<p>This study proposes a novel evaluation concept and method: assessing the value of academic journals by measuring their contributions to the knowledge system. It aims to address the limitations of traditional peer review methods and quantitative approaches based on bibliometrics and altmetrics in the practical evaluation of academic journals. The study hypothesizes that academic journals play a crucial role in the knowledge system by providing valuable information through the publication of research papers, thereby reducing uncertainty within the system. As the knowledge system evolves from disorder to order, its information entropy value tends to decrease, and the academic contributions of journals can be characterized by the negentropy derived from these publications. The study employs the concept of counterfactual research to calculate the information entropy of both the factual knowledge system and the counterfactual knowledge system in the absence of the evaluated journals. The difference in information entropy values indicates the negative entropy contributed by the evaluated journals to the knowledge system. Through empirical data, this study demonstrates that this innovative method can effectively reflect the value of journals based on their actual contributions, and it has the potential to complement traditional evaluations of journal value based on impact after further refinement. The empirical data also reveal that, in general, a small number of journals within each discipline make significant contributions to the knowledge system, while the majority of journals contribute little or nothing. This finding aligns with the nucleus zone of periodicals described by Bradford’s Law.</p>