Stressors can have negative effects on well-being, but little is known about how an individual's inability to precisely forecast upcoming stress could be a risk factor for well-being. Antecedents and outcomes of two stress forecasting variables, anticipated stress level and underestimation errors in stress forecasting (operationalized by the residual change scores obtained by regressing the evening experienced stress on the morning anticipated stress), were investigated. In a daily diary study of 110 undergraduate students over a workweek, poor sleep quality and negative affect reported in the morning predicted a higher anticipated stress of the upcoming day. Poor sleep quality was found to be related to less underestimation errors (i.e., more overestimation). Mispredicting the daily stress level was found to predict greater health complaints and negative affect by the end of the day. Those high on trait resilience were found to make fewer underestimation errors on average. Worse emotional outcomes were associated with underestimation errors during stress forecasting than with overestimation errors. This study demonstrates that examining an individual's experience in forecasting upcoming stressors is an important area for future research in determining points of intervention to promote adaptive management of daily demands.