Nowadays, voltage sag continues to remain a critical PQ issue in the industry. Since it is not possible to install a voltage analyzer on every node, optimal monitoring locations must be determined. However, during the PQ monitoring campaign of the Lithuanian DSO grid, the execution of this task was inhibited by the lack of knowledge and literature about the fundamentals of voltage sags propagation. Therefore, the first part of this paper investigates the propagation paths of voltage sags by using a voltage sag matrix method on the created test grid, which for the first time is BRELL-based. This paper is the first to not only investigate voltage sags propagation paths (both downward and upward) in detail, but also to investigate them complexly, encompassing all four types of short-circuits, both voltages (phase-to-phase and phase-to-ground), and interconnections with other PQ events (interruption, transient, unbalance). The result has established a generalization of voltage sags propagation mechanisms by inductive reasoning and laid the foundation for the further development of PQ theory. The second part of this paper focuses on the Lithuanian DSO grid and is one of the first to investigate voltage sags/interruptions through the prism of both relationship with reliability indexes (SAIFI and SAIDI) and primary causes. For this purpose, we created our own scientific methodology, applying known probabilistic and statistical methods along with newly proposed approaches, in particular for ACR success evaluation and estimation of functions similarity.