The study was designed in order to gain a better understanding of whether the lack of competition stress and/or sampling time had an influence on circulating β-endorphin, adrenocorticotrophin (ACTH) and cortisol modifications after experimental showjumping sessions and to study the effects of fence height on hormone changes. Hormone levels were recorded before exercise in basal conditions and after warm-up, then 5 and 30 min post-exercise. Using a randomized crossover study design, six horses were studied during three experimental showjumping sessions over fences of different heights: 1.00 m (session 1), 1.10 m (session 2) and 1.20 m (session 3). The showjumping exercise did not modify plasma β-endorphin and serum ACTH concentrations after session 1, and tended only to maintain higher values than basal after both session 2 and session 3. The interaction fence height/time was not statistically significant for either β-endorphin or ACTH changes. Sampling time significantly affected both β-endorphin (F = 2.88; P < 0.04) and ACTH (F = 3.84; P < 0.01) changes. Serum cortisol levels were always higher than basal 5 min post-exercise, with levels falling at 30 min. The interaction fence height/time was not statistically significant, while sampling time significantly affected the results (F = 7.96; P < 0.0002). This study demonstrated no significant effects of fence height on β-endorphin, ACTH and cortisol changes. The sampling times adopted affected post-exercise changes in plasma β-endorphin, ACTH and cortisol and could have masked the effects of fence height on hormone modifications.