CDOT Tells Council Improved Red Light Cameras Might Mean More Tickets

Rate of fatal crashes due to red light running has dropped more than rate of fatal crashes due to all other causes between the period Chicago didn't have and did have red light cameras.
In 12 of 14 cities surveyed, fatal crashes due to red light running dropped relative to all fatal crashes after red light camera enforcement began. Source: Insurance Institute for Highway Safety

Chicago Department of Transportation commissioner Rebekah Scheinfeld told the City Council’s Committee on Pedestrian & Traffic Safety that equipment problems that have plagued the city’s red light camera enforcement program have mostly been resolved. Although some aldermen were there to pounce upon the media ruckus around the cameras, other aldermen understood that red light cameras can play a role in keeping Chicago’s streets safer.

Alderman Deborah Graham (29th) recounted her experience with a motorist who ran a red light caused a car crash two years ago. “I was going through an intersection at Lake and Sacramento, and a woman trying to beat the traffic light cut me off, and I struck the CTA support beam,” she told the committee at their regular meeting on Tuesday. “I had to have a major surgery,” she said, adding, “So I have a real concern when we’re approaching intersections and dealing with the (red light) cameras.”

Graham was asking Scheinfeld about how CDOT plans to keep equipment in “top-notch” form since the city changed vendors from Redflex, which used unreliable induction loop detectors, to Xerox, which uses a more accurate radar system that can easily be checked remotely.

Alderman Deb Mell (33rd) expressed her confidence in the changes that CDOT had made in response to the IG’s investigation. “I think people drive way too fast in this city,” she said, adding, “I’m really happy that we had this [hearing], and kind of air it out, and going forward, [red light cameras will] do their job.”

Scheinfeld described changes CDOT has made to their agreement with Xerox, which include financial penalties if Xerox flags too many “false positives.” (Violations are sent to IBM for a second review.) CDOT also reduced the allowable rate of “close calls” from 15 percent in Redflex’s contract to 10 percent in Xerox’s contract. The new contract has already had results: CDOT has already fined Xerox $28,867 for “failure to meet specified performance metrics.”

She said these new “business rules” will be posted on the CDOT website, along with the violation data that’s already been posted to the city’s open data portal. If there are “spikes” in the future, an “early warning system” will alert Xerox staff, who will be required to consult with CDOT staff. Scheinfeld said that CDOT’s own staff will also be monitoring violations data themselves, watching for anomalies.

Chicago Inspector General Joe Ferguson was also at the meeting to answer questions about his report [PDF], which detailed Redflex’s poor record keeping and CDOT’s lax management. Since the report was issued, he said, “CDOT has been extremely responsive to all of our suggestions.”

Scheinfeld told the committee that some of the “spikes” in tickets that the Tribune has reported on so extensively occurred after repairs to the loop detectors. Ferguson explained that a possible reason for the so-called spikes was that the “trigger speed” had fallen at some locations to as low as 5 MPH, a limit that’s since been raised to 13 MPH. The camera only starts recording motorists when they’re traveling faster than the trigger speed, and even the less-sensitive 13 MPH trigger detects 99 percent of motorists who might be running a red light, Scheinfeld said.

Ferguson wrote that the trigger speed problem led to many periods of severe under-enforcement — not over-enforcement, as the “spikes” suggest. In other words, more tickets should have been issued to violators than were issued.

Several aldermen made a fuss about tickets that were issued after a documented yellow light duration of 2.9 seconds, instead of the usual three seconds. This is within the 0.1 second range that an operational traffic signal is allowed to deviate from the programmed time. CDOT narrowed that window when it changed vendors: Redflex had to throw out any violation where the yellow light was less than 3.0 seconds, but Xerox can keep the violation when the yellow light was 2.9 seconds. (Xerox’s software truncates instead of rounds, so 2.99 seconds is rounded to 2.9 rather than three.) Yet that sub-100 millisecond difference does not make or break a motorist’s ability to come to a complete stop at a red light — after all, that’s less than half the 300 millisecond time it takes to blink. CDOT changed the rule in September, requiring Xerox to also throw out violations where the yellow light is recorded as being less than 3.0 seconds long.

None of the committee members doubted that red light cameras save lives. The Insurance Institute for Highway Safety [PDF] examined 14 cities with red light cameras between 2004 and 2008, and found that Chicago was among the 12 cities where red light running crashes made up a declining share of all fatal crashes. Across all of the cities, the rate of fatal red light running crashes decreased by 35 percent — much higher than the 14 percent decrease seen in cities without red light cameras.

