The Tribune's analysis of more than 4 million tickets issued since 2007 and a deeper probe of individual cases revealed clear evidence that the deviations in Chicago's network of 380 cameras were caused by faulty equipment, human tinkering or both.
Chicago transportation officials say they had no knowledge of the wild swings in ticketing until they were told by the Tribune — even though City Hall legally required the camera vendor to watch for the slightest anomaly in ticketing patterns every day. Many of the spikes lasted weeks.
The lack of oversight raises new questions about the controversial traffic enforcement program, the largest in the country, now embroiled in a federal corruption probe into allegations that the city's longtime red light camera manager took bribes from the camera company.
“Something is terribly amiss here,” said Joseph Schofer, an associate dean at Northwestern University's McCormick School of Engineering and Applied Science who reviewed the Tribune's research.
Schofer, who has served as an adviser on city transportation committees, said the findings prove “the system is broken.”
“The only reasonable explanation is that it is something involved in the technology,” he said. “Whether it's diabolical or mechanical or electronic and accidental, I can't look inside people's souls and know that, but the evidence is pretty strong.”
He and three other national experts who reviewed the Tribune's findings suggested that drivers are entitled to refunds, whatever the cause of the spikes.
A 10-month Tribune investigation documented more than 13,000 questionable tickets at 12 intersections that experienced the most striking spikes; similar patterns emerged at dozens of other intersections responsible for tens of thousands more tickets. Among the key findings:
>>Were you ticketed in error? Search the Tribune's database
Cameras that for years generated just a few tickets daily suddenly caught dozens of drivers a day. One camera near the United Center rocketed from generating one ticket per day to 56 per day for a two-week period last summer before mysteriously dropping back to normal.
Tickets for so-called rolling right turns on red shot up during some of the most dramatic spikes, suggesting an unannounced change in enforcement. One North Side camera generated only a dozen tickets for rolling rights out of 100 total tickets in the entire second half of 2011. Then, over a 12-day spike, it spewed 563 tickets — 560 of them for rolling rights.
Many of the spikes were marked by periods immediately before or after when no tickets were issued — downtimes suggesting human intervention that should have been documented. City officials said they cannot explain the absence of such records.
Drivers who appeal red light tickets in Chicago win less than 10 percent of the time, but drivers caught in one severe spike won 45 percent of the time. Yet the vast majority of drivers caught during spikes never appealed and therefore missed an opportunity to have bad tickets thrown out.
The experts said the deviations identified by the Tribune are not supposed to happen in an automated enforcement system and should have set off warning flares from City Hall to the Phoenix headquarters of the city's longtime camera vendor, Redflex Traffic Systems Inc.
"This was a lightning strike, and lightning strikes should not happen in a case like this," Schofer said.
The experts all said the available evidence leads them to only two possible explanations — that ticket procedures were quietly broadened to catch more violators, or that malfunctions led the system to wrongly tag lawful drivers. In either case, they said, fail-safes that should have guarded against such anomalies didn't do their job.
Consistency and fairness are crucial to traffic enforcement because the goal is to make streets safer by changing driver behavior. Any sudden change in enforcement — either by design or by malfunction — undermines that goal, the experts said.
All the experts agreed that the city should consider refunds. Some said the city has an ethical obligation to drivers who were ticketed unfairly, even in those cases in which they were technically in violation of the law.
"The public has to believe that this is a safety countermeasure and not a moneymaking scheme, and that it is a fair system," said Joseph Hummer, professor and chair of civil and environmental engineering at Wayne State University in Detroit. "The fairness is critical to it if people are going to accept these cameras in the city, that they are not arbitrary."
City transportation officials said neither the city nor Redflex made any changes to how violations were enforced. They acknowledged oversight failures and said the explosions of tickets should have been detected and resolved as they occurred. But they said that doesn't mean the drivers weren't breaking the law, and they defended the red light camera program overall as a safety success story. The program has generated nearly $500 million in revenue since it began in 2003.
