The Data-Driven Training of Regan Smith: How Technology Sharpens Elite Performance

Regan Smith has established herself as one of the most accomplished American swimmers of her generation, with multiple Olympic medals and world records in the 200-meter butterfly and 200-meter backstroke. Behind her success lies a meticulously engineered training ecosystem powered by cutting-edge technology. From wearable sensors to advanced video analysis, Smith leverages a full suite of digital tools to monitor every aspect of her swimming performance. This article explores the specific technologies she uses, the metrics she tracks, and how data-driven decision-making has helped her sharpen her competitive edge.

The Tech Stack Behind Regan Smith’s Training

Elite swimmers like Smith do not rely on guesswork. Instead, they build a personal technology stack that collects, processes, and presents actionable data. Smith’s stack includes wearable devices, underwater cameras, heart rate monitors, and cloud-based data platforms that aggregate everything into a single view. This integrated approach allows her and her coaching team to see the full picture of her performance in real time.

Wearable Sensors and Smart Apparel

During every pool session, Smith wears a combination of sensors that track physiological and biomechanical data. The most prominent is a smartwatch or fitness tracker (often a Garmin or Apple Watch) that records lap times, distance, stroke rate, and heart rate. Some of her training sets also use Polar heart rate straps for more accurate cardiac data. Additionally, she sometimes wears a Whoop strap on her wrist or bicep to monitor recovery metrics such as heart rate variability (HRV), resting heart rate, and sleep quality. This 24/7 tracking helps her understand how training load affects her body outside the pool.

Another cutting-edge device in her arsenal is the FORM Swim Goggles. These smart goggles display real-time metrics directly in her line of sight—including split times, stroke rate, and distance—so she can adjust her effort without glancing at a poolside clock. The goggles also log data automatically, syncing to a companion app after each session for later analysis.

Smith also experiments with wearable muscle oxygen sensors that use near-infrared spectroscopy (NIRS) to measure oxygen saturation in her muscles during high-intensity sets. This data helps her coach determine when her muscles are approaching fatigue and whether to adjust the training load on the fly.

Underwater Video Systems

Smith’s coach, Mike Parratto, relies heavily on underwater and above-water video to dissect her technique. Multiple GoPro cameras are positioned at various points along the pool—both below and above the surface—to capture every angle. Slow-motion playback (often at 120 fps or higher) reveals subtle flaws in body position, arm entry, and kick timing that might be invisible to the naked eye. Smith and her coach review these recordings together, often frame by frame, to identify inefficiencies and make immediate corrections.

More recently, the team has incorporated AI-powered video analysis software that automatically flags deviations from Smith’s optimal technique. The software overlays skeleton tracking and joint angles, quantifying the depth of her arm pull or the angle of her head during breathing. This reduces manual review time and provides objective benchmarks for improvement.

Data Aggregation and Visualization Platforms

Raw data from multiple devices would be overwhelming without a centralized system. Smith uses a combination of cloud-based dashboards and custom software to merge data from her wearables, goggles, and video analysis. Many elite teams employ solutions similar to Directus—an open‑source headless CMS and data platform that can integrate with APIs from wearables and video systems. By having a single source of truth, Smith can spot correlations between, for example, changes in her stroke rate and fluctuations in HRV, allowing her to adjust her training intensity proactively.

These platforms also enable her coach to set automated alerts. If Smith’s HRV drops below a certain threshold two days in a row, the system sends a notification recommending a lighter training day. This closes the loop between data collection and daily decision-making.

Key Performance Metrics Tracked

Data is only useful when it translates into specific metrics. Smith’s team tracks a carefully curated set of indicators that directly influence race performance.

Stroke Efficiency: Stroke Index and SWOLF

Two critical metrics are Stroke Index (SI) and SWOLF. Stroke Index is calculated as speed multiplied by distance per stroke; a higher SI indicates greater efficiency. SWOLF (a portmanteau of “swim” and “golf”) adds stroke count to time for a given distance—the lower the SWOLF, the more efficient the effort. Smith aims to improve her SI while reducing her SWOLF, especially during high‑intensity sets. Her wearable devices compute these values automatically, and she reviews trends week over week to ensure she is not sacrificing stroke length in pursuit of speed.

