Carbon emissions trading is essential for reducing carbon emissions, and its role in regional carbon unlocking needs further clarification. This study uses the difference-in-differences (DID) model and synthetic control model (SCM) to evaluate the carbon unlocking effect of China’s six pilot carbon trading provinces. This study found that (1) carbon lock-in effects in China are mainly influenced by technology lock-in and fixed input lock-in; (2) each province’s overall carbon lock-in level presents a decreasing trend yearly, and the regional distribution presents characteristics of “low in the east and high in the west”; (3) carbon emissions trading pilot policies effectively promote the carbon unlocking effect in pilot regions overall, with Guangdong having the most significant unlocking effect. Conversely, Beijing, Hubei, Chongqing, and Shanghai also had different degrees of carbon unlocking. Finally, (4) an assessment of impact mechanisms indicates that technology and institutions have a significant mediating role in effectively promoting carbon unlocking under the carbon trading policy. Conversely, social behavior has an inverse effect, and fixed assets are not affected by the policy. This study demonstrates the carbon unlocking effect of carbon emissions trading and provides a quantitative reference for implementing carbon emissions trading policies and determining carbon unlocking paths.