Enhancing the efficiency of the port cluster in the Guangdong-Hong Kong-Macao Greater Bay Area is of great significance for regional economic integration and the establishment of the “dual circulation” new development pattern. Based on the super SBM model, this paper measures the efficiency of 11 ports in the Greater Bay Area from 2010 to 2023. It comprehensively employs kernel density estimation, standard deviation ellipse, and the Tobit model to reveal the spatiotemporal evolution patterns and influencing factors. The results indicate that, the overall efficiency of the port cluster in the Greater Bay Area shows a fluctuating upward trend, with significant heterogeneity in efficiency changes among individual ports. The efficiency value among ports is obviously stratified, forming a coexistence relationship of competition and cooperation. In terms of the time evolution, the kernel density curve shifts to the right, the main peak flattens, and the number of wave peak evolves from three peaks to four peaks. In terms of the spatial evolution, the center of gravity of the standard deviation ellipse moves to the northwest, the distribution tends to be discrete from concentration, the azimuth first moves counterclockwise and then clockwise, and the oblateness first rises and then decreases. In terms of the influencing factors, industrial structure, port functions, and technological innovation have a significant positive effect on the efficiency of the port cluster, while the degree of opening up has a significant negative effect.