Alderman Graham went on to express her deeper concern for safety in Chicago. “I hope that driving behaviors improve, such that we come to a point where we don’t need the cameras,” she said. Graham asked for all Chicagoans to “take part in better driving habits, and behaviors on the road, and being friendly to our pedestrians and our bikers that are out there — and courteous to our other driving comrades out there.”

  • PhotoRadarscam

    The IIHS study has been thoroughly debunked here:
    The discussion over light timing shows a lot of ignorance. More than 50% of tickets are issued for violations occurring within the first 1 second of a red light, so even a few tenths of a second will actually affect a large percentage of tickets. No cops would ever issue a ticket for a sub-1 second violation as it is virtually undetectable by humans. The right thing to do would be to comply with the FHWA’s guide on photo enforcement which recommends issuing no ticket until the end of the all-red interval (the short time when all lights are red at an intersection), and that this would be fair for drivers, but it wouldn’t make as much MONEY which is what this is all about!

  • alexfrancisburchard

    I disagree, if you’re entering the intersection on the all-red phase, you’re leaving on someone else’s green, no way in hell that is safe driving.

  • CameraScamera

    This is all complete BS. I love that Rebekkah thinks she can act above the law and issue tickets for illegally timed lights. It doesnt matter if it’s “imperceptible”. It’s AGAINST THE LAW! Yet Rebekkah thinks she’s above the law.
    So funny that streetsblog wil Always side against to motorist. If something illegal was happening regarding bikes, Steve and John would be crying like babies.
    Streetsblog- your continued “support” and advocacy of illegal activity by the government makes you just as bad.

  • CameraScamera

    Forgot to mention – great point about how the data presented is completely flawed.
    Streetsblog would NEVER take the time to research and actually present facts. Shame on them. They are crooks too!

  • Barney

    The heavy loading of violations in the first half second of the red signal was brought up in the last chicago RLC discussion here and people cited studies in the comments. This should eliminate ignorance and poor research as reason for the dismissive attitude of a 0.1 second shortening of the yellow signal. Thus it would appear that harassing drivers takes priority over safety.

    Take a look at this light in CA. Adding 0.3 seconds of yellow halves the violations and adding 1.3 seconds drops them from 98%:

    Every fraction of second is money and that’s why it’s important. If safety was important have a longer yellow. Add a camera if you want to. it won’t make money but it might stop a few who run the red on purpose.

  • CameraScamera

    Great information. And you’re right, every 0.1 of a second is about money.
    The city refuses to lengthen the yellow lights, they try to lower the threshold for issuing tickets. Steve- please tell us how this program is about SAFETY.

  • PhotoRadarscam

    Why would a driver with a fresh green light enter an intersection when there is a car still in it? THAT is dangerous driving. In fact, cross traffic should never proceed until the intersection is clear.
    But it’s not my information. I’m trusting the peer-reviewed publications and recommendations of the FHWA (Federal Highway Administration). I think they know what they are talking about, and this recommendation has no agenda.

  • Dennis_Hindman

    From the FHWA website:

    Reducing Red-Light Running Through Longer Yellow Signal Timing and Red-Light Camera Enforcement

    Results of a Field InvestigationRetting RA et. al., Insurance Institute for Highway Safety (IIHS), June 2007.

    “This study shows that the provision of adequate yellow signal timing reduces red-light running, but longer yellow timing alone does not eliminate the need for better enforcement, which can be provided effectively by red-light cameras.”

    Here’s another study on the FHWA website that says it all:

    Red-Lights Mean Stop,

    Hasson, Patrick, FHWA, Washington, DC, 2000

    The Los Angeles County MTA uses red light cameras at intersections along their light-rail lines and the Orange Line BRT which has a parallel bike/ped path.

    I’ve observed several occasions when drivers have gone through a red light at a Orange Line intersection that has a red light camera. Its a very scary situation with drivers several times barely missing pedestrians and bicycle riders (I ride a bicycle along this path) that are starting to go through the crosswalk on a green light.

    A Orange Line bus could also easily smash into some of these red light violators as the peripheral vision of the bus driver can be very small along this narrow two lane roadway. At some intersections there is a soundwall and buildings close to the busway that reduces the bus drivers view of what is approaching on the cross streets. Some bus drivers resort to honking every time they go through some of the intersections. The MTA even imposed a 10 mph speed limit for their bus drivers at intersections along this route to try and improve safety.

