Perceptual studies suggest that the segmentation of a musical sequence is influenced by three accent structures: rhythmic grouping, melodic, and metric accent structures. We investigate whether performers emphasize these types of accents with systematic performance variations (intensity, interonset timing, and articulation). In three experiments, skilled pianists performed sequences of various musical complexities: simple sequences containing only one accent structure (Experiment 1), more complex sequences containing coinciding or conflicting accent structures (Experiment 2), and a concert pianist's performance of a sonata containing coinciding and conflicting accent structures (Experiment 3). In all three musical contexts, similar systematic performance variations were observed in relation to each type of accent. Variations corresponding to rhythmic grouping accents were most consistent across musical contexts and dominated when the accent structures conflicted. These findings suggest perceptual correlates for the accent structures in music performance that may facilitate listeners' segmentation of musical sequences.