Since 1974, the UK has followed a risk-based approach to safety that, in the event of an incident, is enforced through the courts. The legislation is intentionally non-prescriptive and thus requires duty holders and the courts to decide what control measures were reasonable in the circumstances from ex ante and ex post positions. This has proved challenging for all parties involved. This paper describes a series of cases that have shed light on the thought processes of the courts. It appears that there is some variability in decision-making that can be attributed to several factors, including understanding of the word “risk”, the acceptability of residual risk and the validity of historical data as a means of measuring risk. In the aftermath of incidents, there is a real danger that psychological factors may intervene when the prior risk is being assessed for sentencing purposes. It is argued that while the risk-based model continues to enjoy widespread support in the UK and is not challenged, its implementation could be much eased by attention to details. This would serve to simplify courtroom debates, support those practising risk-based regulation and enable risks to be better prioritised.