Despite the marked increase Ecological Momentary Assessment (EMA) research, few reliable and valid measures of momentary experiences have been established. The goal of this preregistered study was to establish the reliability, validity, and predictive utility of the momentary Pain Catastrophizing Scale (mPCS), a 3-item measure developed to assess state pain catastrophizing. Participants in two studies of postsurgical pain outcomes completed the mPCS three to five times per day prior to surgery (N = 496, T = 20,385 total assessments). The mPCS showed good psychometric properties, including multilevel reliability and factor invariance across time. Participant-level average mPCS was strongly positively correlated with trait pain catastrophizing (r = .55 and .69 in Study 1 and Study 2, respectively) and related constructs. Using dynamic structural equation modeling, we then examined whether this dynamic measure improved prediction of postsurgical pain outcomes above and beyond one-time assessment of trait pain catastrophizing. Indeed, greater average mPCS scores (β = 0.30, p = .01) and greater variability in momentary pain catastrophizing (β = 0.34, p < .01) prior to surgery were associated with increased pain immediately after surgery, after controlling for trait pain catastrophizing. Greater average mPCS score prior to surgery was also a small but significant predictor of lesser day-to-day improvement in postsurgical pain (β = 0.27, p = .02), whereas trait pain catastrophizing was not. These results show that the mPCS is a reliable and valid tool for EMA research and highlight its potential utility over and above trait measures of pain catastrophizing.