIT-ACL restored AT. Resection of the ALL increased IR in the fully extended knee, mean difference 2.4 ± 2.1° (p = 0.024). When compared to the intact knee and the knee after HG-ACL-R, MIT-ACL-R knee reduced IR/SPS significantly (p < 0.05) in deep flexion angles (60°-80°), peaking at 80° of flexion. The MIT-ACL-R caused significantly less VRR at 80° flexion (p = 0.02). CONCLUSION: MIT-ACL-R restored AT equally to the HG-ACL-R. The MIT-ACL-R reduced IR and SPS in deep flexion, resulting in overconstraint. MIT-ACL-R can be used as an alternative to standard reconstruction methods.