New technologies have been developed to complement conventional interventions to better target the specific needs of people with stroke, and they have been shown to improve both function and performance. However, it is unknown whether the baseline levels of sensorimotor function and performance interrelate with the improvement in upper limb and daily performance. Thus, the aim of this study was to examine the relationship between baseline levels of sensorimotor function and daily performance and its impact on post-intervention improvement in people with stroke following a robotic intervention. A single-blind, non-randomized, controlled clinical trial was conducted. Participants in the experimental group (n = 9) received a robotic intervention in addition to conventional treatment. Sensorimotor function was measured with Semmes-Weinstein Monofilaments® and the Fugl-Meyer Assessment Upper Extremity Scale. Upper limb and daily performance were measured with the MAL and SIS-16 scales. The multivariate regression models showed that baseline levels of upper limb performance and motor function predicted >95% of the variance in upper limb performance (p < 0.001), while pre-intervention levels of daily performance explained >75% of the post-intervention variance (p < 0.05). These findings indicate that basal upper limb motor function is associated with improved performance following a combined intervention of conventional treatment and robotic intervention.