Establish a comprehensive evaluation system for coaches' work in different events, and use different methods to establish evaluation standards and scales. Generally speaking, we use four categories to describe our measurement standards: work experience, coach score and training ability. In the incomplete model 1, we use the analysis of variance to evaluate the significance of each index. The most important measures are working years, total games, total wins, national champions, league champions and their contribution to professional leagues.

We use analytic hierarchy process to combine these indicators into one. This standard is the evaluation index.

Considering the development of different sports and the time and space factors of coaches in different times, we decided to introduce time factor 2 into the model.

In order to improve the model, we combine principal component analysis and analytic hierarchy process in model 3. Principal component analysis can well extract the statistical characteristics of coaches' multidimensional data.

Our conclusion is that excellent coaches don't always win more championships, and those coaches who can train players can also be considered excellent. We can also draw the conclusion that a coach's ability should be measured against the background of sports development in his time. Good, no need to modify.