The mean, median and mode are the statistical means often used in statistics to describe the trend of a set of numbers sets. However, the three kinds of data are not only different in definition, but also different in application. For the same set of numbers, there is no definite and strict standard whether the mean, median or mode should be used. What's more, using different data may even lead to diametrically opposite conclusions. Even if we finally choose some kind of data, to what extent can this kind of data describe the centralized trend of a group of numbers? Is there any statistical method that can describe the centralized trend of a group of numbers more objectively and fairly? This is not only a very interesting issue, but also an issue of great research value.