Chicago's Misleading Red Light Accident Statistics

The Chicago Department of Transportation (CDOT) has revised their highlighted study on the effectiveness of red light cameras.  (You can find my criticism of an earlier study here).  The new CDOT study reads:

The most recent crash statistics show that between 2005 and 2012 crashes of all types were down at intersections with cameras, and overall safety had improved:

  • Dangerous right-angle (T-bone) crashes decreased by 47% 
  • All crashes at those intersections were down 33% 
  • Crashes resulting in serious injuries were down 22% 
  • Pedestrian crashes were also down 22% 
  • Rear-end crashes were down by 7%

Here are the facts: 

In 2005 Chicago had 119,133 accidents.

In 2012 Chicago had 78,044 accidents.

Running the numbers, you will see that is a drop of 35%!  This means accidents all across the city dropped 35%.  So it is no surprise that accidents at RLC intersections between 2005 and 2012 dropped 33% according to the analysis by CDOT.

So how can this large drop be explained?  First, there is a general trend of accidents dropping because people are driving less the last 10 years.  Second, the definition of an accident changed in 2009 by going up to $1500, so that led a drop in accidents.  Third, as I shown in Chicago (and plenty of research around the world has concurred), red light cameras don't lead to large drops in accidents.


How many cameras in the CHA? - 3,000

It appears that Chicago Housing Authority finished a camera initiative around 2012.  The result is now 3,200 cameras security cameras that are linked with the Chicago Police Department and the OEMC.  The security cameras were purchased with $22.6 million in federal funding under the American Recovery and Reinvestment Act.


Inspector General report on Ticket Spikes

The Inspector General has released their report on the unexplained anomalies in red-light citation counts identified by the Chicago Tribune.  The findings echo some of the basic issues in the IG report on red light camera installation.

The report follows a long line of reports/stories pointing out issues with the red light camera system.  As I have long pointed out, these issues are not a surprise, because the purpose of the red light program is not to improve safety.  My research has long pointed out that there are not significant safety benefits associated with red light cameras.  The motivating factor for the red light camera system is revenue.  Once you realize the city's motivation it is no surprise that little forethought went into the planning and oversight of the red light camera program.  While Xerox is offering transparency, there is still the fundamental problem with the red light camera system is premised on revenue and not reducing accidents.

The report offers a mish mash of actual factual insights into the red light camera program combined with large gaps in our knowledge of how the red light camera program was run.  Here is the summary from the report:


Overall, our review revealed that the City’s management of the RLC program with Redflex was fundamentally deficient. The City did not ensure that Redflex was meeting all of its contractual obligations regarding routine maintenance and monitoring of the program. In addition, monthly reviews of RLC system performance by the City and Redflex failed to identify and timely address violation count anomalies and did not examine trends in RLC violations over time. Such analysis would have allowed the City to programmatically assess whether camera systems were functioning according to specifications. As OIG noted in its May 2013 RLC Audit, such trend analysis is also important for determining if the program is achieving its public safety objectives.

To date, the City has been able to identify and demonstrate the likely causes of anomalies at only three intersections. However, the absence of full historical program records and data makes conclusive identification of the causes of past spikes difficult.

Our review further concluded that CDOT has and continues to take significant steps to improve program management since operations transitioned from Redflex to Xerox, which assumed sole responsibility for the operation of the City’s camera systems in February 2014. 


How many cameras in the CPS? - 7,000

Chicago has more than 7,000 cameras in 268 schools (365 schools have no cameras).  The city has been moving to install newer digital cameras in schools.  The Chicago Public Schools (CPS) monitors the cameras themselves, but the Office of Emergency Management and Communication and individual schools can tap into the cameras’ feed.


How many cameras in the CTA? - 23,000

Keeping track of the cameras in Chicago is like following a spinning number wheel.  The CTA alone now has 23,000 cameras it monitors.  It appears a substantial amount of these cameras use high resolution cameras that would work well with the facial recognition system in Chicago.  Here is the latest on the CTA from their web site:

In May 2014, we announced the completion of a $13.9 million project to retrofit the majority of our rail fleet – more than 840 rail cars – with more than 3,300 360-degree high definition cameras. These state-of-the-art cameras can record and store high-resolution images from all angles, increasing the ability for police to identify criminal suspects.