    Some of these violators just flat out don’t seem to know that they are supposed to stop unless the bright flash for the camera hits their eyes.

    One bicycle rider who ran a red light at a busway intersection had his head go through the windshield of one of the buses.

    One bus driver told me that one time a bicycle rider wearing ear buds ran a red light and the driver had to slam on his brakes. Fortunately he had enough time to stop. The bicycle rider was totally unaware of what happened and turned onto the parallel bike path. The bus driver moved slowly next to her and continuously honked. The bike rider never looked his way, nonchalantly and blissfully riding along. Unfortunately, the red light cameras do not flash when pedestrians or people on a bicycle go through a red light. If they did, safety would be improved for pedestrians and bicyclists at these intersections.

    It isn’t always a timing issue for the red light violators, it can also be a situation where they just aren’t paying attention and drive through a red light.

    The Orange Line BRT had 20 red light violations a month at the Sepulveda Blvd intersection previous to the red light camera installations. After, the cameras were installed the violations fell to 2 a month. Quite an improvement. The LA County MTA only put red light cameras along the Orange Line where they had problems with drivers on the cross streets running red lights.

    The LA County MTA also tried other techniques such as experimenting with a strobe on the side of buses, having red lights embedded in the roadway at an intersection, yellow flashing traffic signals for the bus when going through a couple of intersections instead of a green light. The MTA closely monitors the results. They are not about to use something that does not give them a safety improvement and they have found that red light cameras have reduced the number of red light violations.

  • duppie

    The posts this morning are interesting.

    The general tone is to accuse Streetsblog of reporting biased towards the use of redlight cameras, a claim that Streetsblog writers would likely not dispute.

    But to “proof” that Streetsblog is biased, the only link shared is of a website called The A 1-minute scan of the articles on that website gives the distinct impression that TheNewspaper itself is heavily biased against any action that results in limiting the rights of motorists. And unlike Streetsblog, none of the articles clearly list the author.

    Debunking the IIHS study is fine by me, but it would be nice if you guys link to some reputable media coverage of this study, instead of some fringe blog.

  • Because they’re thinking about something else, staring at the traffic light, and reflexively go when it turns green without thoroughly checking the intersection?

    Though cities tend to, culturally, EITHER jump greens OR run reds, and Chicago is a runs-red city.

  • duppie

    That may indeed the recommendation. But the problem with that recommendation is that you rely on someone else for your safety. That someone else may not be paying attention and simply floor it when his light turns green. I’ve seen that happen many times.

    I’d rather not rely on other road users to keep me safe.

  • I don’t understand all this acrimony about red light cameras. It seems obviously to improve safety–I’ve noticed a change in my driving behavior and the crash data bears this out. Plus it’s a source of revenue for the city from someone other than me!

  • David P.

    What I see most often is a driver thinking about something else and staring at their phone or other nerd-dildo device, and they proceed only when they see other traffic movement in their peripheral vision. This occasionally leads to a sort of standoff wherein no one moves for a few seconds because they’re all spacing out and waiting for someone else to move to signal to them that the light is green.

  • David P.

    The self-pity and sense of martyrdom by commenters (not just here) who self-identify as motorists is really rich.

  • jeff wegerson

    My concern early on with the camera lights was the hesitation during a stale green and the fear of being rear-ended by a too fast stop to avoid a ticket. Hesitant driving is often hazardous.

    But, with the addition of the pedestrian count-down timers, that fear on my part has gone away. Indeed I believe the count-down timers to be the best thing since sliced bread for driving in the city. I can now comfortably and safely stop even a little too soon.

    With the count down timers all the arguments against giving out red light tickets become moot in my opinion. Just about every ticket becomes completely deserved. It’s time for city drivers to take it down a notch. I know I have been.

  • skyrefuge

    Especially odd is what these self-identified motorists are pushing for: an increase in their taxes, AND an increase in traffic congestion.

    Because if longer yellows do in fact reduce red-light violations, then that revenue will have to be made up elsewhere, and that new tax likely won’t be as avoidable as not-driving-through-red-lights is.

    And if longer yellows and longer all-red phases in fact keep drivers out of intersections, that reduces the throughput of every intersection throughout the city, which increases congestion and travel times.

    But hey, congratulations to them, they’re even more concerned about safety than Streetsblog is, and are willing to sacrifice both their time and their money in order to achieve it!