The city was unaware of the spikes until given the evidence by the Tribune in January, said David Zavattero, a deputy director for the Chicago Department of Transportation. In the six months since, city officials have not provided any explanations.
"Trust me when I tell you that we want to know what caused these spikes you have identified as much as you do," Zavattero said. "So far we can find no smoking gun."
He acknowledged that faulty camera equipment likely played a role.
"I would say that is likely in some of these cases," Zavattero said. "I cannot tell you that isn't possible. It is possible. The old equipment was much more prone to break down than the equipment we are currently installing."
He said steps have already been taken to improve oversight of the program under a new vendor hired this year to run the system after Redflex was fired by Mayor Rahm Emanuel amid Tribune reports on the bribery scandal.
"We want the system to be operating fairly and consistently for everyone," Zavattero said. "The level of oversight we have and that we need to take on this program has improved significantly."
But Zavattero and Luann Hamilton, deputy transportation commissioner, balked at the suggestion of refunds. They said a random check of 300 videos from the spikes identified by the Tribune found no instances of tickets being issued erroneously. The city refused to provide those videos to buttress its contention, citing state law that makes them confidential.
Redflex officials declined to be interviewed for this story and referred all questions to city officials.
Emanuel fired Redflex last year after Tribune disclosures about its cozy relationship with former city transportation manager John Bills, who was instrumental in giving the company the contract and oversaw its growth and operation until he retired in 2011. Federal authorities arrested Bills on bribery charges in May.
After the company publicly acknowledged it likely paid as much as $2 million in bribes for the business, the Tribune in March 2013 asked the city to provide a database of all the tickets the company helped generate, as well as information on driver challenges to those tickets. The city refused at first but provided the records in September after months of legal wrangling.
The Tribune also filed other records requests, searching for any shred of paper at City Hall — service records, maintenance reports, email traffic, memos or anything else — that might help explain the reasons for the spikes. The city said it had none.
As part of its analysis, the Tribune charted the daily ticket counts for each camera since 2007, examined thousands of still photos of violations available online, looked at hundreds of videos of violations, sought advice from experts and traffic safety engineers, and interviewed more than 100 ticketed drivers.
In most cases, cameras installed at city intersections performed exactly as advertised. There was a large surge in violations when the cameras were first installed, and then the ticket counts tapered off as motorists adjusted their behavior — the city's stated goal for installing the cameras in the first place.
But in a significant subset, the Tribune found a pattern of problems that raises broader questions about how the program is run.
Driver frustration
Oumou Wague had driven her Chrysler minivan through the intersection at Halsted and 119th streets thousands of times since the city first installed a red light camera in 2004, and never once received a ticket there.
So imagine her surprise in May 2011 when the first computer-generated violation showed up in her mailbox. And then another. And then another. And then another.
In one particularly expensive stretch of 18 days that spring, she received four separate $100 tickets for rolling through a right turn on red on her way home from work — at the exact same spot she had successfully navigated every day without a hitch for years.
"I was stunned," the 42-year-old hair salon owner said in a recent interview. "I knew right away there was something wrong there. I knew that camera was broken, but you can't fight City Hall — and that is a fact."
Wague was caught by a camera system in hyperdrive. After averaging fewer than three tickets per day for years, the camera that tagged her was suddenly averaging 33 tickets a day — many of them issued in error. Then, as abruptly as it began, the ticket spree ended in June of that year.
Spikes in tickets, spikes in appeals
From April 29 to June 19, 2011, one of the two cameras at Wague's West Pullman intersection tagged drivers for 1,717 red light violations. That was more violations in 52 days than the camera captured in the previous year and a half.
Almost all of the tickets were for the same type of violation — right turns on red from the southbound lanes of Halsted onto 119th.
The same thing happened on the Far Northwest and North sides.
