This study aimed to evaluate whether gait speed and grip strength predicted clinical outcomes among older adults with blood cancers. We prospectively recruited 448 patients aged 75 years and older presenting for initial consultation at the myelodysplastic syndrome/leukemia, myeloma, or lymphoma clinic of a large tertiary hospital, who agreed to assessment of gait and grip. A subset of 314 patients followed for ≥6 months at local institutions was evaluated for unplanned hospital or emergency department (ED) use. We used Cox proportional hazard models calculated hazard ratios (HRs) and 95% confidence intervals (CIs) for survival, and logistic regression to calculate odds ratios (ORs) for hospital or ED use. Mean age was 79.7 (± 4.0 standard deviation) years. After adjustment for age, sex, Charlson comorbidity index, cognition, treatment intensity, and cancer aggressiveness/type, every 0.1-m/s decrease in gait speed was associated with higher mortality (HR, 1.20; 95% CI, 1.12-1.29), odds of unplanned hospitalizations (OR, 1.33; 95% CI, 1.16-1.51), and ED visits (OR, 1.34; 95% CI, 1.17-1.53). Associations held among patients with good Eastern Cooperative Oncology Group performance status (0 or 1). Every 5-kg decrease in grip strength was associated with worse survival (adjusted HR, 1.24; 95% CI, 1.07-1.43) but not hospital or ED use. A model with gait speed and all covariates had comparable predictive power to comprehensive validated frailty indexes (phenotype and cumulative deficit) and all covariates. In summary, gait speed is an easily obtained "vital sign" that accurately identifies frailty and predicts outcomes independent of performance status among older patients with blood cancers.