On March 21, 2020, Illinois ordered residents to stay at home because of the COVID pandemic. Within a week, Chicago's daily count of reported incidents had fallen below anything the same calendar week showed in 2018 or 2019. Across the six weeks of the order the city averaged — incidents a day, against — a day for the matching window the two years before. A drop of —.

The mechanism is the same one that runs through the blizzards card. Routine activity needs people in the street: bus stops, parking lots, sidewalks, bars. Pull them indoors and most of what the city counts as crime stops happening. Narcotics enforcement collapses hardest because it depends almost entirely on stops in public places. Theft, burglary, and criminal damage drop too, by smaller fractions.

The chart below shows the whole year in one view. The dark fill across April and May is the lockdown trough. The drop of volume stayed even after the lockdown ended except for the red column rising out of it at the end of May. It is a different story, hidden inside the same year. Press play to look closer.

We go back to the year 2020 bigger picture. Each crime category has its own shape across 2020. Some take more the pattern of lockdown and some peak more during the Floyd window.

The year ran two stories at once. The dominant one was structural: a six-week drop when the city's routine activity stopped. The brief one was reactive: a three-day spike of property crime and disorder when the streets filled again, this time for unrest.

Crime in this dataset is mostly what happens when people are out. Empty the streets and the count falls. Fill them with something other than ordinary life and the count rises.

On the night of October 20, 2014, an officer named Jason Van Dyke fired 16 rounds at 17-year-old Laquan McDonald on South Pulaski Road. A dashcam in a nearby cruiser recorded the whole thing. For 13 months the city kept the file out of public view.

A freelance journalist named Brandon Smith spent most of 2015 trying to find and release publicly the video, first under the Illinois Freedom of Information Act and then through Cook County court. The city's lawyers fought him at every step, arguing that release would prejudice an open investigation. On November 19, 2015, Judge Franklin Valderrama ordered the city to produce the video within five days. Hours before that deadline expired on November 24, the Cook County State's Attorney charged Van Dyke with first-degree murder. The video went out the same evening and the case became a national controversy.

The release detonated a week of demonstrations. Crowds shut down Michigan Avenue on Black Friday and returned downtown through most of December. Mayor Rahm Emanuel asked for and accepted the resignation of Superintendent Garry McCarthy on December 1. Within days, the head of the city's police-oversight body had also stepped aside; the Cook County State's Attorney was facing public calls to do the same. For a stretch of weeks, the city operated in open political crisis.

The case was a scandal and the city went through a year of political fire over it. The question is whether anything changed in how the Chicago Police Department does its job.

Yes, and sharply. The crime did not change. Across the 48 months centred on November 2015, the monthly volume of reported incidents drifts by only ·. The response did. In the 24 months before the marker, roughly · of incidents ended in an arrest. In the 24 months after, ·. A gap of · opened in about a year, and the line settled into a new groove well below the old one. Same incidents. Different department.

Why the response changed is harder to say. The most-discussed explanation is de-policing, sometimes called the "Ferguson effect". It is the idea that after high-profile cases of police violence and the scrutiny that follows, officers pull back from proactive stops and arrests, whether to avoid risk or in response to lower morale. The mechanism is plausible, but it is hard to prove: a drop in arrests is consistent with de-policing, but also with other changes happening at the same time. One of those is concrete. In August 2015, the Chicago Police Department signed an agreement with the ACLU that required officers to file a detailed report for every investigative street stop. That added paperwork, and it changed how stops were recorded.

The break was not a single shock. It was a cascade that ran for years, through City Hall, the police department, and the federal government.

The case did not end there. The DOJ probe opened on December 7 became, by January 2017, a 164-page report finding a pattern of excessive force and constitutional violations by the Chicago Police Department. When the incoming Trump administration declined to pursue reforms, Illinois Attorney General Lisa Madigan sued the city in August 2017 to force them anyway. That suit produced a settlement and, on January 31, 2019, a federal consent decree.

Scandals fade. The next question is whether this one did. Was the break a brief reaction or did the new level hold?

It held. A decade on, the line has not climbed back. The arrest rate fell off a cliff in late 2015 and stayed down.

It is tempting to read the decree onto this picture: the reform that kept the new level in place. The dates make that hard. The drop began in late 2015. The decree was entered in January 2019, more than three years later. Whatever cemented the change was cemented before the decree existed. Reading any post-2019 decree effect is harder still: the pandemic arrives in 2020 and disrupts every series in Chicago's records for the years that follow. What the timing on this chart points to is the case itself (the moment something visible happened inside the department), not the legal mechanism that arrived years later. That is not a verdict on whether the decree worked. It is a reading of when the line moved.

The data on this page fixes the when, sharply. The why is contested, and this article does not try to settle it. Any of these threads from the case itself, de-policing, the stop-report paperwork, federal scrutiny can be pulled.