The camera at Lincoln Avenue and McCormick Road prompted 563 tickets during a 12-day period beginning Dec. 30, 2011 — more than all the tickets issued from that camera in the previous two years. The rate of tickets shot from less than one a day to 47 a day.
The camera at Western and Peterson avenues prompted 410 tickets during a 21-day stretch beginning Sept. 20, 2012, an average of 19.5 tickets per day, up from the typical 2.2 per day the previous two years.
On the Near West Side, the corner of North Ashland Avenue and West Madison Street generated 949 tickets in a 17-day period beginning June 23, 2013. That is a rate of about 56 tickets per day. In the previous two years, that camera on Ashland averaged 1.3 tickets per day.
All four spikes were accompanied by a period when the cameras went dark — two days before the surge at Halsted; immediately after the surge at Ashland; and both before and after the surges at Lincoln and Peterson.
At three of those intersections the surges were also accompanied by an eye-popping increase in appeals. To understand that significance, it helps to know how infrequently people appeal the red light tickets, and how rarely anyone succeeds in having a ticket overturned.
More than 95 percent of all red light camera tickets are never appealed. And the drivers who do contest their tickets lose more than 90 percent of the time.
But drivers who appealed tickets during some of the worst spikes were far more likely to win — nearly half the time in one particularly stark example.
At Halsted, 242 of the tickets during the spike were appealed — 1 in 7. And 109 of them were tossed out.
At Lincoln, 81 of the tickets during the spike were appealed — 1 in 7. And more than a quarter of those tickets — 22 — were tossed out.
And at Peterson, 37 of the 410 tickets during the spike were appealed and seven tossed out.
In almost all of those cases where a ticket was tossed out, it was because the officials hearing the appeal ruled that the vehicle came to a stop.
Experts: Evidence points to trouble
Traffic experts said the appeal rates are alarming, and by themselves indicate a problem.
"It seems there may be quality control problems all the way up and down the line here," said Schofer, of Northwestern University. "Are the people who are supposed to be viewing these videos actually viewing the videos? Given the appeal rates at some of those intersections, it would appear not. I mean, people make mistakes, but they don't make 30 mistakes."
The cases raise many questions the city says it cannot answer.
What prompted the camera system to suddenly target so many drivers? If two people independently review each of the violations before they are sent — a legal requirement of the program — then how did so many erroneous tickets slip through? And how could the wild swings in ticket rates go unnoticed by the company and city officials?
"I would say that very definitely, answers to these questions would be very revealing," said Dr. Rahim Benekohal, a traffic camera expert and engineering professor at the University of Illinois at Urbana-Champaign. "Perhaps some of these people should get their money back."
Benekohal and Schofer also said they are troubled by the periods before and after the spikes when the cameras appeared to be shut down. In some cases it was for days or even weeks.
"There was something strange going on there, no doubt," said Benekohal, who sat on the city selection committee to choose a vendor for the city's speed camera system. "Certainly one of the conclusions that could be reached is that the camera was not functioning properly — that is certainly a strong possibility.
"But typically if a sensor goes down or there is an equipment issue, these things do not repair themselves. There should be a record of it somewhere," Benekohal said.
Schofer said the downtimes suggest intervention of some kind.
"It looks to me like someone intervened, and as a result of that intervention there was a dramatic increase in the number of tickets," he said.
The most credible explanation, Schofer said, is that "the ticketing criterion changed."
"Specifically, it became more stringent. That seems credible. The next question is: Was it too lax in the before period or too strict in the after period?"
City officials insisted the city has not changed its enforcement practices. They also said they have no records indicating camera malfunctions or adjustments that would have affected the volume of tickets.
The lack of records is significant, because Redflex was required to document any time the operation of a camera was disrupted for more than a day, as well as work "that will affect incident volume" — in other words, adjustments or repairs that could increase or decrease the number of violations.
The company was also required to perform a series of daily quality checks to evaluate "the daily activity of the intersection cameras and the central server to determine if there are any anomalies on the data provided."
