Tide–surge interaction plays an important role in the distribution of surges along the coast of China. A comprehensive understanding of tide–surge interaction will provide more accurate estimates of extreme sea levels and storm surges. This study applied a statistical method to hourly tide–surge data from 12 tide gauges located along the coast of China to examine the dependency of residual maxima on tidal phases. Statistical significance has been tested quantitatively for each site utilizing the chi-square test. Results show that significant tide–surge interaction exists at all 12 tide gauge sites, but the strength of the interactions differs according to location. The distributions of peak residuals are significantly different from the uniform distribution and can be divided into three types. In the first type, the residual generally reaches its peak value during the rising tide (2–4 h before high tide), whereas in the second type, the residual reaches its peak mostly during the ebb tide (2–4 h after the high tide). In the third type, the residual reaches peaks during both ebb and rising tides. The main causes of tide–surge interactions are the tidal phase alteration caused by surge and the modulation of surge due to tides at the tide gauge. Tide–surge interactions at Lianyungang and Xiamen are more significant than those at other gauge sites. At Kanmen and Keelung, the tide–surge interactions are weakest. Variations of the tide–surge interaction at Xiamen and Quarry Bay show that the interaction has not varied significantly during the past 40 years and that tide–surge interactions tend to be more stable with the strengthening of the interaction.