The rollercoaster project extracted from HiMCM Contest 2018 the information of 300 rollercoasters around the world and did a series of analysis on various features as well as came up with our own ranking system. We examined the info of the whole dataframe and got rid of the columns with insufficient amount of data. Then, we filled in the few missing values in other columns with the mean of the rest of the data in the respective column so that we could do the further integrated analysis. We then asked several interesting questions regarding the age, geographical distribution and even the names of these rollercoasters. We plotted several graphs exploring the relationships among heights, lengths, and durations. Some interesting bivariate associations were discovered in this process. Then we tried to figure out the logics behind CoasterBuzz Rankings of the rollercoasters and did not find any reasonable explanations. Finally, we came up with our own equation of ranking based on our own preference.