The aim of this study was to test whether or not the threshold to the detection of passive motion (TTDPM) and passive joint position sense on the affected shoulder of patients with rotator cuff tear (RCT) was impaired compared to those on the unaffected side and to investigate the relationship between the tear size and changes in the TTDPM and passive joint position sense induced by RCT. This study included 21 patients with unilateral RCT before arthroscopic rotator cuff repair. To investigate proprioception in this study, we measured the TTDPM and passive joint position sense in abduction and external rotation using an isokinetic dynamometer. The tear size was evaluated intraoperatively under direct arthroscopic visualization. The TTDPM in abduction and external rotation was significantly longer on the affected side than on the unaffected side. However, the angular absolute error in passive joint position sense in abduction and external rotation was not significantly different between the limbs. A comparison according to the tear size impaired proprioception of the TTDPM in the larger tear group showed significantly longer values than in the smaller group. There was impaired proprioception of TTDPM in patients with RCT, and the impaired proprioception was related to tear severity. Impaired proprioception of TTDPM may inhibit consistent muscle recruitment to achieve precise control. Our results suggest that clinicians should consider proprioceptive exercises for impaired proprioception in their treatment for conservative or postoperative patients.