Background and Objective: Early identification of medical students at risk of failing is essential for timely support. Raw formative scores are commonly used but are affected by exam difficulty and cohort variability, reducing fairness and comparability. Normalized scoring adjusts for these factors, providing standardized measures of performance. This study compared normalized and raw formative scores among first-year MBBS students and examined their correlation with summative exam results. Methods: A cross-sectional analytical study was conducted among 200 first-year MBBS students during the 2024–2025 academic year. Raw and normalized scores from the first formative assessment were analyzed using descriptive statistics, Pearson’s correlation, linear regression, and ROC analysis. Results: Normalized scores correlated more strongly with summative results (r = 0.64, p < 0.001) than raw scores (r = 0.56, p < 0.001) and explained a greater variance in performance (R² = 0.41 vs. 0.31). Predictive accuracy for identifying at-risk students was higher for normalized scores (AUC = 0.86) compared with raw scores (AUC = 0.78; p = 0.032). Conclusion: Normalization improved score fairness, predictive validity, and diagnostic accuracy. Incorporating normalized scoring into formative assessments can enhance equity, reliability, and early detection of students requiring academic support in medical education.