The seismic prediction of pore pressure in formations holds great significance for the exploration and development of complex oil and gas reservoirs, reducing drilling costs, hydraulic fracturing, reservoir evaluation, and other researches. The phenomenon of formation overpressure is widespread in reservoirs, with multiple influencing factors including changes in fluid volume, structural squeezing of formations, and compaction. Existing methods for seismic prediction of formation pressure face challenges in obtaining the background trend of formation compaction, typically relying on theoretical rock physics modeling and three-dimensional interpolation. The established background of formation compaction often contains significant errors, leading to inaccurate predictions of formation pressure. Therefore, based on rock physics modeling theory, this paper proposes a pore pressure seismic prediction method based on normal trend ratio. First, through petrophysical analysis, the deterministic and statistical petrophysical relationship between pore pressure trend ratio and elastic impedance is constructed. Under the framework of Bayesian inversion theory, pre-stack seismic data is used to achieve direct inversion of formation compaction trend. Then, the Eaton model is combined to predict formation pore pressure. This method achieves direct seismic prediction of formation pressure trend ratio through seismic inversion, effectively avoids the problem of large indirect errors caused by three-dimensional interpolation of the normal compaction background trend of the formation and improves the accuracy of formation pressure seismic prediction. Finally, using actual well logging and seismic data in a certain area of the Bozhong Depression of Bohai Bay Basin, this method is used to conduct actual seismic prediction of pore pressure, and achieved good application results, verifying the effectiveness and applicability of the method.