Asked to provide any documentation of maintenance or repairs around the time of the worst spikes that would correspond with activity causing those cameras to go dark or to change the rate of ticketing, City Hall provided two dozen single-page maintenance records. Only one documented any problems — a "disabled" lane at the Lincoln intersection.
All the experts interviewed agreed that a change in the camera system — not shifting driver patterns — is the most likely reason for the spikes. They all expressed surprise that such dramatic ticketing fluctuations in an automated system could go unnoticed and undocumented.
"I have never seen anything like this," said Richard Retting, a former senior transportation engineer for the Insurance Institute for Highway Safety and a longtime advocate for the safety benefits of red light cameras.
"Obviously, something very strange is happening during these unexplained and dramatic spikes, no question about it," said Retting, who was dubbed the "father of the red light camera" movement in a congressional report on red lights. "And I would think the system operator must have some record to explain them, especially since they seemed to end as abruptly as they began.
"There must be some record, some maintenance log, something somewhere," he said.
City Inspector General Joseph Ferguson was likewise stymied by a lack of records in his May 2013 audit of the red light program, which sought to analyze the hows and whys of where the cameras were placed but did not look at ticket data as the Tribune did.
"We found a lack of basic record keeping and an alarming lack of analysis for an ongoing program that costs tens of millions of dollars a year and generates tens of millions more in revenue," Ferguson wrote.
How the cameras catch motorists
Since 2003, when Chicago first awarded the contract to Redflex, drivers who run red lights at up to 384 cameras are tagged by what is known as "magnetic induction loop" technology. Oddly, it doesn't determine whether you stop at a red light, but how fast you are going as you approach.
That's because under the law, the camera system has to capture you entering the intersection after the light is red. So the camera must start recording before any violation takes place. Rather than capture images of tens of millions of cars passing by legally, the system is designed to predict which vehicles are most likely to run the red light.
A loop of sensors embedded in the pavement near the intersection measures your speed, and if you are going faster than a preset "trigger speed" — usually 13 miles per hour — the video camera is triggered. The video camera records your car moving through the intersection, the traffic light visible overhead. A separate still camera records your license plate.
Employees of the red light vendor — which until March was Redflex — review the videos, determine whether there was a violation and then forward the video and the information to a second city contractor. Employees of that vendor, IBM, review the violation again to make sure, send the license plate number to state authorities to retrieve ownership information and issue the ticket by mail.
City officials puzzled
There are many legitimate reasons that ticket rates might fluctuate at an intersection, including traffic diversions blocks away, weather events, road construction, even funeral processions. When first asked about the spikes, city officials suggested that some of those reasons might be the cause. But neither transportation officials nor the Tribune could find any evidence of such outside forces affecting the wild swings in two dozen cases presented to City Hall.
When city officials were first presented with evidence in December of the spike at Halsted and 119th, they suggested it was because of funeral processions related to a large cemetery eight blocks away. However, only one of the 109 overturned tickets during the spike was attributable to a funeral procession, according to appeals records supplied under the Illinois Freedom of Information Act.
Regarding the Ashland spike, city officials at first suggested it might be attributable to some event at the United Center, about three blocks away. But there were no significant attractions that would have led to higher traffic, and even if there were, the duration of the ticket spike went well beyond the hours of any event such as a concert.
Likewise, city officials offered no explanation for the times the cameras generated no tickets right around the time of the spikes. They said they had no records to indicate that cameras were taken down during those periods where they inexplicably generated no tickets for days at a time.
Zavattero said Redflex and its subcontractors were responsible for maintaining the cameras and that Redflex usually briefed city officials once a month on maintenance issues.
One former Redflex software engineer who helped develop the Chicago system told the Tribune that Redflex had no computerized warning systems to flag the company when unusual spikes occurred.
"The only time we ever knew when a camera was down was when someone happened to notice, or if we got a phone call," said Michael Schmidt, who was laid off by Redflex last year amid company downsizing.
