Background: The Comprehensive Geriatric Assessment (CGA) has proven to be a valuable tool for providing a more comprehensive health evaluation of allogeneic stem cell transplantation (allo-SCT) recipients. Methods: We prospectively developed and tested a new Simplified Geriatric Score-4 (SGS-4) on 135 consecutive patients aged ≥50 years who underwent allo-SCT between 2020 and 2023. Each CGA component was individually analyzed for its association with overall survival (OS), non-relapse mortality (NRM), and cumulative incidence of relapse (CIR). Then, we performed a two-factor analysis (FA) using oblimin rotation and Bartlett estimation on all CGA components and sex. Based on component weights, a simplified geriatric score-4 score (SGS-4) was created: [Gait Speed] + 2 × [Hand Grip] + Geriatric 8 + 1.5 × [Sex]. ROC analysis defined three fitness groups, frail (≤13), prefrail (>13–22.5), and fit (>22.5). Results: Reduced hand grip strength and impaired mini mental state examination (MMSE) were associated with worse OS and higher NRM. Vulnerable Elders Survey (VES-13) and Fondazione Italiana Linfomi (FIL) scores also indicated poorer OS, though with uneven group sizes. Other CGA domains and the Hematopoietic Cell Transplantation–Comorbidity Index (HCT-CI) showed no significant prognostic value. The SGS-4 effectively stratified patients into three fitness groups, with those in the frail category experiencing lower OS and an increased risk of relapse. Conclusions: The new Simplified Geriatric Score-4 (SGS-4) based on three CGA domains (gait speed, hand grip, Geriatric 8) and sex effectively predicts OS and CIR risk in patients aged ≥50 years undergoing allo-SCT. The study’s small sample size and disease heterogeneity warrant further validation in larger cohorts.