What – if anything – can be learned from studying the career trajectory of successful scientists? To determine whether predicting trends in future scientific breakthroughs might be possible by examining past performance, an international team led by the physicist Albert-László Barabási at Northeastern University in Boston studied the careers of thousands of scientists across numerous disciplines, charting their most cited papers to gauge impact as a measure of success. While the study reaffirmed previous findings about creativity that show high levels of productivity early in scientists’ careers, there was a surprise in the data. When timing of productivity was removed, the research revealed that impact in a scientist’s career is random, and not correlated to experience. In short, success is a complex function of ability, productivity and luck. The full paper is available at Science magazine.