Urban competitiveness encapsulates a city’s comprehensive capacity for development. Utilizing panel data for 282 prefecture-level cities from 2012 to 2020, this study constructs an evaluation index system of urban competitiveness and applies kernel density estimation, standard deviational ellipse, trend-surface analysis, and Dagum Gini coefficient decomposition to examine its spatiotemporal evolution and regional disparities. The results indicate that: (1) urban competitiveness exhibits a V-shaped recovery, with intensified polarization after 2016, widening innovation advantages in the East, and persistent decline in the Northeast; (2) the spatial configuration follows a “dual-gradient, polycentric” structure, characterized by an inverted-U pattern along the east–west axis and an expanding gradient gap along the north–south axis; (3) club convergence and hierarchical entrenchment coexist, as polarization deepens in the East while the Northeast tends toward internal balance; and (4) the competitiveness center shifts southeastward, accompanied by a pronounced fragmentation trend in the Northeast. Based on these findings, the paper proposes differentiated spatial governance, the development of multi-tier innovation networks, and the promotion of green and sustainable development as integrated strategies to systematically enhance urban competitiveness.