This study aimed to characterize scapular muscle timing in stroke patients with and without shoulder pain. Muscle activity of upper trapezius, lower trapezius, serratus anterior, infraspinatus, and anterior deltoid (AD) was measured (Delsys Trigno surface EMG system, USA) in 14 healthy controls (dominant side) and 30 stroke patients (hemiplegic side) of whom 10 had impingement-like shoulder pain. Participants performed 45° and full range anteflexion, in two load conditions. The impact of group, anteflexion height, load condition, and muscle was assessed for onset and offset of the different muscles relative to the onset and offset of AD, using a 3 (group) × 2 (height) × 2 (load) × 4 (muscle) mixed model design. Recruitment patterns were additionally described. Across all load conditions and groups, serratus anterior had a significantly earlier onset and, together with lower trapezius, a significantly later offset in 45° compared to full range anteflexion tasks (p < 0.001). In stroke patients without pain, lower trapezius had furthermore a significantly earlier onset in comparison to stroke patients with shoulder pain (all tasks, p = 0.04). Serratus anterior also showed a significantly earlier offset in stroke patients with shoulder pain in comparison to controls (p = 0.01) and stroke patients without pain (p < 0.001). Analysis of muscle recruitment patterns indicated that for full range tasks, stroke patients without pain used early and prolonged activity of infraspinatus. In stroke patients with shoulder pain, recruitment patterns were characterized by delayed activation and early inactivity of serratus anterior. These timing results can serve as a reference frame for scapular muscle timing post-stroke, and when designing upper limb treatment protocols and clinical guidelines for shoulder pain after stroke.