  • jeff wegerson

    Why do you use the word “crooked?” What do you mean by it? I’m pretty sure that Steve is not unfairly making money from drivers. He’s likely not even making money fairly from drivers.

    Also Rebekkah is really unlikely to be personally benefiting from the income from the cameras. Perhaps CDOT is benefiting. Still I don’t understand your purpose in using the word and I suspect it merely serves to undermine your creditability as a commenter.

  • skyrefuge

    Can you point me to some data showing where both (1) yellow intervals were increased city-wide, and (2) the decrease in violations was permanent?

    Because it’s fairly obvious that any lengthening of a yellow interval would result in an immediate drop in violations, since drivers are still expecting the light to turn red earlier. And then it’s also fairly obvious that if only one intersection has a longer yellow relative to other intersections in the same city, that intersection would see fewer violations because drivers are conditioned to believe that *all* intersections have the shorter yellow interval.

    I wouldn’t be shocked if there is such a case, I just haven’t seen it; the page you got the graph from ( ) shows both #1 and #2, but never #1 and #2 together.

  • Fred

    Brian has a personal vendetta against traffic cameras. Read all of his posts with that bias in mind.

  • BlueFairlane

    I tend to assume all the common first names without surname or buddy icon are Brian and therefore not worth addressing. There seems to be a common tone and theme to any such comment, and there have been a rash of them lately.

  • Barney

    I don’t understand the desire to keep short yellows unless safety is not the goal. I don’t understand the vigorous defense of Chicago’s outdated three seconds just about everywhere method. It’s always ‘do you have more proof?’… what about this detail or that detail? It’s always raising the bar to a level of data collection and study. For what?

    The Chicago method is outdated. We know short yellows make the money on errors in the first half second of red signal. We know that well set yellow signal length + RLC means near zero violations. Mr. Vance cited such a study in the last discussion. Longer yellow + RLC = very very few violations. I don’t know why that study didn’t look at the first half second of red, but they didn’t, but even three times a very small number is still a very small number. Longer yellow + RLC will probably offend nobody. Few people will get tickets and then only those who ran the light on purpose.

    The problem with the data you ask for is that red light cameras are removed after the timing is fixed. Nobody wants to lose money for years and years on end. We only know after a few months in most cases because of the money motivation. When the cameras go away the data collection stops.

    RLC’s are always at the very least picked because something isn’t right with them. Intersections that don’t get RLCs are usually ok or at least not as bad. Only in states where by law the yellow must be longer with an RLC do all of them get fixed at once. Then the programs are shut down so there is no long period of ‘after’ data.

    In this plot we see that they yellow was increased, but probably not enough. Only 0.7 seconds. However the drop in violations was 75% It then stayed that level for two years. Increase the yellow another half second and it will probably drop to 4 violations a month.

    That’s the best the data gets.Most places the cameras are pulled and never re-instated best that I know. I know of no intersections where cameras were re-instated after being pulled because the violations dropped with a longer yellow. If this ‘habitualized theory’ had any basis in reality there would be.

    So why not keep the RLCs and set the yellow timings correctly? Everyone will be happy enough but the city and xerox. The people who want vengence against red light runners will have it and regular drivers won’t be scammed. Those who want safety will actually have it. I don’t understand why this is so difficult. The only possible objection is by people who just want to harasses drivers with fines by deliberately mis-engineering the signals and intersections. If safety is the goal just do it. What’s the risk? If it works we get safety. If it doesn’t there will be no change. I don’t
    understand why safety people don’t want to do this.

  • PhotoRadarscam

    So you’d rather rely on a camera for your safety?

  • PhotoRadarscam

    You can keep citing IIHS materials but they might as well be written by the camera vendors.
    And there is no single measure that will eliminate red light running and make people perfect drivers.
    But the causes of crashes and problems at every intersection are unique and need to be studied. It has been proven quite extensively though, that if there is a problem with red light running that extending yellow light times is VERY effective and tend to reduce RLR by well over 50%. This solution costs almost nothing and is quick and easy to implement. But cities aren’t interested. Why? No MONEY in it.

  • Barney

    But RLC ticket revenue goes to the corporate fund. It is not counted as a motorist paid tax.

    The reduction in throughput is trivial even if no other compensation is made. The most profitable RLC in Chicago issues just under 1700 tickets a month, 56 cars a day. Even doubling or tripling that value is trivial compared to the traffic volume. However these prolific RLCs usually rely on right-on-red to make money. Many people don’t turn right on red at RLC intersections, so that decreases throughput all by itself.