The remaining rail cars not retrofitted with cameras are in the process of being replaced with our newest generation of rail cars – the 5000-Series, which come equipped with multiple surveillance cameras.
The rail car cameras are the latest addition to our already extensive rail system camera network, which more than doubled in size following an aggressive plan announced in June 2011, to install 1,800+ cameras in rail stations and platform in less than six months. All 145 CTA rail stations are now fully equipped with multiple, high definition cameras, which can provide a live-feed to the CTA Control Center and the Office of Emergency Management (OEMC).
In addition, our entire bus fleet has been cameras equipped since 2003, with up to 12 cameras on each vehicle depending on model.
Collectively, there are more than 23,000 cameras across the CTA. To maximize the effectiveness of this network and ensure police have quick and easy access to footage and live camera feeds, we created a new, modern video surveillance room.
Occupying a former library space at CTA headquarters, the new video surveillance room is approximately 2,800 square feet and is more than 12 times larger than the previous video room. Existing resources, including surplus furniture, computers and display monitors were used in creating the new room, resulting in no added costs to the CTA. Security staff and police detectives who work with the CTA on a regular basis have access to 20 terminals with 35 displays to view video from rail stations, rail cars and buses. In addition, there are seven dedicated workspaces for CTA investigators and security specialists; a quad-screen video panel for large-scale or multi-viewing purposes; and a team conference room.

Red Light Camera Study - 2014

Here is a snippet of my followup red light camera study.  The full report contains more extensive details and you can run your own analysis at my MiningChi site.

Executive Summary:

Chicago leads the nation with over 380 red light cameras (RLCs) at 190 intersections.  Four years ago, I did a short study that showed accidents in Chicago decreased between 2001 and 2008.  In contrast, accidents at intersections with RLCs increased.  This result threw doubt on the city’s claims of significant accident reductions because of RLCs. I am now conducting a follow-up study by analyzing the accident data and ticket data between 2009 and 2012.

The results showed that accidents have gone down overall about 6%.  This is expected, since there has been a reduction of 8% in the miles driven in Chicago.  Consequently, for a safety intervention to be effective, it would need to exceed the overall accident decrease of 6%.

Analyzing RLCs that were placed during the study, it was found that in the first year after a camera was introduced, there was actually an increase of 5% for accidents at the intersection.  This result was based on a very small sample size of 11 cameras and should not be considered definitive. 

A second interesting finding about RLCs was that accidents dropped at RLC intersections at a rate of 11% between 2009-2012.  This translates into 250 fewer accidents at RLC intersections between 2009-2012. The drop is largely due to a one-year 7% drop in accidents at RLC intersections during 2012.  It’s not clear what the cause of the drop is or whether it will persist. Nevertheless, in 2012 RLC intersections became safer than other traffic signal intersections.

The third point focused on using ticket data to explain why they are not reducing accidents. I offered two explanations for this: the inconsistent ticketing, e.g., spikes, and the rarity of accidents.

The results here mirror my earlier study.  The findings here are mixed and at best, the RLCs have a slight reduction on accidents.  Despite the million of dollars invested in RLCs and half a billion dollars in tickets, there is no evidence that the RLCs have had a significant safety benefit.


Accident Heat Map for Chicago

With the help of Aaron Moore, I have created an accident heat map for Chicago.  The map is colored by the crash rate for a street.  The streets in red are the more dangerous streets, while green has a reduced crash rate.  You can see the crash rate for a given segment by hovering on a street.  The crash rate takes into account the traffic volume for a street. 

I have embedded the map below, but you can see a full screen map here as well.  My hope is that this will allow people to understand what are the more dangerious areas in their community and use this to make their community safer.



Rethinking Privacy in a Digital Age

State Senator Daniel Biss is holding a panel on Rethinking Privacy in a Digital Age in Skokie on August 26th. Its a great lineup of panelists and an opportunity to ask questions.
Here is the complete info:
I hope you're enjoying the last days of the summer! I'm writing to invite you to join me at the next forum in our Critical Issues Series: Rethinking Privacy in a Digital Age. I'll be joined by fantastic panelists who will discuss the importance of balancing access to new and useful technologies against our desire for and right to privacy. As you may know, I've passed three bills that regulate law enforcement's use of these technologies, including drones and location tracking devices. In fact, the governor signed the follow-up drone legislation just two weeks ago. As technology continues to evolve fast, and as citizens continue to feel very legitimate concerns about the potential of overreaching surveillance techniques by our government entities, I intend to keep working on these issues and I very much look forward to your input.

Here are the details of the event, and I sincerely hope you'll be able to join us.

Event Details:

  • Date: August 26th
  • Time: 7:00 p.m.
  • Location: Skokie Public Library, 5215 Oakton Street
  • Panelists: Rajiv Shah, Adjunct Assistant Professor at the University of Illinois at Chicago; Jeremy Staum, a representative of Restore the Fourth Chicago; and Adam Schwartz, Senior Staff Council with the ACLU Illinois.