Schmidt said that during his 10 years working on the Chicago contract for Redflex, cameras routinely malfunctioned or stopped working, or sensors embedded in the roadway went out either on their own or because a road worker accidentally severed the connection. Schmidt said he had no knowledge of whether Redflex shortened trigger speeds for some cameras but agreed with the experts that such a change could result in the spikes.
"When I look at this data I notice right away that a lot of them happened after or before the camera going dark," Schmidt said. "That tells me somebody is out there working on it, or adjusting it. There really is no other explanation."
Yellow light timing
All of the experts, as well as Schmidt, said another explanation for these kinds of spikes would be shorter yellow light intervals. They said fractions of a second make a huge difference on violations, as drivers readjust to the shorter yellows.
Zavattero said the city has not tinkered with yellow lights, and that all of them meet state and federal standards that require at least three seconds on all yellows. He said any yellow light that falls out of whack and decreases its interval to 2.8 seconds automatically triggers the traffic system to a flashing red until it can be recalibrated.
"I can tell you there have been no changes to the trigger speeds, and there have been no changes to the yellow light intervals," Zavattero said. He added that while Redflex had the ability to alter trigger speeds in its roadway sensors, the company had neither the authority nor the ability to change traffic light intervals.
For that, the company would have had to call the city, he said. "And I can tell you it was not allowed,'' he said.
But the yellow light interval during one of the most severe spikes fluctuated wildly, according to information extracted from a review of every ticket generated by the camera at 6200 N. Lincoln Ave. for more than a year.
For the 100 tickets issued in the six months before the spike, the yellow light interval at that intersection ranged from 4.03 to 4.08 seconds. In the 563 tickets issued during a 12-day spike there, yellow light intervals dropped to about three seconds during 406 of them, then bounced back and forth to more than four seconds intermittently during the others.
Two of the tickets showed a yellow light time of 0.05 seconds, according to city records, though in both cases the images indicate the light was red long before the violations happened.
"There is no way the amber times should be bouncing around like this," said Schofer, shown the evidence. "Obviously, if you shorten up the amber times you are going to write more tickets."
Experts agreed that a one-second shortening of a yellow light interval is more than enough to throw off drivers who go through the same intersection every day and set their mental clocks by the longer interval.
A firestorm erupted in Florida last year after news reports that state government officials had quietly approved shorter yellow lights at red light camera intersections, boosting ticket rates. Public outrage from the story prompted state officials to reverse course.
Chicago officials adamantly deny tinkering with yellow light intervals to increase revenue. Zavattero said all the city's yellow intervals are set to no lower than three seconds.
Focus on right turns
Another striking finding in the examination of the spikes was how many tickets during the spikes came on right turns.
A random sample of 1,100 red light citations around the city shows that about one-third are given to drivers for not coming to a complete stop before making a right turn on red. But almost all of the tickets during two of the most dramatic spikes went to drivers making a right turn, according to the Tribune's review of still images from more than 2,500 tickets at the Halsted and Lincoln intersections.
The same was true for tickets thrown out from those intersections, according to records that include notes from administrative law judges who preside over hearings and handle appeals by mail.
One such hearing officer, Eileen McHugh, tossed out two tickets in one day for retired Chicago schoolteacher Debra Blackmon-Parrish, 60, who received both while she turned right on red at Halsted and 119th during the 2011 spike.
"Based on the video, I am going to find that you made a stop prior to making the turn, and I am going to find that you are not liable on the ticket," McHugh told Blackmon-Parrish, according to an audio recording of the hearing provided by the city.
"You know, I think there's a flaw with that light or something," Blackmon-Parrish told McHugh. "There are some issues with that light."
"I have seen a couple at this particular" location, McHugh responded.
Blackmon-Parrish said she was told by a city representative that day that the city was aware of a problem at the intersection.
