One of the challenges in the study of music and emotion is to achieve a comprehensive understanding of the role of listener attributes involved in musical emotion induction. Studies have typically focused on one class of listener features, such as particular personality traits or musicianship status. Drawing from the induction rule model (Scherer & Zentner, 2001), which accords special significance to expertise, stable dispositions, and current mood state, we adopted a systematic approach to the study of listener attributes by examining the relative influence of musical expertise, objectively assessed musical aptitude, the Big Five personality traits, and positive and negative mood on musical emotion induction. To this end, 113 participants (45% musicians, 55% nonmusicians) provided ratings of felt emotion in response to 12 excerpts of Western classical music, selected to evoke different types of emotions. Ratings were obtained with Geneva Emotion Music Scale (GEMS)-25, a domain-specific scale for the assessment of music-evoked emotion, and analyzed according to their intensity and their granularity. Findings suggest that expertise mattered most, whereas the effects of mood and personality traits were more dependent on the type of outcome, such as the type of experienced emotion and the type of emotional responding (intensity vs. granularity), as well as on the presence of other listener factors. In total, listener features accounted for around 30% of the variance in musically evoked emotions. The significance of these findings, beyond their contribution to research on musical emotion induction, may extend to applications in which personalization of musical listening is desirable.