By Dashun Wang
In science, 1905 is known as the annus mirabilis, or “miracle year,” the period when Albert Einstein, at the age of 26, published several discoveries that changed physics forever.
What happened for Einstein in 1905 can be described as a “hot streak,” or burst of seemingly miraculous success and impact. Our understanding of creative careers to date has suggested they’re unlikely to include hot streaks. For example, my earlier coauthored work found that a scientist’s biggest research hit occurred completely randomly in their sequence of published works: We called this phenomenon the “random impact rule.”
While intriguing by itself, the random impact rule raises puzzling implications: What happens after we finally produce a breakthrough? Indeed, if every work in a career is like a random lottery draw, then one’s next work after a hit may be more mediocre than spectacular, reflecting regression toward the mean.
Most of us—including me—would like to believe that if we produced a big hit, it would help us produce more hits afterward. Are we really regressing toward mediocrity after we break through? To answer these questions my student Lu Liu and I along with other collaborators studied the careers of about 30,000 scientists, artists and film directors. We find that across these diverse careers, the random impact rule holds firm.
In fact, it’s not just the biggest hit that occurs at random: The second-biggest and third-biggest also hit randomly. This finding paints an unpredictable view of creativity, with an outsized role of chance in individual success. If our careers are indeed like lotteries, should we just keep drawing and hope for the best?
Given when someone produced their best work, when would their second-best work be? They’re just around the corner. Creative careers are characterized by bursts of high-impact works clustered together in sequence. Hot streaks are ubiquitous: For each domain we studied, about 90 percent of individuals had at least one hot streak.
We also found that hot streaks usually last for short periods. Moreover, the timing of the hot streak is random. Hence while periods of relative success were common, there was no way to predict when these would emerge in a given career, in line with the random impact rule uncovered earlier. Unexpectedly, hot streaks were not associated with greater productivity. Thus we don’t produce more during hot streaks than we typically would, but what we create is substantially better than our remaining body of our work.
But perhaps the most important—and uplifting—implication is for the individual innovators out there striving to make their mark on the world. The conventional view is that an individual’s best work will likely happen in their 30s or 40s. Our findings indicate that the hot streak may emerge with any work you put out, resulting in a near-term cluster of relative successes. Your big break, it appears, may arrive at any time in your career.
Dashun Wang is an associate professor at the Kellogg School of Management, Northwestern University.
