Soft tissue balancing in TKA has traditionally relied on surgeons' subjective tactile feedback. Although sensor-guided balancing devices have been proposed to provide more objective feedback, it is unclear whether their use improves patient outcomes. We conducted a randomized controlled trial (RCT) comparing freehand balancing with the use of a sensor-guided balancing device and evaluated (1) knee ROM, (2) patient-reported outcome measures (PROMs) (SF-12, WOMAC, and Knee Society Functional Scores [KSFS]), and (3) various surgical and hospital parameters (such as operative time, length of stay [LOS], and surgical complications) at a minimum of 2 years of follow-up. A total of 152 patients scheduled for primary TKA were recruited and provided informed consent to participate in this this study. Of these, 22 patients were excluded preoperatively, intraoperatively, or postoperatively due to patient request, surgery cancellation, anatomical exclusion criteria determined during surgery, technical issues with the sensor device, or loss to follow-up. After the minimum 2-year follow-up was accounted for, there were 63 sensor-guided and 67 freehand patients, for a total of 130 patients undergoing primary TKA for osteoarthritis. The procedures were performed by one of three fellowship-trained arthroplasty surgeons (RPS, HJC, JAG) and were randomized to either soft tissue balancing via a freehand technique or with a sensor-guided balancing device at one institution from December 2017 to December 2018. There was no difference in the mean age (72 ± 8 years versus 70 ± 9 years, mean difference 2; p = 0.11), BMI (30 ± 6 kg/m 2 versus 29 ± 6 kg/m 2 , mean difference 1; p = 0.83), gender (79% women versus 70% women; p = 0.22), and American Society of Anesthesiology score (2 ± 1 versus 2 ± 1, mean difference 0; p = 0.92) between the sensor-guided and freehand groups, respectively. For both groups, soft tissue balancing was performed after all bony cuts were completed and trial components inserted, with the primary difference in technique being the ability to quantify the intercompartmental balance using the trial tibial insert embedded with a wireless sensor in the sensor-guided cohort. Implant manufacturers were not standardized. Primary outcomes were knee ROM and PROMs at 3 months, 1 year, and 2 years. Secondary outcomes included pain level evaluated by the VAS, opioid consumption, inpatient physical therapy performance, LOS, discharge disposition, surgical complications, and reoperations. There was no difference in the mean knee ROM at 3 months, 1 year, and 2 years postoperatively between the sensor-guided cohort (113° ± 11°, 119° ± 13°, and 116° ± 12°, respectively) and the freehand cohort (116° ± 13° [p = 0.36], 117° ± 13° [p = 0.41], and 117° ± 12° [p = 0.87], respectively). There was no difference in SF-12 physical, SF-12 mental, WOMAC pain, WOMAC stiffness, WOMAC function, and KSFS scores between the cohorts at 3 months, 1 year, and 2 years postoperatively. The mean operative time in the sensor-guided cohort was longer than that in the freehand cohort (107 ± 0.02 versus 84 ± 0.04 minutes, mean difference = 23 minutes; p = 0.008), but there were no differences in LOS, physical therapy performance, VAS pain scores, opioid consumption, discharge disposition, surgical complications, or percentages of patients in each group who underwent reoperation. This RCT demonstrated that at 2 years postoperatively, the use of a sensor-balancing device for soft tissue balancing in TKA did not confer any additional benefit in terms of knee ROM, PROMs, and clinical outcomes. Given the significantly increased operative time and costs associated with the use of a sensor-balancing device, we recommend against its routine use in clinical practice by experienced surgeons. Level I, therapeutic study.