Properly managing the relationship between food security, ecological protection, and urbanization, and coordinating the trade-offs among these three factors for land demand are extremely important for environmental management and sustainable development. In this study, we attempt to analyze the state of land use trade-offs from a dynamic perspective in terms of both potential and efficiency. We have innovatively proposed a new land use trade-off analysis framework integrating the Estimation System for Land Productivity (ESLP) model, machine learning algorithms, ecosystem service value assessment, and spatial analysis method. By applying the framework, the potential and efficiency of the three land use functions of urban development, ecological protection, and agricultural production on the Huang-Huai-Hai (HHH) Plain were comprehensively estimated, and the trade-off relationship between the three land use functions was identified. The results showed a prominent conflict between urban development and agricultural production (around 8%) on the HHH Plain, especially in the Beijing–Tianjin–Hebei urban agglomeration and the southern Jiangsu urban agglomeration. In the mountainous areas, such as northern Hebei and central Shandong, there was an obvious trade-off between ecological land and agriculture land. Most cities had a trade-off between ecological land and urban land (approximately 6% of the study area), but it was relatively more relaxed in comparison. Finally, we found that on the HHH Plain, where land resources are fiercely competitive, spatial planning and land resource control depend not only on the suitability or potential of the land unit, but also on whether the efficiency of land use has reached an appropriate range. The smart way to use land resources is to scientifically trade-off different land use functions and improve the efficiency of land use to achieve maximum benefit.