The Skokie Review and Evanston Now covered the panel and offers snippets of the discussion.


Bloody weekends in Chicago: A Historical Perspective

Every week this summer, we get a fresh set of statistics on how many homicides and shootings occurred over the weekend.  From listening to the numbers, they sound outrageous.  We know every homicide has enormous impacts on society in immeasurable ways.  But we also want to know, is the city becoming more dangerous?

As a starting point, last year there were 415 homicides (the lowest in a long time).  So roughly one murder a day and so on a normal weekend, you would expect 3 to 4 homicides.  This is a quick mental calculation people would do to see if it was a truly bad weekend.

I started by looking at roughly 6500 murders in the last 14 years from my MiningChi crime site.  I found 48% of them occurred on the weekend, which is defined as Friday, Saturday, and Sunday.  So that is the first interesting nugget, half of the homicides occur on weekends.

Next for 2011, 2012, and 2013, I took a look at homicides that occurred between May 1st and August 31st.  This time period of about 17 weeks actually accounted for 40% of all the homicides!  Bringing this all together, in the last three years, there has been an average of 4.7 homicides every weekend, during the summer.


Questioning Facial Recognition in Chicago

The Verge has an excellent story on Chicago's facial recognition.  It shows what happens when a reporter asks a few questions and checks their facts.  It is a MUST READ.  Here are a few snippets:


When Pierre D. Martin was arrested on robbery charges, it didn’t seem like an unusual case. There were nearly 12,000 robberies reported to the Chicago Police Department last year, and Martin was charged in two of them after allegedly carrying out daytime stickups on commuter trains. But a July 2013 Chicago Sun-Times article pointed out why Martin’s case stood out: he had been identified using facial recognition technology.

Martin was convicted in June this year. At that time, the state’s attorney, Anita Alvarez, told the Sun-Times: "This case is a great example that these high-tech tools are helping to enhance identification and lead us to defendants that might otherwise evade capture." But those tools — purchased by CPD through a $5.4 million federal grant — may not be as necessary as Alvarez would have you believe.

Martin's attorney told The Verge that facial recognition tech wasn’t even mentioned at trial. And a Freedom of Information Act request with CPD revealed "no responsive records" related to the department’s use of facial recognition technology in any arrests, including Martin’s. 

 . . . .

According to the Sun-Times, a Chicago Transit Authority surveillance camera picked up an image of a man who had just taken another man’s cellphone. CPD detectives ran the photo through NEC’s NeoFace computer program, and Pierre D. Martin’s name came up as the most likely suspect. The rest was history. Martin was arrested and charged. "This was our first success," said Jonathan Lewin, a CPD commander in charge of information technology, to the Sun-Times in reference to Martin’s case being CPD’s first facial recognition arrest. A year later, in June this year, Martin was convicted of robbery and sentenced to 22 years in prison.

 . . .

First of all, a cursory look at Martin’s rap sheet shows that he’s 25 years old — born in September 1988 — not 35, as the Sun-Times claimed. And, at least according to his public defender, John J. Kwon, Martin’s conviction had nothing to do with facial recognition. "I don’t know where that information came from," he told The Verge.Kwon stressed that Martin was identified in a lineup — and that the young man admitted guilt to investigators. An arrest report confirms both these things.

. . . 

Separate from the FOIA request, CPD’s public affairs office did not respond to repeated requests for comment about the department’s use of facial recognition technology. Emails to NEC Corporation’s marketing department went unreturned. And while Commander Jonathan Lewin assured The Verge that facial recognition was being used by the department "as a tool to help solve crimes" — both in Martin’s case and generally, in accordance with the department’s directives for facial recognition technology — he declined to say how many cases have been helped by facial recognition technology or the kinds of cases in which it’s being used.

. . . 

Furthermore, this was a man who, according to police records, was identified in CPD’s gang-monitoring system as a possible member of the Latin Kings street gang. He also had two prior convictions. This was a man very much on CPD’s radar.

As such, Martin’s case doesn’t really tell us anything about facial recognition’s effectiveness. What it shows is that NeoFace can do investigative work that pretty much any competent police officer could easily do — that is, if it was used at all.

. . . .

Cases like Martin’s "are often trotted out as major successes, without going into detail as to whether or not traditional police tools helped to capture the suspect or could have drawn the same conclusions on their own," Hall continued. "It is also important to look at the amount of data collected and money spent in proportion to the number of crimes solved by the particular technology. If Chicago has been using this tech for a year, and this is its first and only success story, that’s not a great record."