This study employs the Diebold and Yilmaz (Int J Forecast 28:57–66, 2012) spillover model to investigate volatility spillover effects between crude oil and soft agricultural commodities. The results indicate a significant increase in systemic risk following the global financial crisis (GFC) of 2008–09, highlighting the importance of understanding the drivers of risk transmission. The findings underscore the need for strategic portfolio selection to mitigate extreme market fluctuations. Despite offering higher returns, crude oil exhibits substantial volatility in terms of spillover reception and transmission in the postcrisis period. Additionally, commodities such as rubber, sugar, and coffee are identified as major contributors to spillover intensification after the crisis. To assist in asset selection, this study proposes an asset allocation strategy based on a connectedness network derived from a spillover matrix designed to minimize systemic risk. The approach is applied to three crisis periods: the GFC (2008–09), the global commodity crisis (2014–15), and the COVID-19 pandemic (2020–21). In each case, a portfolio of four assets is selected, with the following commodities chosen: GFC (2008–09): wheat, rapeseed, and corn; the global commodity crisis (2014–15): cotton, rapeseed, and rice; and the COVID-19 pandemic (2020–21): wheat, rubber, and rice. The proposed approach provides valuable insights for policymakers, facilitating informed decision-making to address inflationary pressures and improve portfolio resilience during periods of market volatility.