One of the most significant rewards in science is peer recognition, often bestowed in the form of awards. However, little is known about what sets apart award-winning contributions, how award committees determine prize-worthy contributions, and why some scientists are more likely to be recognized than others, particularly in the field of computer science. Using a mixed method approach that includes qualitative and quantitative techniques, this study investigates the characteristics of award-winning contributions, and the education and career factors associated with recipients of the Turing Award, a Nobel equivalent award in computer science, and compares them to those of a matched group of non-winning scientists. In regard to award-winning contributions, the study finds that the Turing Committee was just as likely to recognize contributions related to practice ("applied research") as to theory ("basic research"). In regard to education and career factors, the study reveals that neither scientific productivity nor the quality of contributions differentiated winning from non-winning scientists and their contributions. However, early advantages, visibility to the awarding association, prior eminence, and affiliation with a top computer science department distinguished award winners. These findings suggest that excellence in computer science is a quality that has not been defined, explained, or communicated by the award committee to the computing community or to the public. The findings call attention to the limitations of peer reviews and the importance of improving the design of nomination, evaluation, and selection procedures as well as citations accompanying the Turing Award and other computer science awards.