Despite its current popularity and relative success, endoscopic reconstruction of the anterior cruciate ligament (ACL) using a bone-patellar tendon-bone (BPTB) graft has not yet been perfected. Using a recently developed robotic/UFS testing system, we assessed the overall stability of porcine knees following ACL reconstruction with different sites of tibial graft fixation--proximal, central, and distal. Testing of the intact knee was performed first to determine the normal anterior-posterior (A-P) displacements and in situ forces of the ACL under 110 N of anterior tibial loading of 30 degrees, 60 degrees, and 90 degrees of knee flexion. The knee was then reconstructed with a BPTB autograft, and the distal end of the graft was fixed sequentially at three different locations in each specimen--proximal, central, distal. A-P testing was repeated for each fixation site, and the resulting knee kinematics and the in situ forces of the grafts were compared to the intact case. The site of tibial fixation was demonstrated to have a significant effect on the resulting anterior displacement and internal rotation of the tibia as well as the in situ forces of the graft. Proximal fixation produced the most stable knee (A-P displacements reduced to 120% of intact at 30 degrees and 170% at 90 degrees), becoming significantly less stable with more distal fixation. These results suggest that proximal graft fixation may provide the most acute stability of the reconstructed knee.