<p>Many human interactions may be thought of as a series of anticipations of what others will do and the resultant tailoring of behaviour. This paper looks at police-public interactions in terms of actions relative to expectation and the implications for applicable research. The decision of victims not to report a crime is used to illustrate the approach. Specifically, the proportion of decisions to report a crime that were police related, changes in this trend over time as well as reasons for it are analysed. Analysis of responses to the Crime Survey for England and Wales between 2001 and 2023 suggests that anticipated police reactions feature in between 40 and 52% of decisions not to report. Such reasons include anticipated police impotence; ‘could do nothing,’ or police disinclination to act ‘would not be bothered/interested’. Some victims tried but failed to contact police. Non-report because of bad experiences in previous encounters with the police was rare. The pattern was consistent across crime types. The rated seriousness of crimes not reported for police-linked reasons was significantly higher than those not reported for other reasons. This has implications for current allegations of ‘two-tier policing’. The discussion section argues for abandoning conventional attitude surveys of public views about the police. Novel approaches to police-public interaction are outlined.</p>