Abstract: This paper puts forward the problem of weight failure implied in academic evaluation, namely, when some evaluation indexes “increase and decrease at the same time”, no matter how the weight combination of these indexes changes, the ranking of evaluation results is almost unchanged, making the weight almost of no impact. This paper makes a further analysis on the formation mechanism, characteristics, measurement methods, influencing factors and consequences of weight failure, and takes JCR2019 economics journal as an example to carry out simulation experiments. The results show that weight failure is a common phenomenon in academic evaluation, which should be paid enough attention to. Weight failure is a double-edged sword—it can reduce the impact of weight fault and weight error, but also affects the quality of evaluation. The influencing factors of weight failure include index correlation, total weight and evaluation scale. This research enriches the evaluation theory in Information Science.