To address the issue of ship-to-bank collision probability for large container ships under the influence of wind loads and bank effects in port channels, this paper proposes a prediction approach based on maneuverability analysis and Monte Carlo simulation. Taking a 10,000 TEU container ship navigating in the Yangshan Port channel as the research scenario, probability distribution models for ship route deviation and wind conditions are constructed using AIS (Automatic Identification System) data and meteorological records. Numerical simulation is conducted using the MMG (Maneuvering Modeling Group) model, accounting for stochastic factors such as ship speed, container stacking configuration, wind speed, wind direction, and ship-to-bank distance. Through maneuvering motion simulations, the relationship between the ship-to-bank collision probability and ship speed, as well as stacking pattern, is evaluated. Results show that maintaining 40% of the service speed significantly reduces the collision probability. The aerodynamic characteristics of stacking patterns have an important influence on the stopping and collision avoidance process. The comb-shaped stacking pattern has a higher probability of control loss but a lower final collision probability, while the empty aft deck stacking pattern shows the opposite effect. This research provides a quantitative analytical approach for identifying the environmental factors that affect the navigation safety of ships entering and leaving ports, as well as for supporting ship maneuvering decision-making.