This study has adopted the Grasha-Riechmann model to assess the preferred teaching styles as self-reported by a group of mathematics faculty members at a state university of higher education. The teaching styles have become a mechanism used to convey how teachers define and carry out their instruction. This study aimed to examine how these styles were utilized, implemented, and influenced students' achievement. Statistical fractal analysis was used to examine students' test scores. Results have presented evidence that styles are blended, and the implementation level depends on the learning needs or preferences. There is a likelihood that the blended styles are utilized both in a teacher-centered and a student-centered learning method. It has clear indications that teachers have leveraged the blended styles so that its implementation could have reduced the learning variability and positively affect scores. The fractal dimension quantifies this variation, which is evident in the scores. The large fractal dimension indicates greater teacher control in the learning process. This numerical information provides a better understanding for teachers of the hidden dimensions in the learning outcomes of their students. Although teaching styles are inherent in mathematics education, only a few have indicated effective blended styles; it also demonstrates how it affects student achievement. This paper calls for a comprehensive study of teaching style preferences involving behavioral variables: social and cultural, integrated into the cognitive aspects to provide a generalized style, and utilizing fractal statistical analysis. The results could have pedagogical implications in mathematics education.