Background: Big data, or datasets with large numbers of variables and observations, may be analyzed to reveal previously unknown patterns. This is especially helpful when attempting to identify predictors of rare outcomes, such as suicide ideation (SI) and attempt (SA), and nonsuicidal self-injury (NSSI). Method: The present study used recursive partitioning, a type of machine learning, on the American College Health Association National College Health Assessment (ACHA; n = 5,313) data to identify from 100s of possible predictors salient predictors of these outcomes. Results: Among females who reported SI but not SA, one of the salient protective factors against ever engaging in NSSI was the Flourishing Scale item “I lead a purposeful and meaningful life.” Participants who responded either “agree” or “strongly agree” had lower risk of ever having engaged in NSSI. Among males with SI, higher total score on the Flourishing Scale, which indicates lower psychological well-being, were associated with a greater risk for reporting ever having engaged in NSSI. Conclusion: Advanced analytics on big data can identify previously unknown risk and protective factors for SI, SA, and NSSI. Flourishing may serve as buffer against self-injurious behavior, especially among those considering suicide. Interventions that enhance Flourishing should be developed.