This paper examines the often-invoked principle that diversifying international trade partners reduces economic risk, and argues that an uncritical application of this concept at the country level can be misleading. Using historical context and a simple theoretical model, we show that while spreading trade across many countries – akin to portfolio diversification – can mitigate idiosyncratic shocks, it provides little protection against systemic shocks and can mask dangerous concentrations at the firm and supply chain levels. Key insights include: (1) Countries do not trade; firms do – and export flows are often dominated by a small number of firms or commodities, so national diversification statistics may conceal micro-level vulnerabilities. (2) Simply increasing the number of trade partners yields diminishing returns in risk reduction, especially when partners’ economies are correlated; more partners do not automatically mean less exposure. (3) A diversified macro-level trade profile does not ensure that individual firms or industries are diversified or resilient. We formalize these ideas in a mean–variance portfolio framework and illustrate how true risk reduction depends on having independent streams of trade (not merely multiple streams). Policymakers are cautioned to look beyond aggregate metrics: without examining who trades what with whom, efforts to diversify trade can create a false sense of security – a phenomenon akin to “diworsification” in finance. The paper concludes by suggesting a multi-level approach to trade diversification that emphasizes firm-level and supply-chain considerations for genuine economic resilience.