Measured in Motion: How Science Is Redefining Rider Skill from Dressage to the NFR

New research across multiple disciplines — from the NFR to dressage biomechanics — shows that equestrian performance can be quantified, analyzed, and improved far beyond traditional subjective observation.

 If you’ve been watching the Wrangler National Finals Rodeo and thinking, “Wow, these rides look like pure magic — or pure chaos,” science would like a word.

A growing wave of equestrian research is proving that performance in our sport — whether it’s a shoulder-in in a dressage arena or an eight-second bareback ride under the bright lights of the Thomas & Mack — isn’t just luck, vibes, and whatever mood your horse (or bronc) woke up in. From sensor-based systems that quantify the rider–horse partnership, to machine-learning models predicting gait scores, to studies showing how stress, asymmetry, and biomechanics influence outcomes, researchers are dissecting the very things we often chalk up to unpredictability.

And, as the Wrangler NFR reminds us every year, rider skill, animal performance, and environmental factors all collide to separate “good ride” from “great ride.” The data is clear: success can be measured, trained, and understood — and it’s time we lean into that.

In the article “Equimetrics – Applying HAR principles to equestrian activities,”  the authors introduce a sensor-based system that applies human activity recognition (HAR) methods to equestrian sports. By placing wearable inertial sensors on both the rider and the horse, the system collects detailed motion data to objectively analyze rider posture, horse movement, and their interactions. This allows researchers and trainers to distinguish between rider skill and horse behavior, providing a more data-driven understanding of performance beyond subjective observation. The system can classify gaits, detect subtle posture changes, and optimize training and competition outcomes, making it a significant advancement for performance measurement in equestrian activities.

According to the article, Rider Skill Affects Time and Frequency Domain Postural Variables When Performing Shoulder‑in: Using inertial sensors placed on both horse and rider, researchers compared eight novice vs. eight advanced rider–horse pairs performing sitting trot (straight line) and “shoulder‑in” maneuvers (left and right). The study found that advanced riders had objectively different biomechanical signatures than novices: e.g. greater hip extension and outside‑leg external rotation, especially during shoulder‑in, which corresponded with higher judge‑assigned scores. This shows that rider skill is measurable and has a tangible effect on posture and motion independent of horse variability; providing evidence against the idea that performance is purely luck with an animal.

The study, “Money Bull: Analyzing the Application of Ranking Methods to Rodeo” examines the ranking system for bareback riders in the Professional Rodeo Cowboys Association (PRCA), arguing that total earnings alone may not accurately reflect rider skill. The authors apply classical linear algebraic ranking methods (Colley, Massey, Keener, and PageRank) to PRCA performance data, finding that each method highlights different aspects of performance, such as average earnings, total score, and rider score. The study suggests that a more nuanced, holistic ranking system can better distinguish rider skill from variability caused by external factors, like the performance of the animals or prize pool inconsistencies.

In the review, Potential Effects of Stress on the Performance of Sport Horses, the authors survey literature on how stress, from environment, training, competition, and handling, influences physiological and behavioral traits of sport horses, such as temperament, gait quality, and overall performance. They argue that stress responses may bias performance: while short-term stress might occasionally enhance alertness or performance, chronic or repeated stress (from poor management, frequent competition, or unsuitable environments) tends to impair performance and reduce the reliability of the horse as a constant. Hence, what might be dismissed as “unpredictability” due to the animal may in part be explained by measurable stress and environmental variables, which, if controlled or recorded, could reduce noise and clarify true performance consistency.

In the study, Supervised Machine Learning Techniques for Breeding Value Prediction in Horses: An Example Using Gait Visual Scores, researchers applied machine‑learning methods, artificial neural networks (ANN), random forest regression (RFR), and support vector regression (SVR), to predict breeding values for visual gait scores in a population of over 5,000 gait evaluations in Brazilian gaited horses. They found that the ML models achieved accuracy comparable to traditional multiple‑trait models (MTM), though with slightly increased bias and dispersion; still, the results show that ML techniques are a feasible alternative for dealing with subjective measurements and complex, non-linear traits. This suggests that even when animal performance traits are influenced by many variable factors (like gait, comfort, style), data-driven statistical and machine‑learning approaches can extract meaningful, repeatable patterns, which undermines the notion that animal unpredictability makes skill or quality unmeasurable.

In the article, “The Effect That Induced Rider Asymmetry Has on Equine Locomotion and the Range of Motion of the Thoracolumbar Spine When Ridden in Rising Trot,” the authors explore how asymmetry in the rider, uneven posture or uneven guidance, affects the horse’s locomotion and spinal motion during rising trot. They note that rider asymmetry can interfere with the rider’s aids and the horse’s movement symmetry, potentially degrading performance or causing suboptimal results. This indicates that variability attributed to the horse might sometimes stem from the rider, implying that skill (or lack thereof) in riders can significantly influence outcomes, which again supports the idea that performance is not purely random or unpredictable.

The authors of the review, “Psychological Factors Affecting Equine Performance,” discuss how temperament, mood, stress, emotional state, and other behavioral/psychological variables in horses influence their performance across different disciplines. They argue that horses are not just machines: their welfare, psychological state, and environmental conditions (like confinement, social isolation, training load) significantly shape their athletic output. Thus, variability in horse performance (which some attribute to unpredictability) may reflect measurable psychological or welfare-related factors, again opening the door to objective assessment rather than labeling outcomes as random.

Relying on the animal alone cannot explain performance outcomes, and measurement, data analysis, and rider training remain essential. Also, the horse’s stress level, environment, temperament, and the rider’s skill all play critical roles in determining performance consistency. Collectively, the findings show that evaluating equestrian performance requires a multi-layered approach; rider skill can be quantified, but only when combined with careful monitoring of the horse, objective data collection, and appropriate analytical methods.