During taper periods before major competitions, Smith’s focus shifts entirely to SI and SWOLF. She drills at race pace while keeping her stroke count within a narrow band. If her SWOLF creeps up by even a point, she and Parratto identify the root cause—often a slight drop in body position or an inefficient turn—and correct it before it becomes a habit.

Tempo and Rhythm

Smith uses a tempo trainer (a small waterproof device that beeps at set intervals) to maintain a consistent stroke rhythm during endurance sets. The data from her FORM goggles also logs stroke rate variations across different phases of a race (e.g., turns, mid‑pool, finish). By analyzing stroke rate in conjunction with lap times, she can identify where she tends to accelerate or decelerate and adjust her pacing strategy accordingly.

For instance, in her 200-meter backstroke events, Smith noticed from the data that her stroke rate dropped significantly after the first turn. This indicated she was slowing her turnover while maintaining distance per stroke, but the net effect was a loss of momentum. She worked on maintaining a steady rate through the first 50 meters, and the result was a sequence of negative splits in competition.

Heart Rate and Lactate Threshold

Heart rate data is collected continuously during practice. Smith’s coach monitors heart rate zones to ensure that each set is performed at the correct intensity—whether for aerobic base building, lactate threshold improvement, or anaerobic power. Regular lactate tests (blood samples taken after specific sets) are correlated with heart rate data to fine‑tune her training zones. Over time, this data helps Smith push her lactate threshold higher, allowing her to sustain faster speeds for longer periods.

One particular insight came from correlating lactate levels with subjective Rate of Perceived Exertion (RPE). Smith found that her RPE sometimes lagged behind physiological markers: she felt fine but her blood lactate was already elevated. By relying on data rather than feeling, she could back off before entering a state of accumulated fatigue that would require extra recovery days.

Psychological Readiness: Heart Rate Variability (HRV)

Physical performance is inseparable from mental state. Smith tracks her HRV each morning using her Whoop or a similar device. HRV reflects the balance between the sympathetic and parasympathetic nervous systems. A low HRV reading often indicates fatigue or stress, signaling that a lighter training day or extra recovery is needed. By respecting these signals, Smith reduces her risk of overtraining and ensures she is mentally sharp for high‑stakes competitions.

Smith also uses HRV to gauge her response to travel and competition stress. Before international meets, she tracks how long it takes her HRV to stabilize after a flight. If it remains depressed, she extends her acclimation period. This data-driven approach to jet lag management has become a key part of her race preparation.

Turning Data into Action

Collecting data is only half the battle; the real value comes from how it is used to modify training and racing strategies.

Real‑Time Adjustments During Practice

During a practice set, if Smith’s FORM goggles show her stroke rate drifting above her target while her lap times stagnate, she immediately focuses on lengthening her stroke. Her coach also monitors a tablet displaying live heart rate and pace data. If her heart rate is spiking earlier than expected, they may lower the intensity of the next interval to prevent early fatigue. This closed‑loop system turns every practice into a dynamic, responsive session rather than a fixed, one‑size‑fits‑all plan.

An example of this occurred during a pre-Olympic training camp. Smith was doing a set of 10×200 meters on a tight send-off. Halfway through, her stroke rate began climbing while her SI dropped. The data suggested she was compensating for fatigue with increased turnover rather than maintaining efficiency. Parratto immediately called an audible, reducing the send-off interval and adding 30 seconds of extra rest between repeats. The remaining repeats were executed at a higher quality, and Smith finished the session without accumulating unnecessary fatigue.

Post‑Session Analysis with Coaches

After each session, Smith and Coach Parratto review a combined dashboard that overlays video clips with numerical metrics. They highlight specific laps where her body position dropped (detected via video angle) and cross‑reference those laps with a dip in Stroke Index. They then create a targeted drill set for the next session. For example, if the data shows her head lifts slightly during breathing, they might add a “hip‑up” drill to reinforce alignment. This continuous iteration is a hallmark of elite training.

The video analysis also captures turn technique. Smith’s underwater camera setup includes a dedicated camera at the wall that records her flip turns from multiple angles. By measuring the time from her feet hitting the wall to her breakout (commonly called "wall time"), she and Parratto have shaved tenths of a second off her turns. Data from the video shows that even a 0.1-second improvement per turn across 50 meters translates into a meaningful race advantage.

