Skip to main content

‘Digital Twins’ Give Olympic Swimmers a Boost

In the Paris 2024 summer Olympics, swimmers will be guided by their digital twin. Here’s how they work to help the fastest swimmers break records

Underwater photo of Kate Douglass competing

Kate Douglass of Team United States competes in the Women's 200-meter Individual Medley at the Doha 2024 World Aquatics Championships on February 11, 2024, in Doha, Qatar.

Adam Pretty/Getty Images

In July sports fans worldwide can see the world’s fastest swimmers launch from the starting blocks into the Olympic-size pool at the Paris La Défense Arena in Nanterre, France, a western suburb of Paris. For the Olympians, the chance to compete in these games will be a dream come true.

Turning Olympic dreams into reality requires years of total commitment, with only the very, very few making it. Only the nation’s top 60 to 80 athletes will receive an invitation to the Olympic trials for each swimming event. And only the four best performers in those trials—two in the men’s category and two in the women’s category— will make the Olympic team. Adding to these tall odds, races are often decided by mere hundredths of a second.

How should coaches prepare Olympic hopefuls? Should they instruct their athletes to swim like Katie Ledecky and Michael Phelps, with dreams of replicating their success? Definitely not. Athletes come in different shapes and sizes and have different strengths and weaknesses. The stuff of Olympians is nothing like the do-it-yourself movement of home repair, where a simple Internet search reveals the one correct answer.


On supporting science journalism

If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.


Mathematics, physics and technology have instead revolutionized swimming. The idea is that biomechanical and hydrodynamic minutiae are variables in a complex physical and mathematical problem. By optimizing these “variables,” swimmers can reach near perfection. Today the advent of sensor technology has turned this idea into a reality in which mathematics and physics produce useful information so that coaches can “precision-train” 2024 Olympic hopefuls. The results have been enormously successful.

Swimming Forces

The universal nature of Newton’s laws of motion governs not just our solar system but also the minute movements of a swimmer. When a swimmer dives into a pool and begins undulating to propel themselves forward, Newton’s laws govern the connection between the propulsive forces generated and the resulting acceleration of the swimmer’s body.

For example, in the Olympic 50-meter freestyle final, eight athletes thrash their limbs with the goal of completing one lap of the pool first. Rather than a competition between athletes, the race is each swimmer’s individual battle against the physics of inertia (described in Newton’s first law) and the force of drag (described in Newton’s second law), and they must create forces that move their body to the finish (Newton’s third law) in the hopes of a gold medal.

Digital Twin Training

This Summer Olympics will be the first time that nine of the elite swimmers will be guided by their digital twin. Since 2015 teams of researchers at Emory University and the University of Virginia, led by one of us (Ono), have been equipping swimmers with devices called inertial measurement units to record their body’s acceleration, orientation and force. Unlike typical digital video, which records 24 frames per second, these sensors capture information 512 times a second.

While the swimmers go through a battery of tests wearing these sensors on their wrists, ankles or back, the data show the impact on their acceleration from every rotation, splash, pull and kick.

Recently we started using advanced sensors that measure force generated by an athlete’s hands. These high-tech bands measure the pressure differential between the palm and the side of the hand, revealing the direction of the force. What was previously evaluated purely by looking at the swimmer above the water can now be distilled into a series of charts and graphs that show the distribution of force in all the forward, sideways, and upward and downward directions. Force applied in any direction other than forward is wasted force.

We use these streams of numbers to create an athlete’s digital twin, which captures their movements down to the millisecond. We have now assembled a massive database of digital twins from more than 100 of the best U.S. swimmers.

With such digital twins, we can make recommendations that immediately improve technique, offer suggestions for race strategy and point to long-term aspirational goals—all in pursuit of the optimal race plan.

In terms of technique, we can digitally identify an athlete’s comparative strengths and weaknesses without a live race. If we find a technical flaw, a coach can offer immediate precision training to fix it. The digital twin even quantifies the severity of a flaw. And thanks to Newton’s equations and the acceleration data, we can accurately predict the time savings that an athlete can expect with a given change. It boils down to the numerical integration of the acceleration data because these values are part of the calculations for velocity. Thank you, Newton and your calculus!

