The combination of \emph{nondeterminism and probability} in concurrent systems lead to the development of several interpretations of process behavior. If we restrict our attention to linear properties only, we can identify three main approaches to \emph{trace} and \emph{testing} semantics: the \emph{trace distributions}, the \emph{trace-by-trace} and the \emph{extremal probabilities} approaches. In this paper, we propose novel notions of \emph{behavioral metrics} that are based on the three classic approaches above, and that can be used to measure the disparities in the linear behavior of processes wrt.\ trace and testing semantics. We study the properties of these metrics, like \emph{non-expansiveness}, and we compare their expressive powers.