<p>Natural gas (NG) is a cost-effective and relatively clean energy source, with North America’s market offering profitable opportunities for storage and trading via price arbitrage. Trading companies take advantage of price fluctuations by purchasing low, storing, and then selling high. Optimizing injection and withdrawal schedules is key to maximizing profits, but this is challenged by price uncertainty and operational constraints. While prior research has offered valuable insights into NG storage optimization, existing models often rely on complex stochastic price processes, which overlook actionable futures price predictions already available to trading companies. Moreover, dynamic inventory ratcheting constraints, which are critical for ensuring feasibility, are largely ignored, and modern risk management techniques such as chance-constrained programming (CCP), robust optimization (RO), and distributionally robust optimization (DRO) remain underexplored. This paper addresses these gaps by developing a decision-support framework tailored to the needs of NG trading companies. Three mathematical models incorporating constant deliverability, dynamic inventory ratcheting, and path-dependent constraints are developed. The used approach leverages observable market data, ensuring model transparency and practical relevance. Additionally, risk-aware formulations are introduced using CCP, RO, and DRO to hedge against prediction uncertainty without relying on extreme assumptions. Numerical experiments, based on real-world data, validate the models’ tractability and practical effectiveness. The inclusion of ratcheting constraints proves essential for technical feasibility, while risk-aware strategies improve worst-case performance and reduce profit variance. These combined contributions offer a robust foundation for optimizing Future-to-Future NG trading decisions in uncertain market environments.</p>