OBJECTIVES: To examine the relationship between angular insertion depth (AID), modiolar proximity, and speech recognition outcomes for cochlear implant (CI) recipients of a precurved electrode array. STUDY DESIGN: Retrospective review. SETTING: Tertiary academic referral center. PATIENTS: Thirty-five adult CI recipients (n = 40 ears) of precurved electrode arrays listening with a CI-alone device. INTERVENTIONS: Cochlear implantation with postoperative computed tomography. MAIN OUTCOME MEASURES: Consonant-nucleus-consonant (CNC) word recognition at 6 months post-activation. RESULTS: A multivariate regression model demonstrated that both deeper apical AID and closer modiolar proximity in the basal turn were independently associated with better CNC word scores at 6 months (F2,37 = 7.264, p = 0.002). A deeper basal insertion depth was positively correlated with apical AID (r = 0.754, p <
0.001) but negatively correlated with modiolar proximity in the basal turn (r = -0.766, p <
0.001). CONCLUSIONS: These data suggest that both apical cochlear coverage and modiolar proximity independently confer speech recognition benefit with a precurved array. However, these benefits are mutually exclusive for current precurved array designs as a deeper basal insertion depth results in greater apical coverage but lateralization of electrodes away from the modiolus in the basal turn. Future work is needed to elucidate mechanisms behind these findings that may motivate electrode array design modifications to further optimize outcomes for CI users.