With the deepening advancement of educational informatization and the advent of the 2.0 era, online autonomous learning has become an important component of college English teaching and a key breakthrough in its reform. Based on the Self-Determination Theory and Self-Regulated Learning Theory, this study analyzes the learning data from the “Lighthouse Reading” platform in the first half of 2025 at a university to examine the current status of college students’ online autonomous learning in English. The results reveal a significant polarization in students’ online learning, reflected not only in learning duration and task completion rates but also in the differences in learners’ self-regulation abilities. Additionally, the study finds that learning engagement and outcomes are positively correlated, but nonlinear—mere time investment does not guarantee learning quality, as the key lies in the application of effective metacognitive strategies for deep learning. These findings provide important theoretical foundations and practical guidance for optimizing college English online teaching models.