  • Barney

    The Insurance Institute for Higher Surcharges has never seen an enforcement scheme it didn’t like or create a study to support. The more safe motorists are ticketed the more money its members make. (In some states camera tickets do come with points) Because car insurance is often regulated they look for ways to charge motorists unlikely of a claim more money.

    (note there are errors in this list, including Illinois, but I spot checked and some are correct)

  • duppie

    I dont rely on RLCs nor FHWA recommendations. Instead I rely on myself for my safety.

  • Actually drivers do not change behavior on expectations of the length of yellows. When drivers see a yellow, drivers do what they must do according to the laws of physics. For a 45 mph road, it takes a driver 7 seconds to stop. DOTs set the yellow by federal standard to 4.5 seconds. Drivers will run the red in the difference between 7 and 4.5 seconds, that is up to 2.5 seconds. Reduce the yellow to 4 seconds and drivers run the red up to 3 seconds. Increase the yellow to 5.5 seconds, drivers run the run up to 1.5 seconds. No exceptions. This is what the data shows for 140,000 tickets issued in Cary, NC.

    Think about your own driving. When you see a light turn yellow, are even aware how long the yellow is?

  • Onix AQua
  • skyrefuge

    Yes, I make driving decisions all the time based on my assumptions of how long the yellow will be. The most frequent example is when I’m going, say, 30mph on a 45mph road (due to congestion). When I see the light turn yellow, I know that I’ll have more than enough time to stop, but also know that I’ll more than enough time to continue through the intersection, so I continue through. If I’m going the same 30mph at the same distance from the intersection but on a road with a 30mph limit, then I always stop because I know the yellow interval is shorter.

    It takes drivers 7 seconds to stop from 45mph? You must live on a different planet than me, with different laws of physics or different drivers. I know of no yellow light that’s 7 seconds, and no drivers have difficulty stopping in a time much shorter than that.

  • skyrefuge

    Yes, I wasn’t assuming that a specific “motorist paid tax” would be used to make up for the decrease in RLC revenue. It would be more likely to come from property taxes or sales tax or even income tax. The point is, money always has to come from *somewhere*.

    In terms of intersection throughput, I wasn’t saying that eliminating the red-light-runners would decrease the throughput, I was saying that increasing the amount of time the intersection goes unused would decrease the throughput. As a toy example, imagine if the yellow interval was increased to 10 seconds. That would leave 6 to 7 seconds at every light change where no one was moving. That’s at least 3 fewer cars that can make it through for every green phase. That’s one of two major reasons why yellow intervals aren’t 10 seconds long: it would totally snarl traffic.

  • skyrefuge

    Thanks, I had not considered that “data after yellow-interval increases” simply doesn’t exist because the cameras get pulled. I’m not sure how the contracts with camera vendors are written (particularly Chicago’s), but I thought it was something like the vendor doesn’t charge the city anything up-front for the cameras, they just take a cut of the violations? But then whose decision is it to take down the cameras when they stop being able to pay for themselves? The city or the vendor? The only way I could see cities actually taking them down for revenue reasons (rather than driver-complaint reasons) is if the contracts put the city on the hook for maintenance/operation when they start losing money.

    So I think Chicago *should* increase all yellow intervals citywide, AND keep all the cameras in place for at least two years, just so we actually can have some real data on both violation changes and the effects on traffic congestion that the rest of the country/world can use to help form their own decisions. Heck, maybe some national anti-RLC groups would be happy to kick in funding to pay for the cameras over that time just to collect good data that supports their view.



Crashes Doubled After Houston Banned Red Light Cameras

Law enforcement officers warned there would likely be an uptick in collisions when Houston debated banning red light cameras in the early part of the decade. Turns out they were absolutely right. Houston voters banned the life-saving technology in 2010, with the press mostly cheering them along. Last year Houston PD examined how that’s impacted safety at intersections. According […]

Vox Pulls Back the Curtain on “Scam” to Save Lives With Red Light Cameras

You can usually count on Vox for accurate, research-based explainers of public policy issues. That’s why the new Vox video on red light cameras is so monumentally disappointing. Researchers have established that red light cameras make streets safer by reducing potentially fatal T-bone collisions, though they do lead to more rear-end crashes, which tend not to be very serious. […]