Background: In low/middle-income countries, most nutritional assessments use the latest weights, without reference to growth trajectory. Objectives: This study explores whether velocity, in addition to the latest weight, improves the prediction of wasting, stunting, or mortality in the fi rst 2 years of life. Methods: We analyzed a combined data set with weight and height data collected monthly in the fi rst year of 3447 children from Pakistan, Malawi, and South Africa, with height and survival recorded till 24 m. The main exposures were weight-for-age z-score (WAZ) at the end of each 2-m period and weight velocity-for-age z-score (WVZ2) 2 ) across that period. The outcomes were wasting, stunting, or all-cause mortality in the next 1-2 - 2 mo. As a sensitivity analysis, we also used WVZ over 6 mo (WVZ6), 6 ), with matching WAZ. Cox proportional hazard models with repeated growth measures were used to study the association between exposures and mortality. Mixed Poisson models were used for stunting and wasting. Results: Children who were already stunted or wasted were most likely to remain so. Higher WVZ2 2 was associated with a lower risk of subsequent stunting (risk ratio [RR]: 0.95; 95% confidence fi dence interval [CI]: 0.93, 0.96), but added minimal prediction (difference in AUC = 0.004) compared with a model including only WAZ. Similarly, lower WVZ2 2 was associated with wasting (RR: 0.74; 95% CI 0.72, 0.76) but the prediction was only marginally greater than for WAZ (difference in AUC = 0.015). Compared with WAZ, WVZ6 6 was less predictive for both wasting and stunting. Low WVZ6 6 (but not WVZ2) 2 ) was associated with increased mortality (hazard ratios: 0.75, 95% CI: 0.67, 0.85), but added only marginal prediction to a model including WAZ alone (difference in C = 0.015). Conclusions: The key anthropometric determinant of impending wasting, stunting, and mortality appears to be how far below the normal range the child's ' s weight is, rather than how they reached that position.