I outline recent advances in the theory of attractor flow trees, demonstrating this to be a powerful tool in the analysis of the BPS spectrum of Calabi-Yau compactifications. As an application, I derive a significantly refined version of the OSV conjecture.