To compare the graft length change patterns in nonanatomic adductor-transfer (AT) and adductor-sling (AS) medial patellofemoral ligament (MPFL) reconstruction with those in anatomic MPFL reconstruction (MPFLR) and to investigate the favorable isometric ranges (FIRs) of knee flexion. Eight small fresh-frozen cadaveric knees were mounted in a knee-customized jig with tensioned muscles to measure graft length changes from two patellar points to four femoral attachments using a linear variable displacement transducer. Femoral attachments were at the MPFL footprint center (MPFL-C) in anatomic MPFLR, adductor magnus (AM) footprint center (AM-C) in AT, and at 5-mm (AM-5) or 10-mm (AM-10) points proximal to AM-C in AS. The FIRs of these femoral attachments were determined after zeroing the graft length changes at different initial fixation angles. Anatomic MPFL-C resulted in almost isometric graft length changes from 0° to 90°. At AM-C, the graft length changes displayed an increase from 0° to 45° and significantly greater length changes than those at MPFL-C from 60° to 90°. The graft length changes at both AM-5 and at AM-10 continuously increased with knee flexion angles and presented significantly larger graft length changes than those at MPFL-C and at AM-C from 30° to 90° and 60° to 90°, respectively. After zeroing graft length changes at different fixation angles, the FIRs of the MPFL-C covered all knee flexion angles, regardless of the patellar attachments and initial fixation angles. Moreover, with the smaller FIRs of AM-C observed at any fixation angle when compared with MPFL-C, fixing the graft at 0° to 30° in AT allowed the FIRs to cover the whole functional flexion range. However, the significantly larger graft length changes of both AM-5 and AM-10 resulted in extremely limited FIRs at any fixation angle. Anatomic MPFL-C resulted in a favorable graft length change range (less than 2 mm) at 0° to 90° of knee flexion, which was close to the isometric graft behavior. However, nonanatomic attachments of AM-C in AT, and both AM-5 and AM-10 in AS caused significant anisometric graft length change patterns and limited FIRs. Moreover, AT had a smaller range of graft length change but wider FIRs compared to AS, indicating superior graft behavior for MPFLR. Anatomic MPFLR is preferable for the treatment of in skeletally immature patients, followed in preference by nonanatomic AT. Nonanatomic AS should be performed cautiously.