Teaching quality evaluation of higher education teachers is an essential aspect of higher education evaluation. It is also the core factor that drives the development of high-quality teaching and learning in universities. How to evaluate teaching scientifically is a common concern of the current educational evaluation reform in universities. A new method of evaluating the teaching quality of university teachers based on the value-added of students' performance is proposed to address the shortcomings of scientific and objective nature and the large consumption of human and time resources of most current university evaluation work. Firstly, we analyzed the status data of university students to build a table of four features related to university course performance containing six subfeatures and selected "College English" and "Advanced Mathematics" courses as samples to predict students' course performance. Firstly, we analyzed the status data of university students to build a table of four features related to university course performance containing six subfeatures and selected "College English" and "Advanced Mathematics" courses as samples to predict students' course performance. Then, by comparing the difference between student's predicted and actual test scores, we obtained the value‐added of teachers' performance due to the difference in their teaching quality. We selected this value‐added to quantify the teachers' teaching quality at the class level. Through the combination of university course performance‐related features table and student performance prediction and value‐added evaluation, this method realizes the evaluation of university teachers' teaching quality by using student performance value‐added, which provides a new method for university teachers' teaching quality evaluation with high reliability and high efficiency. It is a valuable exploration of university education evaluation reform in the digital era.