Background Tobacco use and obesity are major public health problems and part of the leading causes of preventable disease and death worldwide. The prevalence of overweight children has escalated over the years; making the weight gain in children a critical issue for health professionals and academics alike. This study examines the association between maternal tobacco use and body weight abnormalities among South African children. Data and methods The study uses data from the National Income Dynamics Study (NIDS), conducted from 2008 to 2017. The survey is available in five waves which have been merged into a panel dataset of 211,718 children aged between 0 to 5 years old, drawn from a nationally representative sample of households followed across time. Thus, the subgroup of children aged between 6 and 15 years old was excluded from the analysis. The empirical investigation employs a logistic regression model to estimate the marginal effect of maternal smoking on childhood overweight prevalence measured by three anthropometric based proxies, namely risk of child overweight, child overweight and child obesity. This framework assumes unobserved child characteristics to be uncorrelated with independent variables (random effect assumption); hence allowing to control for time-invariant sociodemographic factors which are likely to affect child nutritional health. Results In addition to sociodemographic and health characteristics of mothers, empirical findings suggest that maternal smoking exhibits a significant odds and/ or probability of weight abnormalities in children. Notably, children of smoking mothers are likely to be obese, overweight, or at risk of becoming overweight with a possible coexistence of mother-child overweight. Similarly, weight irregularities in children are likely to increase with mother’s age. Conversely, married mothers although associated with increased odds of children having high BMI, display a reduced probability of children being abnormally weighted. Finally, child support grant has the potential to improve children nutritional health as children whose mothers are recipient of such grant have a negative probability of having high BMI. Conclusion Overall, maternal smoking contributes to child overweight and/or obesity in south Africa; suggesting that maternal healthy lifestyle could be an alternative strategic tool to fight against overweight in children. However, mothers’ inability to remain and/or follow a healthy life style is plausible as age increases, with spillover effect on child care. Thus, policymakers should prioritize programs to reduce smoking, especially amongst pregnant women and caregivers, to minimise the risk of overweight in children. Promoting the consumption of healthy foods accompanied by physical activity may reduce mothers’ stress levels and their incentive to self-medicate using tobacco substances. In addition, comprehensive action programs including child support grant and recommendations for treatment plans that address the problems of children who are already suffering from overweight, remain essential.