Predictive Policing: Stopping Crime Before it Starts

February 20, 2019

By Jill K. Sanders, Esq.

Nearly two decades into the twenty-first century, it’s clear that computers have changed nearly every aspect of our lives. When it comes to crime, the age of technology has impacted each part of the criminal justice system. Predictive policing is one of the newer ways law enforcement is harnessing the power of technology to fight crime.

What is Predictive Policing?

Predictive policing is a broad term used for techniques used by law enforcement to identify potential criminal activity. It falls into four general categories:

  1. Methods for predicting crimes
  2. Systems for predicting offenders
  3. Methods for predicting perpetrators’ identities
  4. Techniques for predicting victims of crime

While not necessarily telling what will definitely happen in the future, predictive policing identifies where crime is likely to happen. This is not meant to replace more traditional forms of policing but rather to be used in conjunction with other proactive strategies.

How Does The Software Work?

A popular program, PredPol, is already being used by police forces across the nation. PredPol has police dedicate their resources towards petty crime. The program generates place-specific crime forecasts to areas as small as 500 by 500 square feet. The forecasts can include when crime is more likely to occur as well as a prediction on who may commit the offense.

“PredPol has a precise definition of predictive policing. For us and our customers, it is the practice of identifying the times and locations where specific crimes are most likely to occur, then patrolling those areas to prevent those crimes from occurring. Put simply, our mission is to help law enforcement keep communities safer by reducing victimization.”

PredPol assumes similar crimes are more likely to occur in the same place in the future. PredPol uses anywhere from three to 10 years of crime data, and runs it through the program’s algorithm. The forecasts are generated based on data relating to prior crimes. This includes data on when, where, and by whom prior crimes were committed.

How Does Predictive Policing Help Communities?

The goal of predictive policing is to stop crime before it happens. Results are varied as to the effectiveness of software like PredPol, but the data collected thus far is promising. The Los Angeles Police Department (LAPD) found predictive policing software to be twice as accurate as its current practices. Another jurisdiction in California also saw a 19 percent drop in burglaries over a 6-month period after its police implemented predictive policing. And in the United Kingdom, one jurisdiction found that PredPol predicted the location of street crime better than police analysts. Further studies are need to determine the true efficacy of the software.

Can It Impact Civil Liberties?

One criticism of this software is that it is based on the “broken windows” theory that has been widely criticized as discriminatory. “Broken windows” theory is based on the idea that if police crack down on minor crimes to create an atmosphere of safety and lawfulness, more serious crimes will be prevented. Often times, those arrested for minor crimes are persons of color and those from poorer communities.

Another criticism is that because software like PredPol is based on historical crime data, it may continue to perpetuate bias that previously existed. For example, if a police department over-policed certain communities in the past, the software will be more likely forecast crimes in those same areas.

Finally, there is a privacy concern for the data storage – particularly as it pertains to individuals. Will it be used in other ways? Is it being stored securely? Police should use caution when using and sharing information about possible offenders and victims.