Carbon trading is an effective market emission reduction mechanism. In order to formulate effective market policy, it is necessary to access a deep understanding of carbon trading policy implementation. So based on the panel data of 30 regions in China from 2001 to 2016, we investigated the factors of carbon emissions through LMDI decomposition approach, and then discuss the effect of policy implementation in eight pilots using synthetic control methods. The main conclusions are as follows: The distribution tendency of carbon emissions in China is obviously high in the east and low in the west, and the proportion of these regions’ emissions to the total are respectively around 48.7%, 34.0% and 17.3%. The total carbon emissions effect can be divided into five parts, which respectively are energy structure, energy intensity, economic development, urbanization and population size. The gap between real and synthetic carbon emissions in each pilot is enlarged after implementation. The probability of the real policy effect is 4.35%, which at a 5% significant level refuses the original assumption that there is no policy effect, and shows that the carbon trading policy can effectively reduce emissions and achieve the goal of total control.