Predictive Policing and Prevention Fit

This article develops a theorem-driven criminological analysis of predictive policing through the concepts of prevention fit and prevention misalignment. Building on the crime-process framework developed in Crime as a Mediated, Gated, and Path-Dependent System, it argues that predictive systems should not be evaluated solely by predictive accuracy or operational performance, but by the degree of alignment between predictive signals, institutional translation, preventive intervention, and the level of the crime process at which intervention is directed. Rather than treating predictive policing as inherently beneficial or inherently problematic, the article develops a processual diagnostic framework capable of distinguishing theoretically aligned from misaligned preventive interventions. It introduces explicit propositions for analysing level congruence, cross-level escalation, mediation displacement, and persistence substitution, thereby providing a structured basis for empirical evaluation and comparative research. Within the broader theorem-driven crime-process research programme, this paper constitutes a first-order prevention-theoretical extension of the core framework. It operationalises the concepts of prevention fit and prevention misalignment in the specific context of predictive policing and algorithmic crime prevention. It is a downstream theoretical application rather than a substitute for the programme's core crime-process theory.