Fashion trends may feel unpredictable but—according to new research—they follow a surprisingly consistent mathematical pattern.
Scientists from Princeton and Northwestern universities analyzed more than 35,000 images of women’s clothing dating back to 1869 and found that styles tend to rise in popularity, fall out of favor and then return roughly every 20 years—lending scientific backing to the long‑held “20‑year rule” often cited by fashion insiders.
The study, which examined features such as hemlines, necklines and waistlines, used a newly developed mathematical model to track how fashion evolves over time.
The researchers say that the findings do not just confirm popular beliefs about trend cycles but also offer insight into how ideas spread and resurface in society more broadly.
“Fashion insiders have talked about this for decades, but this is the first time we’ve been able to demonstrate it quantitatively,” said paper author and applied mathematician Emma Zajdela of Princeton University in a statement.
“The cycle we uncovered in the data—about 20 years—matches industry knowledge almost exactly.”
Zajdela presented the research at the American Physical Society’s Global Physics Summit in Denver, Colorado on March 17. According to the team, the findings are an exemplar for how ideas circulate in society more broadly.
To conduct their study, the team assembled one of the most extensive quantitative datasets of fashion ever created. Drawing from historical sewing patterns housed in the Commercial Pattern Archive at the University of Rhode Island as well as modern runway collections, the researchers analyzed tens of thousands of garments spanning more than 150 years.
Using custom‑built tools, the team measured specific physical features of dresses—including where the hemline, neckline and waistline fell on the body—transforming visual designs into numerical data.
The team developed a mathematical model which revealed that fashion trends behave like waves: styles grow popular, saturate the market, decline and eventually resurface. Bell bottoms offer a textbook example—surging in the 1970s, disappearing in the 1980s, returning in the late 1990s and early 2000s, and now creeping back into wardrobes once again.
To conduct the study, researchers compiled one of the most extensive quantitative fashion datasets ever assembled, drawing from historical sewing patterns and runway collections dating back more than 150 years. Using custom tools, they transformed visual designs into numerical data that could be analyzed decade by decade.
The model itself is based on a familiar tension: the desire to stand out while still fitting in. Once a style becomes too common, designers and consumers move away from it — but not so far that the clothing feels impractical.
“Over time, this push to be different from the recent past causes styles to swing back and forth,” said Daniel Abrams, a professor of engineering sciences and applied mathematics at Northwestern. “The system intrinsically wants to oscillate.”
One of the clearest examples appears in skirt length. Over the last century, hemlines have repeatedly shortened and lengthened—from flapper dresses in the 1920s to longer silhouettes in the 1950s, followed by the miniskirts of the late 1960s.
But the researchers also found that fashion cycles have become messier in recent decades. Since the 1980s, the data show multiple hemline lengths existing at the same time, suggesting that trends are fragmenting rather than converging on a single dominant look.
“In the past, there were two options—short dresses and long dresses,” Zajdela said. “Now there are many options at once. There’s more variance and less conformity.”
That fragmentation may explain why today’s fashion revivals—including the return of bell bottoms—feel less universal than they once did. Trends still come back, the math suggests, but they now share the spotlight with many others.
So if you’re wondering when a long‑retired style might make its comeback, the answer appears to be simple: about 20 years—give or take a hemline.
Do you have a tip on a science story that Newsweek should be covering? Do you have a question about fashion? Let us know via science@newsweek.com.
Reference
Zajdela, E., Abrams, D., Caticha, A., White, J., & Kohlberg, E. (2026, March 17). Back in Fashion: Modeling the Cyclical Dynamics of Trends. APS Global Physics Summit, Denver, Colorado. https://summit.aps.org/smt/2026/events/MAR-J62/6