DescriptionPerformance analysis (PA) involves the systematic observation and analysis of factors identified to enhance performance in a specific sport to improve athlete decision-making. PA is commonplace in human sports, yet despite potential advantages, its uptake is limited in equestrianism. This study aimed to evaluate if factors anecdotally associated with performance in elite showjumping influenced competitive success. All rounds of horse and rider combinations competing in the 2nd Round of the FEI Nations Cup 2017 competition in European Division 1 at 10 different 5* events were reviewed. Courses contained on average 15 jumping efforts (JE) and field size ranged between 18 and 23 combinations. Fences were classified by jumping effort (incremental), by type (e.g. upright, oxer), by approach line (straight vs. not-straight). Total faults and the distribution of faults for every quarter of the course were also calculated. A total of 3052 JE were reviewed (no faults: 93.6%; n=2857, faults: 6.4%, n=195; frequency: 6.39 faults per 100 JE). Fence level variables were analysed through univariable analysis to inform multivariable model building. The final multivariable model was refined using faults/no faults as the dependent variable using a backwards stepwise process, likelihood ratio: P<0.05. Combinations which completed their round above the optimum time were 1.1 times more likely to have faults for every 0.1 second they were over the time (P=0.0001). Faults were also on average 4 times more likely to occur at JE 3,4,5 and 8 in the first half of the course (P<0.03). The probability of scoring faults then increased to being 9 times more likely in the second half of courses at JE 9,10,11,12,13 and 14 (P<0.006). A straight approach to a JE reduced the chance of faults by (P=0.0001) by 48% compared to a non-straight approach. Interestingly, although fence type was not significant in the model, 49% of JE were faults occurred were upright fences and 41% were combination fences. Distribution of faults also varied significantly (ANOVA: P=0.0001) with the number of faults increasing sequentially between the 1st (n=13, mean faults=4) and 3rd (n=53, mean faults=6, LSD: P=0.03) and 4th quarters (n=83, mean faults=7, LSD: P=0.0001), the 2nd (n=43, mean faults=5) and 4th quarters (LSD: P=0.0001) and 3rd and 4th quarters (LSD: P=0.03). Understanding the impact of factors that influence horse and rider performance can inform training and competition strategies. These preliminary results suggest patterns exist within fault accumulation in elite showjumping, supporting the application of PA in equestrian sport.
|Event title||26th Equine Science Symposium|
|Location||Asheville, United States, North Carolina|
- performance analysis
- training strategies