Background Normal knee kinematics is characterized by posterior femorotibial rollback with tibial internal rotation and medial-pivot rotation in flexion. Cruciate-retaining TKAs (CR-TKAs) do not reproduce normal knee kinematics. Questions/purposes We hypothesized a more anatomic reconstruction of the medial femoral condyle, simultaneously preserving the tension of the PCL and medial collateral ligament, resulted in (1) medial-pivot rotation and tibial internal rotation, (2) lateral femoral rollback, and (3) reduced liftoff. Patients and Methods We compared 10 patients who underwent CR-TKA using the new technique at their 1-year followup to a matched control group of nine patients using a traditional gap-balancing technique at their 2-to 4-year followup. All patients received lateral radiographs in extension and flexion, which we utilized for threedimensional implant matching to calculate tibial internal rotation, lateral rollback, and lateral liftoff in extension and flexion. Results The new gap-balancing technique resulted in a median of 3.5°tibial internal rotation with 2.7-mm rollback of the lateral femoral condyle relative to the medial condyle in flexion, which was different from the control group. We found no differences in liftoff between the groups. Conclusions The new technique resulted in tibial internal rotation with flexion and lateral rollback comparing the lateral to the medial condyle in flexion, but no differences in condylar liftoff. These preliminary results were comparable to published kinematic results of an asymmetric CR-TKA or medial-pivot CR-TKA but not to symmetric CR-TKA.