Abstract: It’s of great importance to detect research front under the context of utilizing key words, but the method based on the word-frequency usually ignore the distribution characteristics of the key-word in the time window. The key words included in articles during 2008 to 2018 in the field of statistics recorded in CSSCI are used as empirical data to fit the cumulative distribution function. Based on the cumulative distribution function, two indicators, which are heat and potential degree for key words, are designed to present cumulative velocity and cumulative acceleration for key word frequency. The integration of the two indicators, heat and potential degree, are used to explore the research front and its dynamic evolution. The result found that the new method totally retains the original distribution characteristics of keywords and describes the evolution process of keywords more precisely. Compared with the method based on the word-frequency, the new method can get different results, which is a useful trial for the temporal cumulative evaluation.