The optimization of the cooperation network is a key link to accelerate the high-quality development of regional tourism. Taking the Beijing–Tianjin–Hebei region as an example, this paper measures the tourism cooperation intensity with the modified gravity model, on which the original, binary, and Top networks are generated to identify the spatiotemporal evolution characteristics from the multi-dimensional difference–association–agglomeration model, and provide insight into the determinants by the GeoDetector model. The results show that (1) the cooperation network reveals a diffusion trend with Beijing–Tianjin as the main axis chain, and southward expansion, and the overall differences tend to moderate at a slow pace, among which the north is the weak area. (2) The robustness of the cooperation network association structure is enhanced, showing that the outgoing equilibrium is improved, while the cohesion is strengthened and accessibility changes little. Furthermore, the cities show a core–edge distribution pattern in terms of power roles. (3) The cooperation network shows the phenomenon of hierarchical agglomeration gradually with the expansion of network scale, and eventually evolves into two camps: the Beijing–Tianjin cooperation circle and the Shijiazhuang–Xingtai cooperation circle. (4) Tourism cooperation belongs to the locational traffic constraint type, and making up for the shortcomings of rural development is another key to further enhancing regional tourism cooperation. The future optimization of regional tourism cooperation needs to seek multifactorial promotion paths.