CRIME BEGETS CRIME
Mohler got his undergraduate degree in mathematics (and his slap-shot training) in his native Indiana, and went on to the University of California, Santa Barbara, where he researched mathematical modeling of polymers and fluids. After graduating, he got a job offer in that field but found himself more intrigued with a stranger one. Two UCLA professors, anthropologist Jeffrey Brantingham and mathematician Andrea Bertozzi, were working with the LAPD to develop algorithms to predict crime. They saw Mohler’s résumé and wanted him on board; turns out some of the mathematical models Mohler had been working with that describe pattern formation in polymers were similar to those the UCLA professors were using to predict burglaries. “I read their papers, and it made a lot of sense,” says Mohler. “I thought what they were doing was really cool.” He took the job.Mohler isn’t exactly in close touch with the mean streets. He spends most of his days in a sparsely furnished office in the basement of O’Connor Hall on the Mission Campus, at the end of a subterranean corridor bedecked with posters advertising upcoming math conferences and job openings for computer scientists. He’s on the skinny side of thin, with an easy, oversize smile that gives him an almost alarmingly cheerful look. With his rectangular geek-chic glasses, lace-less Converse sneakers, plaid shirt, and flop of black hair, he could be a San Francisco website designer or an indie rocker—which he has been, actually. In his spare time, he played bass in a couple of bands. The music has gone on hiatus since he and wife Courtney Elkin Mohler, an assistant professor in SCU’s Department of Theatre and Dance, became parents in 2011. But George Mohler still manages to strap on skates for some ice time as part of an adult hockey league.
The team gathered years’ worth of data from the LAPD on the time and place where home and car burglaries and auto thefts had taken place. (They focus on those crimes mainly because there are lots of them, providing a rich data set.) One of their key early insights was that crime tends to beget crime: If a house gets broken into, the probability of neighboring houses getting broken into soon after rises. Most crimes, like burglaries and car thefts, are not planned in advance but are opportunistic: A bad guy sees an unlocked window and ducks in. “Burglars typically don’t travel far. They tend to commit crimes in their own neighborhoods,” explains Mohler. “They have a lot of information: They know when their neighbors are at work and which houses are easy to get into. And when they succeed, they go back again. You see it in the data.” Mapping those patterns can give police an edge in figuring out where to deploy extra cars and cops to catch bad guys—or, better yet, keep them from opening that unlocked window in the first place.
In some ways, the notion of predicting where crimes will happen based on where they’ve happened in the past is obvious. That one event increases the likelihood of similar events occurring nearby in space and time is well established in other fields of research. In fact, you can see it everywhere in ordinary life: A punch thrown in a bar increases the chances of more punches. One kiss leads to another. Analysts and academics use the principle more methodically to predict where banana trees might be found, or where corporate defaults will cluster. One of Mohler’s main contributions to work on a new model for predictive policing was to find and adapt an algorithm developed by seismologists to help predict where aftershocks will strike after an earthquake.
If there have been a lot of muggings on a particular street for the last 50 weeks, there will probably be some the following week. Cops know that, of course. But the idea is to make those assumptions and guesses more accurate and to turn up patterns that aren’t so readily apparent.
Corporations have long used similar predictive analytics to anticipate consumer demand, and have found that cross-pollinating data can yield unexpected results. A famous example comes from Wal-Mart’s analysis of what its customers in coastal areas stock up on before hurricanes. The list includes duct tape and bottled water, naturally, but also a surprise item: strawberry Pop-Tarts.
Analyzing crime data can similarly yield counterintuitive conclusions. Most people think good lighting makes an area safer, for instance, but studies have found that it actually increases the chances of being victimized. It seems that muggers want to be able to see their potential targets clearly.