Terzaghi’s (T), Bjerrum-Eide’s (BE) and the slip circle (SC) methods are widely used to evaluate the basal heave stability in reliability-based design of braced excavations. The three methods produce different reliabilities for the same excavation problem. It is still unclear which of these methods is most conservative or most economical. Hence, this study compares the reliabilities of basal heave stability for the three methods for both wide and deep excavations. The probability of failure (P _f ) for a given factor of safety (FS) and coefficient of variation (COV) for soil parameters is calculated using the first-order reliability method. The P _f − FS design charts for different methods are compared. The results demonstrate that the SC method is more conservative than the other two methods. For wide excavations, the same level of the factor of safety results in a much smaller P _f for the SC method than for the T method. For deep excavations, the P _f for the SC method is also smaller than for BE method. In addition, the P _f − FS design curves for T are more sensitive to the penetration depth of the wall than the other two methods. These results serve as a guideline for the method selection of basal heave stability evaluation in geotechnical practice.