Experience sampling method (ESM) studies have found an association between daily stress and paranoid symptoms, but it is uncertain whether these findings generalize to physiological indicators of stress. Moreover, the temporality of the association and its moderating factors require further research. Here, we investigate whether physiological and self-rated daily stress predict subsequent paranoid symptoms and analyze the role of emotion regulation as a putative moderator. We applied ESM during 24 h to repeatedly assess heart rate, self-rated stress, and subclinical paranoia in a sample of 67 psychosis-prone individuals as measured with Community Assessment for Psychotic Experiences (CAPE). Adaptive and maladaptive emotion regulation was assessed at baseline with the Emotion Regulation Skills Questionnaire (ERSQ-ES) and the Cognitive Emotion Regulation Questionnaire (CERQ). Linear mixed models were used to analyze the data. Heart rate (b = 0.004, p < 0.05) and self-rated stress (b = 0.238, p < 0.001) predicted subsequent paranoia. The reverse effect, paranoia as a predictor of subsequent heart rate (b = 0.230, p = 0.615) or self-rated stress (b = –0.009, p = 0.751) was non-significant. Maladaptive emotion regulation was a significant predictor of paranoia (b = 0.740, p < 0.01) and moderated the path from self-rated stress to paranoia (b = 0.188, p < 0.05) but not the path from heart rate to paranoia (b = 0.005, p = 0.09). Our findings suggest a one-way temporal link between daily stress and paranoia and highlight the importance of emotion regulation as a vulnerability factor relevant to this process.