Research into other sports has shown that excessive levels of anxiety can undermine motor function during performance. This study aims to investigate the effects of anxiety and self-confidence on equestrian performance. Forty riders (12 male, 28 female; 15 elite, 25 non-elite; 12 dressage, 17 showjumping, 11 eventing) completed the Revised Competitive Sport Anxiety Inventory 2 (CSAI-2R), which measures the levels of somatic and cognitive anxiety (arousal) as well as levels of self-confidence. Two-way between-subjects MANOVA tests were used to examine competence-by-discipline interactions and gender-by-discipline interactions in CSAI-2R scores. Post hoc analysis was conducted using one-way univariate ANOVA tests. Spearman's rank correlation tests were conducted between each of the CSAI-2R subscales according to competence, discipline and gender. Most important findings include lower somatic arousal and higher self-confidence in elite compared with non-elite riders, with 15.2 ± 4.4 vs. 19.0 ± 5.0, F1,34 = 5.8, P < 0.05; 31.5 ± 4.8 vs. 25.1 ± 7.4, P < 0.01, respectively. Negative correlations between cognitive arousal and self-confidence were found among elite riders, non-elite riders, showjumpers and female riders (r = − 0.69, P < 0.005; r = − 0.41, P < 0.05; r = − 0.52, P < 0.05; r = − 0.33, P < 0.05). Greater riding-specific skills in the elite rider may result in increased self-confidence. Lower levels of somatic anxiety may further increase fine motor skills in elite riders. Practical implications are that non-elite riders would benefit from sport psychological interventions increasing levels of self-confidence and reducing symptoms of somatic arousal to improve performance.