Flaws include poor head position, legs that are too low (which can cause an anchoring and slowing effect), unbalanced body rotation and inefficient breathing. Consider the execution of streamline in breaststroke, which is the underwater glide phase of the stroke. The goal is to preserve as much speed as possible off the opening dive and after pushing powerfully off the wall in exiting turns. You might think there is little opportunity for improvement in these phases of breaststroke because the swimmer seems to be doing nothing. Races can be won or lost, and records set, during this innocuous-seeming glide, however.

In November 2020 we created a digital twin of one of us (Douglass), who was then a collegiate swimmer. Though the 200-meter breaststroke was not on Douglass’s list of races because of her time, within hours of compiling her digital twin, we knew that she had both the physical ability and aerobic capacity to compete at the world championship level. We ran the simulations and then made a list of targets of opportunity if she chose to pursue breaststroke.

Photographs of Lilly King and Katherine Douglass in streamline use overlaid dotted lines to highlight how different head positions affect the swimmers’ speed.

Ken Ono (images); Amanda Montañez (lines and labels)

Head position was her key to success. Take the streamline position of 2016 Olympic gold medalist Lilly King (upper panel in image above) for comparison. Because Douglass’s head was tilted from the plane of her body (lower panel), you can see how that would introduce extra turbulence and drag. Her digital twin let us quantify the significance of this flaw. Using mathematical equations arising from Newton’s laws of motion, we predicted that she stood to gain 0.1 to 0.15 second per streamline glide by adjusting her head position. In the 200-meter breaststroke event, an athlete performs four of these streamline glides, so we predicted that this recommendation could save 0.4 to 0.6 second.

After three years of work, Douglass’s improved technique shaved 0.44 second off her 200-meter-breaststroke time. A few months later, in 2023, she broke the American record in the event with a time of two minutes, 19.3 seconds, dipping under the previous record from 2012 by 0.29 second.

Digital twins also play an important role in devising race strategy. Analysis of a twin can lead to suggested changes in tempo, the timing of body movements, the number of kicks taken in various phases and recommended breathing patterns. Should an athlete breathe on both sides in freestyle? How many breaths should they take in a 100-meter sprint?

By virtually experimenting with an athlete’s digital doppelgänger, we can easily run different race scenarios to determine that swimmer’s optimal race plan, their “formula” for success.

No two formulas are alike. When trying to improve strategies for two elite breaststrokers, we compared the acceleration of their digital twin during the first phase of a “pullout,” which takes place underwater and consists of a powerful push off the wall followed by a streamline glide and, at the end, a dolphin kick. We graphed the acceleration in the direction of the swim and found that the “orange” swimmer in the graphic below had an extraordinary streamline, with almost no deceleration. The orange breaststroker also had a weaker dolphin kick, which she executed almost one second earlier than the other swimmer. In terms of strategy, the orange swimmer might consider delaying the dolphin kick because of her superior streamline and weak kick. Meanwhile the blue swimmer decelerated significantly in the glide but showed a powerful kick. That breaststroker might want to kick earlier to mitigate the inferiority of her glide. By running different simulations, we confirmed these speculations, figured out optimal timing of each stroke phase and estimated the expected time savings to boot. Why guess?

Line chart shows acceleration patterns of “orange” and “blue” swimmers over about three seconds, from push off through dolphin kick.

Finally, this quantitative approach to swim analysis can help to formulate aspirational goals that can become reality after months or years of extensive training. Some of the desired simulations are not realistic given an athlete’s current aerobic capacity. After all, the digital twin doesn’t feel the pain of burning muscles and oxygen-starved lungs. One such goal might be the addition of another dolphin kick off the dive in the 100-meter butterfly sprint, something that requires more oxygen consumption but can shave 0.1 second off a swimmer’s time. A coach might be able to help an athlete increase their aerobic capacity by transforming that unrealistic simulation into a genuine race strategy at the highest stage of competition.

