This paper questions the assumption that performance measures inevitably lead to gaming behaviours. Using (and extending) Patrick's Perverse Policing Model, the rise and fall of out of court disposals (OOCDs) from 2000-2020especially penalty notices for disorderis assessed in light of the introduction and removal of the offences brought to justice target (in place from [2002][2003][2004][2005][2006][2007][2008] and the Best Use of Stop and Search Scheme (BUSSS, in place from 2014). All four forms of gaming set out in the Perverse Policing Model were found with regard to the offences brought to justice target and a fifth form, materialising, was added to the model. However, gaming was not apparent in the use of OOCDs post-BUSSS. Reasons for this difference are considered. Namely, distinction is drawn between performance targets and performance measures. It is argued that all performance measures begin life as presentational rules. They either remain as such or become enabling (encouraging gaming) or inhibitory (encouraging compliance) depending on which response attracts the least 'within the job' trouble. This paper draws on national statistics on OOCD use from 2000-2020, a case study on one force's use of PNDs from 2010-2011 and exploratory data on positive outcomes following stop and search from July 2015-December 2020. These varied datasets allow for a detailed consideration of how these powers have been used in practice, the various opportunities afforded for gaming by OOCDs and an exploration of why those opportunities are, and are not, realised.