The probable maximum flood (PMF) is the highest design criterion for major hydraulic projects whose failure might cause severe loss of life and property. The probable maximum precipitation (PMP) is the corresponding extreme design rainfall, which is defined as the theoretically greatest depth of precipitation for a given duration over a particular area under modern meteorological conditions. The moisture and wind maximization (MWM) method is widely used in deriving 24‐hr PMP. However, variables in MWM factor (e.g., maximum precipitable water and wind speed) are collected from full study durations regardless of their respective main moisture inflow directions in each year. This issue could lead to higher PMP estimates which could lead to costs in the construction of hydraulic structures. To alleviate this possible problem, an improved moisture and wind maximization (IMWM) approach is proposed based on the concept of a “wind rose diagram”, data (e.g., rainfall, wind speed, dew point) are grouped according to the main moisture inflow directions of a study catchment. Analysis shows the IMWM approach has the potential to provide lower PMPs, which would have economic implications for hydraulic designs. The storm transposition approach is applied for safety, extraordinary storms occurring in the homogeneous rainfall region are transposed to a study catchment after being maximized by the IMWM method. In addition, statistical methods, including storm formulation, trend analysis and hydrological frequency analysis, are used for deriving short‐duration PMPs. The Nash instantaneous unit hydrograph coupled with regional empirical formulas is then applied to estimate PMFs. As an example, PMP/PMF estimation approaches are applied to an important flood control project located in a small ungauged catchment of China. The study can provide a scientific basis for the construction of the study project as well as a reference for similar high‐hazard projects.