The return of the Olympics to Paris after a 100-year hiatus offers an elegant opportunity to reflect on a central pillar of the Olympics—its offering of a consistent tradition to a constantly evolving world. During the competition, the Seine will still flow, and the Eiffel Tower will still preside over many of the same events as it did a century ago. Among these stoic landmarks, however, are a city and games that have been unquestionably altered by the modern era, one full of science, electronics and an abundant supply of resources. These leaps in technology will result in equally magnificent leaps in performance, with athletic feats that would have been quite literally unimaginable 100 years ago. These athletes, armed with troves of data, refined training techniques and complex analytics, demonstrate the beauty of the games as both a driver and display of what humans and technology can achieve, redefining our common limits.

Millions will watch the swimming events unfold in Paris La Défense Arena. Many American swimmers will make Olympic history with medals and records, and for some, their digital doppelgänger will be hidden on a computer—out of sight but somehow also part of the team.

Katherine Douglass is currently working on a statistics master’s degree at the University of Virginia (UVA). She received her undergraduate degree in statistics from UVA in 2023. During her time as an undergraduate at UVA, she was a member of the swim team and was a multiple-time team and individual National Collegiate Athletics Association (NCAA) champion. Along with NCAA swimming success, she also won a bronze medal at the 2021 Tokyo Olympic Games and is a multiple-time American record holder and world champion. She will be competing in the Paris Olympics in three individual events for Team USA.

More by Katherine Douglass

Augustus Lamb has a master’s degree from the University of Virginia (UVA) School of Data Science. He received his undergraduate degree in computer science from UVA in 2023. He was on the university’s swim team for five years and competed as part of the school’s National Collegiate Athletics Association (NCAA) and Atlantic Coast Conference (ACC) teams in addition to representing the program at the U.S. Olympic Swimming Trials in 2021. He placed 10th in 50-meter freestyle at the 2024 U.S. Olympic Trials.

More by Augustus Lamb

Jerry Lu is a graduate researcher at the Massachusetts Institute of Technology’s Sports Lab. He received a master’s degree from the M.I.T. Sloan School of Management after completing his undergraduate in systems engineering at the University of Virginia. Lu has been working on performance optimization and race strategy design for elite swimmers and triathletes and has worked on analytics with AusCycling, the governing body for cycling in Australia, to help inform performance outcomes. His research interests include applied math, sports quantitative research, athlete biomechanics and financial markets. Outside of research, Lu likes sailing and tennis, and he was a semiprofessional poker player for a year.

More by Jerry Lu

Ken Ono is STEM adviser to the provost and Marvin Rosenblum Professor of Mathematics at the University of Virginia. He is also a professor by courtesy of data science and a professor affiliate in statistics. He received a B.A. in mathematics from the University of Chicago in 1989 and a Ph.D. in mathematics from the University of California, Los Angeles, in 1993. His research interests include arithmetic geometry, combinatorics, number theory and mathematical physics. He is a fellow of the American Mathematical Society and a recent recipient of the University of Chicago Alumni Medal for Professional Achievement. In the 1980s he raced bicycles professionally, and from 2012 to 2014 he competed as a member of Team USA in the World Cross Triathlon Championships.

More by Ken Ono

William Tenpas is a student at the University of Virginia (UVA), where he is earning a master’s in data science. In 2023 he graduated with a B.S.E. in mechanical engineering and materials science from Duke University, where he was also a four-year member of the swim team and two-time captain. He has spent the past year studying and swimming at UVA. He is interested in the crossover of science and engineering with sports, having previously worked for a start-up producing 3D-printed medical devices (PROTECT3D). He also won second place in his high school talent show for consuming an inordinate amount of chocolate milk onstage.

More by William Tenpas

The Mathematical Intelligencer is a quarterly journal written in an engaging, informal style for a broad audience. It features expository articles about mathematics (broadly defined), mathematicians (ditto), and the history and culture of mathematics in its intellectual, social and scientific context. Puzzles, poetry and fiction appear in its pages too.

More by The Mathematical Intelligencer