In modeling interconnected electric power systems, the role of input parameters is crucial. For example, when considering the flow series that define the hydrological state of hydroelectric plants, these inputs can directly influence operational economic value and cause variations in generator dispatch to satisfy demand. This study focuses on evaluating the reduction in the dimensionality of the stochastic state space using a CEGH (Correlations in Gaussian Space with Histogram) synthesizer to generate hydrological data. Using advanced electrical modeling techniques, the medium-term modeling of a real interconnected system is analyzed. This system includes wind, solar, and thermal generators, along with four hydroelectric plants and CEGH inputs. Regarding the serial synthesizer, the variation fields are assessed by reducing the state from six to three. Variations in the states are considered with initial ranges such as low (5%–10%), medium (30%–45%), and high (60%–85%), enabling the identification of trend changes and the development of a robust variation matrix. This research develops indicators to assign weights to the simulated cases using the open-source, freely available SimSEE platform. These indicators facilitate the identification of economic impacts resulting from the operating policies derived from the case matrix, with resulting sample variations in MUSD ranging from 4% to 15% in operating costs. Additionally, statistical analysis shows differences in computational costs: cases without reduction require approximately 122 hours of modeling, whereas applying the serial analysis synthesizer reduces this to 2–6 hours, demonstrating significant time savings. Furthermore, the study proposes indicators to assess the flexibility of generators under these hydrological variations.