Nationwide out-of-school suspension and expulsion rates show historically underserved groups of students leading discipline disproportionality reports (i.e., 1.1 million African-American, 660,000 in special education, 600,000 Latino, and 210,000 ELL students; U. S. Department of Education, Office for Civil Rights, 2018). While Schoolwide Positive Behavioral Interventions and Supports (SWPBIS) effects on racial discipline disproportionality have been promising, they have been insufficient (McIntosh, Girvan, Horner, and Smolkowski, 2014; Vincent and Tobin, 2011), and empirical work studying the interrelation between classroom management, culture, behavior, and teacher decision-making is needed for a cohesive and theoretically sound approach to addressing the racial discipline gap (Gregory and Roberts, 2017; Fallon, O'Keeffe, and Sugai, 2012; McIntosh et al., 2015). The purpose of this study was to explore the extent to which teachers' perceived classroom management abilities predict racial discipline disproportionality in office disciplinary referrals (ODRs), and how teachers' behavioral expectations of students mediate racially associated discipline differences. Thirty-three teachers in 28 classrooms completed the Culturally Responsive Classroom Management Self-Efficacy Survey (Siwatu, Putnam, Starker-Glass, and Lewis, 2015), and reported their expectations for all students in their classrooms (N=496) using a modified version of van den Bergh, Denessen, Hornstra, Voeten, and Holland (2010) Teacher Expectation scale. The discipline history of classroom students was measured with ODRs during the 2017-2018 academic year. Using multi-level models, a racial discipline gap was evident for African-American students in comparison to White students. Further, through multi-level models and Ordinary Least Squares (OLS) regression with standardized errors corrected, teachers' CRCMSE strength index score was not shown to be associated with the racial discipline gap. Additionally, some of the difference between the number of ODRs received by African-American and White peers appears to be due to the distinction in teacher expectations for these students. Findings also support that teacher expectations have a stronger influence on the ODRs received by Latino students in comparison to their White peers. A summary of findings, limitations to this work, contributions to the literature, and possible implication for future research are discussed.