Unregulated naked selling of virtual power plants (VPPs) in day-ahead markets poses inherent risks to grid security and market fairness. This paper proposes a joint electricity-reserve trading model for VPPs as a strategic measure to mitigate the negative impacts of naked selling. This model systematically evaluates the economic advantages and risks of naked selling, utilizing metrics such as user comfort and conditional value at risk (CVaR). Furthermore, a sophisticated combination of a data-driven level-set fuzzy approach and advanced algorithms, including support vector quantile regression (SVQR) and kernel density estimation (KDE), is employed to quantify the uncertainties related to prices and reserve activation precisely. The results of case studies demonstrate that integrating default penalties within the proposed trading model diminishes the overall revenue of VPPs engaging in naked selling, thereby serving as a robust decision for mitigating the adverse effects of the naked selling of VPPs.