Despite their central role in Emergency Department safety and flow, modern five-level triage systems consistently deliver performance that is “good but not excellent.” This chapter analyzes the persistent “0.7 paradox”: across settings and validated frameworks, discriminatory performance typically plateaus around AUROC 0.70–0.80, rarely exceeding 0.8 despite revisions, training, and digital implementation. The chapter explores why this ceiling likely reflects a convergence of factors rather than a single design flaw: structural properties that privilege safety, cognitive variability, and context-dependent decision-making under uncertainty and clinical complexity that is only partially captured at first contact. It also addresses methodological barriers to improvement, including fragmented evidence, heterogeneity of outcomes, and the absence of universally accepted, clinically “urgency-sensitive” indicators, arguing for more proximal, syndrome- and time-based endpoints (e.g., time-to-key-action) and retrospective urgency recording. Finally, the chapter advocates a shift from repeatedly “reinventing” new scales toward a collaborative, evidence-based refinement of existing validated systems.