"She told us that the sensor in the street was having a problem," Blackmon-Parrish said. "She said it had been malfunctioning and that it was being worked on."
But those comments weren't captured on the recording.
A Tribune review of more than 100 tickets generated by the Halsted camera before the spike shows that tickets there are typically for right turns on red.
But some other spikes demonstrated dramatic increases in tickets for drivers making right turns on red. Consider the camera at 6200 N. Lincoln Ave.
An examination of 100 tickets generated by that camera in the six months prior to the spike shows that only 12 were issued to owners of vehicles making a right turn on red. The other 88 violations were written for owners of cars who went straight through the red light, the photographs show.
But during the 12-day spike of 563 tickets, all but three of the citations — more than 99 percent — went to owners of cars that turned right.
"It sure as hell should not look like this," Schofer said. "Suddenly everyone at that intersection decides to only turn right? I don't think so."
A dozen drivers who were tagged for rolling right turns at the 6200 N. Lincoln camera told the Tribune that drivers are forced to creep into the intersection because the cross street of McCormick merges at an angle.
"If you don't move out into the intersection, you can't see oncoming traffic at all," said Thomas Garrity, 73, a retired Chicago police sergeant who was ticketed during the spike that started Dec. 30, 2011.
Photographs and video of the alleged infraction show a bicyclist just feet from the front of his car when the camera recorded him, and Garrity said he can't imagine he didn't stop to avoid hitting the bike.
"They claimed I didn't stop, but I am pretty sure I did stop," he said. "I always make sure I stop, but no, I didn't bother to contest it. From my experience you might as well try to explain to them there are aliens.
"I know how this city works, and you just don't get a fair shake," he said.
The video of Garrity's offense shows his car approach the intersection in the right turn lane, cross over the crosswalk, brake and come to what appears to be a stop as the bike passes in front of him.
Changes to state law made after early negative reaction to the first red light cameras in Illinois allow drivers to stop past the stop line or crosswalk as long as pedestrians or bicyclists are not present.
A Tribune review of video from some of the spike intersections that remained on the city's website found numerous cases in which drivers appeared to make a complete stop with no pedestrians or bicyclists present, and dozens more where drivers appeared to crawl slowly through a right turn during the red light.
At 6200 N. Lincoln, hearing officers threw out 22 tickets appealed from the spike, explicitly noting in most cases that the driver had stopped. A Tribune review of those videos found that drivers hit their brakes and slowed to a crawl as they went around the corner, suggesting the hearing officers considered that behavior excusable.
Change in enforcement?
Experts said the surge in tickets for so-called rolling right turns suggests a decision was made to begin enforcing violations that were not previously enforced. If the goal is to change driver behavior, it's not fair to alter enforcement without notice, they said.
"The way we do that is by getting the word out," Wayne State's Hummer said. "So if those were real violations that they decided to enforce that day and then stopped the next day, where is the notice there?
"The camera is a safety countermeasure that is specifically installed to try to help with a particular problem, and if one day they decide, 'Well let's go after these other problems over here,' that is violating what I consider the public trust," he said.
For drivers like Garrity, Blackmon-Parrish and Wague, the reasons for the spikes that cost them so much time and money are not relevant.
Wague has gotten other red light tickets. Since 2003, records show, she has received eight — four during the spike at Halsted and 119th less than a mile from her hair salon and another four sprinkled at different intersections over the years.
Records show she tried to appeal only two of the tickets.
"It's just not worth it. Nobody listens, and all they want is their money," she said. "Maybe now they will start listening. I think we should all get our money back. It's what is fair."
Northwestern's Schofer agreed.
"I think if I discover the process is broken, it is sort of like the General Motors vehicle recall thing. We are all over General Motors because they didn't recall in a timely way.
"In a sense, you have an obligation to recall that ticket and say, 'Something was broken in my machine. I don't know what it was, but I am going to make the fair assumption that it wasn't you.'"
©2014 the Chicago Tribune
