Two studies are presented that attempt to test commonly held assumptions regarding mathematics preparation as a predictor to success and retention in engineering programs. Many engineering educators, high school teachers and guidance counselors recommend that engineering programs should only be attempted by students who excel in accelerated and advance placement high school mathematics. The underlying assumption behind this belief is that success in high school mathematics is the best predictor of success as an engineering student and, by extension, a practicing engineer. Boise State University, as a metropolitan university in a region that is not supported by a community college system, is uniquely situated to test these assumptions. We have found that more than 30% of our successful students (those who earn bachelors degrees) began their college career at the eighth or tenth grade algebra level. Even more significantly, we found that the grade earned in their first college level mathematics class was significantly correlated to whether or not they persisted in engineering while the level at which they began mathematics study at the university was not. These results, if they prove to be representative, have significant implication for policy, admissions standards and program development to improve retention in engineering schools.