We provide descriptive analyses of discrimination against black, Latino, and Asian homeseekers using the 2012 Housing Discrimination Study. Our chief methodological innovation is to assess discrimination in the rental and sales market at three levels of aggregation: all units seen, units seen in common, and units not seen in common. We measure discrimination in two ways: first, in agent treatment of testers at both the audit level and the unit level for all units and for units seen in common. Second, we measure differences in unit and neighborhood characteristics for units for all units shown to testers and for units not seen in common. We find that the overall incidence of poor agent treatment of minority testers is rather low. Exceptions include the price quoted to black and Asian testers in the sales market and the total number of rental units shown to Asian testers relative to their white counterparts. In terms of differences in unit or neighborhood characteristics, we find some evidence that black and Asian testers are steered into less white, more minority, and somewhat poorer neighborhoods in terms of median household income. We conclude that, unlike in past iterations of the HDS, minority homeseekers are not uniformly treated poorly. We suspect that housing discrimination increasingly serves to steer white and minority homeseekers into different neighborhoods, rather than agents’ acting on a blanket “taste for discrimination” against minority group members.