A riverbank landslide with a high speed that enters the water can poses a great threat to public facilities and the lives of residents on both sides of the riverbank. Therefore, the precise calculation of speed is very important for disaster assessment. Water resistance is one of the key factors affecting the speed of riverbank landslide. In order to quantify the resistance calculation of reservoir bank landslides and further provide experimental data and theoretical basis for the analysis of the velocity of reservoir bank landslides, this study has developed an experiment to investigate the water resistance coefficient, featuring a water tank, a sloped section, and a smooth section. The sloped section serves to accumulate kinetic energy in the landslide prior to its entry into the water, while the smooth section is employed to research the water resistance coefficient. This paper has developed a model for calculating the comprehensive water resistance coefficient, grounded in the dynamic and kinematic equations for underwater test blocks. The experimental results were analyzed through dimensionless methods, examining the impact of various dimensionless factors on the water resistance coefficient. The model for calculating the comprehensive water resistance coefficient was derived from multiple regression analysis. The theoretical formula for water resistance coefficient has a fitting degree of 0.77, indicating good accuracy. The study has proposed a sophisticated model for calculating the water resistance coefficient, enhancing our understanding of this coefficient. This advancement enhances the precision of predicting the inflow velocity of landslides into reservoirs, thereby improving the predictive accuracy of tsunami models triggered by such landslides.