This paper introduces a structured framework for integrating probabilistic risk metrics and cost considerations into power system resilience analysis. It distinguishes between Risk of Failure—quantifying the likelihood of disruption—and Conditional Value-at-Risk (CVaR), which captures the severity of losses in extreme scenarios. By modelling these metrics as functions of the separation between stress and strength distributions, the framework reveals key trade-offs between reliability and economic impact. A risk-adjusted formulation is proposed to reflect stakeholder preferences, enabling planners to assess design robustness not only through failure probability but also through tail-sensitive loss estimates. While the methodology is illustrated through representative scenarios, no numerical results are presented. Future work will include quantitative validation and sensitivity analyses to demonstrate practical applicability for guiding investment decisions under uncertainty and for evaluating resilience strategies in the face of High-Impact, Low-Frequency events.