An 18-month, $25m experimental programme, called readi, tries to change individuals’ behaviour in the most violent districts of Chicago. It is one response—funded by Heartland Alliance, a big non-profit group, and philanthropists—to a surge in violence in 2016, when Chicago saw 762 murders. The police force is unable to solve 80% of murders and 95% of all shootings. Readi might make a difference. It mixes job training with months of intense efforts to teach habits of restraint. Unusually, the scheme applies lessons from a study in Liberia, in west Africa, after years of civil war left young, homeless men involved in crime, especially in Monrovia, the capital. Researchers there recruited 999 “hard-core street youth”, picking individuals deeply involved in crime. Some got grants to start a business, others a few weeks of therapy to change impulsive behaviour and teach basic skills for legal ways to make a living. Those who got both grants and therapy turned out to be much less likely to be involved in crime a year later, says Chris Blattman, a researcher at the University of Chicago who worked in Liberia. He now helps to advise the readi programme, which began in 2017 and ends its first phase this year. As in Liberia, the programme targets the hardest cases. An algorithm developed by the city’s Crime Lab trawled police data for individuals’ arrest history, age, address, social networks, and for those who already know victims of violence—all indicators of who is likeliest to pull a trigger next. Mr Bocanegra says 91% of participants have been arrested before, on average 17 times each. The focus makes sense: one study found 70% of non-fatal shootings and 46% of fatal ones occur inside a network of just 6% of city residents.