This study assesses the relative influence of soil moisture memory and tropical sea surface temperature (SST) in seasonal rainfall over the contiguous United States. Using observed precipitation, the NINO3.4 index, and soil moisture and evapotranspiration simulated by a land surface model for 61 years, analysis was performed using partial correlations to evaluate to what extent land surface and SST anomaly of El Niño–Southern Oscillation (ENSO) can affect seasonal precipitation over different regions and seasons. Results show that antecedent soil moisture is as important as concurrent ENSO condition in controlling rainfall anomalies over the U.S., but they generally dominate in different seasons with SST providing more predictability during winter while soil moisture, through its linkages to evapotranspiration and snow water, has larger influence in spring and early summer. The proposed methodology is applicable to climate model outputs to evaluate the intensity of land-atmosphere coupling and its relative importance.