Long‑Term Trend Analysis and Periodization

Smith’s data is also used for macro‑level planning. Over a season, she and her team look at trends in HRV, average heart rate during threshold sets, and stroke efficiency declines during heavy training blocks. This informs periodization—strategically alternating high‑volume and high‑intensity phases with recovery weeks. By analyzing historical data, they can predict when Smith is likely to peak and adjust her taper accordingly. For instance, if her resting heart rate drops and HRV stabilizes during a taper, it’s a strong indicator she is ready to race at her best.

The system also tracks cumulative training load using a metric called Training Stress Score (TSS). By monitoring TSS alongside HRV, Smith’s team can identify when she is approaching an overtraining threshold. They have developed a personalized “red line” that, if crossed, triggers an automatic recovery day. This data-driven approach has reduced the frequency of minor illnesses and overuse injuries that once interrupted her training cycles.

The Role of Goal Setting and Visualization

Technology also supports Smith’s mental preparation. She uses a goal‑setting app to define specific, measurable targets for each training cycle—e.g., “Hold a 1:02 pace for 15×100 m on 1:30 send‑off by the end of March.” The app shows a progress bar and provides visual feedback when she meets milestones. Additionally, she uses biofeedback during visualization sessions: while imagining a perfect race, she wears a heart rate monitor, and her goal is to keep her heart rate low and consistent, reflecting calm focus. If her HR spikes during the visualization, she repeats the exercise until she can maintain control.

Smith also practices “data-driven mental rehearsal” by reviewing race simulations on her dashboard. She watches video of a simulated race while the system overlays her real-time biometrics—heart rate, pace, stroke rate. This helps her associate specific physical sensations with optimal performance. Over time, she learns to replicate those sensations during actual competition.

Outcomes: How Tech Has Elevated Her Performance

The quantitative approach has paid off for Smith. Since integrating this technology suite, she has seen measurable improvements in key areas:

  • Faster reaction times: By analyzing start data from pressure‑sensitive starting blocks, she shaved 0.05 seconds off her take‑off. The blocks measure force exerted by each foot, allowing her to optimize her starting position.
  • Improved underwater dolphin kicks: Video analysis led to a more streamlined body position, increasing her breakout speed by 0.3 m/s. She now maintains a tighter streamline off each wall.
  • Better pacing: Real‑time feedback eliminated the common amateur mistake of going out too fast; she now negative‑splits her 200 m events consistently. Post-race data shows she is typically 0.5 to 1.5 seconds faster on the second half.
  • Reduced injury risk: HRV and sleep monitoring flagged early signs of overreaching, preventing overuse injuries that had plagued her earlier career. Her shoulder issues, once chronic, have been largely eliminated.
  • Stronger mental resilience: Biofeedback training has improved her ability to stay calm under pressure. During the 2024 Olympic Trials, her heart rate remained within 5 bpm of baseline during the warm-up for the 200 m butterfly final.

Her world‑record performances are, in part, a testament to the precision that data‑driven training enables. But Smith is quick to emphasize that technology is a tool, not a replacement for hard work. The data merely guides her efforts; she still must execute in the pool.

The Future of Tech in Swimming

Technology in swimming is evolving rapidly. Smith is already experimenting with smart‑pool systems that automatically log every lap without requiring a wearable—using cameras and computer vision. These systems can track swimmer position in 3D and provide instantaneous feedback on distance per stroke and body roll. Meanwhile, her team explores AI‑driven video analysis that can automatically flag technical flaws and suggest corrections, reducing manual review time.

Another emerging area is predictive analytics. By feeding years of training data into machine learning models, Smith’s coaches can simulate different race strategies and predict optimal pacing for specific competitions. This allows them to fine-tune race plans before ever stepping on the blocks.

For fans and aspiring athletes, many of the tools Smith uses are becoming more accessible. Entry‑level smart goggles and heart rate monitors are now affordable, and open-source platforms like Directus can help coaches of any scale integrate data from multiple sources. While few will reach Olympic heights, anyone can adopt the same systematic approach to turn raw numbers into steady improvement.

Regan Smith’s success illustrates that the integration of technology into training is not about replacing intuition—it is about augmenting it. By combining world-class work ethic with a data-driven feedback loop, she has set a new standard for what elite swimmers can achieve.

Further Reading and Resources