Programs increasingly rely on randomization in applications such as cryptography and machine learning. Analyzing randomized programs has been a fruitful research direction, but there is a gap when programs also exploit nondeterminism or, more generally, when they exhibit concurrent behavior. In this talk, we will present a framework based on Outcome Logic to develop program logics to reason about mixtures of effects in programs, including nondeterminism, concurrency, and randomization. We will show how one can integrate concurrent and probabilistic separation logics principles into Outcome Logic to introduce new compositional reasoning principles.