This week covered the second half of EDA — statistical foundation and validation design. Honestly summarizing data (statistics), reading relationships between variables (correlation), avoiding pitfalls that contaminate evaluation (leakage), and reliably estimating generalization performance (validation). These four form the final gate of MLS-C01 Domain 2 and hide the answer to the practical question "why does my model fail in production?" Today we review four days as one integrated flow.