Introduction Superior capsule reconstruction (SCR) has been used for the tendon grafting of massive rotator cuff tears when primary repair is difficult. We examined the postoperative outcomes of SCR for massive rotator cuff tears and the risk factors for postoperative retear. Materials and methods Through this retrospective comparative study, we evaluated 35 patients with an average age of 75.3 (57-90) years who underwent SCR using the technique developed by Mihata et al. Clinical outcomes were evaluated 1 year postoperatively using the Japan Orthopedic Association (JOA) score, University of California Los Angeles (UCLA) shoulder score, elevation angle and the Sugaya classification, which uses a 5-point scale evaluation on magnetic resonance imaging in which types 4 and 5 are considered retears. We also investigated the progression of fatty degeneration before and after surgery and the rupture site of the graft. Acromio-humeral distance (AHD), before and after surgery was measured through X-rays. Rotator cuff tear-related shoulder arthritis was evaluated on plain X-rays using the Hamada stage. Risks of retear were identified using multiple regression analyses for sex, age, Hamada stage and JOA score. Results The JOA score improved from 62.3 ± 9.49 (SD) preoperatively to 84.6 ± 5.66 (SD) postoperatively (P < 0.001). The UCLA score improved from 15.3 ± 3.77 (SD) preoperatively to 30.1 ± 3.11 (SD) postoperatively (P < 0.001). AHD increased from 4.03 mm preoperatively to 6.23 mm postoperatively (P < 0.001). Postoperative retear was observed in seven of the 35 patients. Moreover, retear was observed in five of nine patients with a Hamada stage ≥ 4. Multiple regression analysis revealed that age ≥ 80 years, male sex and Hamada stage ≥ 4 were risk factors for retear. Conclusions While the postoperative outcomes of SCR are favorable, age, sex and degree of arthropathic change should be considered for surgical indications of SCR.