While the term “predictive policing” has been in use for years, the growth of our software and reliance upon it has given us insight into common myths regarding its definition. This post examines 5 myths about predictive policing, discovered through recurring questions that relate to predictive policing using PredPol’s technology.
Predictive Policing is Just CompStat’s “Rapid Deployment” Principle
Much of modern law enforcement strategy and technology is based on the CompStat process, which enforces principles of tracking and rapidly responding to crime in an organized manner. Predictive policing is not simply faster reactive policing where patrol officers responding to a crime receive information in the field from many different sources very quickly. Although predictive policing owes its roots to the CompStat process, its goal is to prevent crimes from happening, not responding to them faster.
Predictive Policing is the same as HotSpot Mapping
Predictive policing is more than traditional hotspot mapping that simply maps past crimes to make judgments about future crimes. While hotspots and heatmaps are a minimally acceptable way to allocate limited police resources, it is only half as accurate as actual forecasting technology that includes high level mathematics, machine learning, and proven theories of crime behavior. In fact, a 6 month randomized controlled trial found that crime analysts using PredPol technology in addition to their existing tools are twice as effective as experienced crime analysts using hotspot mapping alone.
Crime Predictions Will Guarantee an Incident or Arrest
Predictive policing does predict that a crime will happen in the microplace prediction areas, but provides the locations that are at highest risk of a crime occurring during the prescribed window of time. The presence of police officers in the prediction areas creates a deterrence and suppression effect, thus preventing crime in the first place.
Predictive Policing is Profiling
Because PredPol does not collect, upload, analyze or in any way involve any information about individuals or populations and their characteristics – PredPol’s software technology does not pose any personal privacy or profiling concerns. The algorithm uses advanced mathematics and computer learning to generate predictions using only three pieces of data – type, place, and time – of past crimes. Patterns inherent in the crimes themselves provide ample information on where and when crimes are likely to occur in the future. For example, if one house on a block is burglarized, there is a higher likelihood that other houses in the neighborhood will also be at risk for burglary.
Predictive Policing Requires a Major Investment for Agencies
Law enforcement often assumes that this cutting edge technology is very expensive, requires new hiring, and means big hardware costs. None of this is needed. PredPol can be set up to work with a Department’s own technology within weeks and generates actionable predictions for each shift in one tap or click. PredPol information can be printed or accessed securely through computers or mobile devices in patrol cars and in the field. The annual subscription cost is straightforward, based on population served rather than amount of crime or number of users. Law enforcement agencies using PredPol have no new hardware to buy, no new hires to make, and no limit on the number of law enforcement users or updates.
For more information about PredPol, find out how it works or contact us anytime.