This study evaluates the outcomes of multilevel soft tissue surgery in 31 ambulatory children (n = 39 sides) with cerebral palsy. All children had undergone rectus femoris transfer, hamstring lengthening, and gastrosoleus lengthening for the purpose of correcting sagittal plane abnormalities. There were no simultaneous bony surgeries. Preoperative and postoperative evaluation consisted of clinical assessment and gait analysis, including 3-dimensional kinematics and kinetics. Results demonstrated improvements in knee and ankle function. At the knee, there was a decrease in mean flexion at initial contact (from 31 degrees [SD, +/-8 degrees] to 21 degrees [SD, +/-10 degrees]) and in stance (mean stance, 22 degrees [SD, +/-12 degrees] to 16 degrees [SD, +/-11 degrees]) associated with a decreased mean internal extensor moment in stance (from 0.09 Nm/kg [SD, +/-0.24 Nm/kg] to -0.03 [SD, +/-0.22 Nm/kg]). At the same time, knee flexion was preserved in swing and occurred earlier. At the ankle, mean dorsiflexion improved at the time of examination (from 8 degrees [SD, +/-9 degrees] to 14 degrees [SD, +/-11 degrees] with the knee in extension), in terminal stance (peak from 7 degrees [SD, +/-9 degrees] to 12 degrees [SD, +/-8 degrees]), and in swing. Peak ankle power generation in stance was preserved and shifted later in stance toward push-off, with no functional weakening of the ankle plantar flexors. A longer-term assessment of a subset of patients with a second postoperative gait analysis at a mean of 4 years after surgery showed that gains measured at 1 year were maintained during the longer term. A subgroup demonstrating a jump knee gait pattern (as defined by excessive knee flexion at initial contact followed by rapid knee extension to full knee extension in midstance) had a tendency to go into knee hyperextension in stance with resultant net knee flexor moment after surgery. This raises concern about the indications for hamstring lengthening in this patient group.