When Garry McCarthy took over the Chicago Police Department, he brought in an approach made famous by William Bratton, a precise and data-driven structure made famous in Los Angeles and New York, as Noah Isackson wrote in his Chicago profile of the city's top cop:

Bratton had been the top cop in Boston, and his crime-fighting strategies as commissioner of the NYPD in the mid-1990s made him a legend in police circles. As McCarthy explains: “He kind of changed the way we did things so that they had less to do with seniority and more to do with achievement.”

McCarthy is referring to Bratton’s use of the CompStat system, which collects data about everything from curfew violations to murders and then spits out reports that show how each district is performing. These numbers allow the police chief to hold officers and their superiors accountable for results, or lack of them, mainly through frequent no-holds-barred meetings with the command staff. “I got to show my wares at CompStat,” says McCarthy, who earned his bosses’ attention by detailing his successes. “And, quite frankly, I did very well.”

He says that much of his policing philosophy grew out of Bratton’s approach, which worked so well that the city’s subsequent police chiefs rode it to historic drops in crime. Simply put, Bratton did three things: First, he implemented CompStat. Second, he acted on the “broken windows” theory; that is, he instructed cops to crack down on minor offenses—curfew violations, loud music, graffiti, public intoxication, and so on—because any bad deed could lead to more serious criminal behavior. Third, he collapsed the layers of bureaucracy, eliminating middle managers and emphasizing the job of the beat cop. “It has to do with accountability,” explains McCarthy. “If [officers] don’t have a beat, if they’re not accountable for anything, all they’re doing is going from job to job.”

Bratton popularized the broken-windows theory, but it comes out of the social sciences—specifically, a paper by James Q. Wilson and George Kelling, and before that, the streets of Newark, where McCarthy would later run the police department. Bratton had intuitively been moving in the direction of Wilson and Kelling's idea, but after their 1982 paper he fully embraced it, and as New York's police commissioner, installed it as a crime-fighting tactic in the nation's biggest city. It's an attempt to get ahead of crime, attacking the root causes of crime as much as the people who commit it.

As Isackson details, McCarthy's east-coast offense has been controversial in Chicago, in part because it superceded an approach many (including McCarthy's predecessors) believe to have worked here—an intensely focused, highly specialized anti-gang force:

[I]t’s hard to argue with the effectiveness of specialized units. Consider what happened when Cline began using them to target gangs, guns, and drugs in certain neighborhoods on the South and West Sides. Chicago went from being the murder capital of the nation in 2003, with 601 people killed, to 453 the next year (see "Murder Capital of America?"). That was the first time since the 1960s that fewer than 500 murders had been recorded.

Homicides remained under the 500 mark until 2008. The previous fall, in the wake of ongoing scandal involving the rogue officers, Dana Starks, the interim superintendent between Cline and Weis, shelved the specialized units. Homicides began to rise. After Weis took the job, he convinced Mayor Daley in September 2008 to reinstate the units, promising that, as chief, he’d implement better training and more accountability.

According to Weis, the city’s sub-500 murder totals in 2009, 2010, and 2011 were a direct result of that decision. “Sometimes the answer is staring you right in the face,” he says.

After this violent summer, McCarthy's approach appears to be evolving, and toward a gang-oriented approach, in part because of the nature of the city and its gang problem:

“Gang violence in Chicago is something that’s totally different from what I experienced in New York and Newark,” says McCarthy, “where we had low-level bands of thugs who called themselves Bloods or Crips but were narcotics dealers.”

But like his New York training, it also evolves out of the social sciences, this time research that's closer to the streets of Chicago, as David Kennedy writes in a fascinating new Daily Beast piece:

In direct response to the street scene driving the violence, Chicago’s police superintendent, Garry McCarthy, is working with criminologists and other researchers to spearhead a new response that could represent the next major advance in how America polices serious violence.

At its core, the new approach focuses not on crime “hot spots,” the traditional target of law enforcement, but on “hot people”—the small number of individuals who account for the vast majority of the crime and murders.

One of those criminologists is Andrew Papachristos, a Loyola grad with a Ph.D. from the University of Chicago. He's currently an associate prof at UMass-Amherst and a fellow at Harvard, but the bulk of his research has been done in his hometown of Chicago. One of the problems that McCarthy has run into in dealing with Chicago's gangs is what they are now—less the established, hierarchical gangs of old, which have been undermined by his predecessors' approach, but the decentralized, disorganized, factionalized groups that are more fluid and less predictable than their predecessors. Papachristos's research into the social networks of crime lines up with this new reality: association over affiliation.

Starting with homicide and shooting victims and using readily available police data—say, who was arrested with whom—Papachristos can isolate social networks of staggeringly high risk. For example, simply being in an arrested-with network with a homicide victim increases your own chance of being murdered by 900 percent. While those with the closest links to homicide victims are at the most elevated risk, each additional “handshake” away reduces that risk by nearly 60 percent.

It's reminiscient of a chilling but sensible approach Chicago Public Schools tried recently: predicting the likelihood that a given CPS student would be the victim of violence in a short-term timespan. In 2011, Steven Levitt of Freakonomics fame, along with Dana Chandler of MIT and John List of the University of Chicago, went back and looked at their model, and found some success in predicting the fates of at-risk students:

A number of behavioral factors predict violent victimization: serious misconduct at school, suspensions, and incarceration history. The strongest predictor is having spent time in a juvenile detention center, which raises the likelihood of being shot by 2.5 percentage points—a tenfold increase over the mean probability. Other factors correlated with victimization are being overage (implying that the student previously failed a grade), low test scores, suspensions, absences, and an indicator for whether the student had spent time in an adult detention center. The school-related variable with the greatest explanatory power is per capita shooting rate.

[snip]

The 1,000 students predicted to be most at risk had a victimization rate of 1.4 percent. The next 4,000 students had a victimization rate of 0.83 percent. Students ranked between 5,000 and 10,000 were shot at a rate of 0.44 percent. Therefore, the model was able to identify a high-risk group that had victimization rates more than ten times higher than the population as a whole and six times higher than the typical black or Hispanic male.

Papachristos has found similar results, according to Kennedy: "simply being in an arrested-with network with a homicide victim increases your own chance of being murdered by 900 percent." Using the principle of social networks, it's a bit more powerful than a list of indicators. And both are not unlike CeaseFire, only one step closer from the potential incident and to its root. It's somewhere between science fiction, addressing precrime before it becomes crime, and common sense, that a host of minor troubles signal major ones on the horizon.

PS: Another sociologist Kennedy writes about in the context of her influence on McCarthy is another Chicago-turned-Massachusetts academic, Tracey Meares, whom I've written about before.

Photograph: Chicago Tribune