Despite the ubiquity and complexity of grading, there is limited contemporary research on grading students in schools. There is, however, an outpouring of publications and consultants promoting new approaches. Many eliminate effort and behavior scores, remove the zero, adopt a four-point system, advocate rubrics, or promote their own software packages. To study changes in grading, we collected data in two New York high schools undergoing a year-long professional development program on rethinking grading. We not only used existing literature on grading to frame our study but also relied on institutional theory and teacher identity as frameworks. We found that productive teacher change occurred, but it was partial, tentative, contingent on school-wide support, and not without frustration on the part of teachers.