All the Crashes Red-Light Cameras Are Missing; Previewing a Third Ward Sunday Streets

Downtown Houston with Sunflowers

Photo: Russell Hancock via Swamplot Flickr Pool


16 Comment

  • Shocking that the HPD would issue poorly parsed data to mislead the public about how fantastic red light cameras are at stopping accidents. Other cities have shown extending the yellows and the “all directions red” by a second or so can really cut down on major t-bone collisions. If serious accidents are really happening at these locations at a higher rate, they should work harder on some of those more innovative technologies.

  • Do not dignify the HPD’s red light information with the term ‘data’. At best it is an anecdote.

  • juancarlos31, yes, shocking. As in “I’m shocked, shocked to find gambling going on in here.” :-)

    It’s also fascinating that they had all that data, and didn’t chose to share any of the intersection-specific numbers with us. Hmmm.

  • I’m shocked that some individuals believe other individuals would run a red light whether cameras were present at the intersection or not. I thought we were all pre-wired to do the right thing [like not run a red light].

  • I’m pleased to see this recent information about the intersections that once had red-light cameras, because it explodes the argument favored by red-light runners that the cameras *increase* the likelihood of an accident when they are present and turned on. It was something to do with causing rear-end collisions.

    Added to that, the responses generated so far are good fun. juancarlos31 makes a neat deflection away from what the data are telling us and toward other methods of reducing collisions because of red light runners—methods that merely accommodate red-light runners by giving them more time to do it.

    Gisgo and Pffft, take the road more often traveled of simply dismissing evidence that they don’t like. Gisgo goes the extra mile on that road with the Orwellian assertion that the HPD statistics are anecdotal rather than data, thereby reversing the definitions of those two words.

    There isn’t much else to conclude from the short history of red-light cameras except that a lot Houstonians are determined to run red lights and will make any argument to prevent being penalized for it, much less caught.

  • Inasmuch as HPD’s numbers are not normalized to traffic density, or corrected for other factors, they are just numbers, not statistics. But it’s useless to argue with a poster who suffers severe innumeracy.

  • @Houstonreader: The data does not prove what you seem to think it proves. You’re assuming causality when there isn’t enough information in the linked article to determine any sort of causality.

  • HPD was candid in pointing out that it didn’t take traffic volume into account when they reported an average doubling of accidents at the 50 intersections that once had the cameras. Gisgo would have us believe that the increase in crashes is the result of increased traffic volume, as if our driving population had suddenly doubled.

    Gisgo also mentions “other factors,” but doesn’t name them or comment on how they might have skewed the results of the HPD study.

    Innumeracy, indeed.

  • You’re not getting your $$$-generators back, HPD. This pathetic attempt to try and make the citizens feel guilty for scrapping them isn’t working. Take a few officers off of your freeway u-turn ticket writing squads and put them at the intersections at peak times if you think they’re so dangerous. It’s all about ‘safety’, right? Rolling my eyes

  • It’s worth calling out the political undercurrents that affect the discussion on red light cameras…

    1. Mayor Parker and HPD love cameras because they are an easy source of revenue for the cash-strapped city; they will support them at all costs, even risking moderate political backlash from the electorate.
    2. For the past 5 years, every Chronicle article concerning red light cameras has been pro-camera.
    3. Never has a full set of analyzed data (or a peer-reviewed article) been published to public; instead, we have been fed anecdotes, press releases, and statements that attempt to summarize the data for us – leaving out all those precious details.

  • @takeshi: You’re correct that I’m assuming causality (in addition to correlation). I do so because no other causes would account for the average doubling of accidents at those 50 intersections. The other possibilities would include an average doubling of accidents citywide, or an average doubling of the volume of traffic at those fifty intersections.

    Because the driving population of the city has not doubled between the periods 2006-2010 and 2010-2014, I look at what has changed and draw the most likely conclusion.

    I could be persuaded that there are other causes involved, but I’d *still* favor red-light cameras because the evidence strongly suggests their deterrent effect. After all, isn’t fewer red-light runners and safer intersections what we want?

  • @ Houstonreader: The alternative possibilities that you cite are absurd, I think intentionally so as to create a strawman argument. Here are some that aren’t:

    1) Another possibility is that the number of accidents at those intersections is actually very low to begin with so that one’s ability to make inferences from data is limited by a small sample size.

    2) Another possibility is that the traffic has indeed increased but that the rate of accidents is geometrically proportionate to the traffic level, meaning that for every 1% increase in traffic throughput, an intersection might yield a 5% increase in accidents.

    3) Another possibility is that human factors have changed; for instance that the people in the impacted neighborhoods (which have historically tended to be poorer than most) have found gainful employment and that they have a reason to want to hurry to get somewhere.

    4) Another possibility is that the method for counting the number of crashes in a particular intersection is either flawed or systematically biased in some way (which happens with crime stats all the time).

    5) There exist other plausible possibilities.

  • @Houstonreader: When cameras were still up and running, I nearly rear-ended someone on countless occasions when they suddenly slammed on their brakes because the light turned yellow, for fear of being caught in the intersection and issued a ticket. I’d like to see data on how many people were rear-ended two cars back from the intersections with cameras – of course, that data would not cite the “intersection” as the site of the accident, but rather the segment of road between two intersections.

    Your last sentence is classic political speak… And I agree – by the same logic, the national speed limit should be 20 mph. Statistics show that only 2% of traffic fatalities occur at speeds under 20 mph. And none of us want to die, right?

  • @ Superdave: Your nearly rear-ending someone is anecdotal evidence, and it may say more about your driving than it does about red-light cameras. Fact is, there’s no collision in your anecdote, so it doesn’t count for much.

    That said, I too would like to see the statistics concerning rear-end collisions at those intersection. And then I’d like to compare them to T-bone collisions at the same intersections without the cameras.

    I don’t know what you mean by “political speak,” but that bit about driving 20 miles per hour is classic false equivalence: red-light cameras had no effect that is comparable to instituting a 20mph national speed limit.

  • @Niche: asserting that my points are absurd isn’t an argument. If you think I’ve created a straw man, then explain it.

    The rest of your post doesn’t change much: your scenario #1 might well be the case, but that wouldn’t contradict or invalidate the effect of red-light cameras. #2 & #3 are imaginative hypotheses without any supporting data. And, funny, #3 looks very much like a straw man argument with its creation of newly employed poor folk hurrying to work. Really, Niche? #4 reverts to Gisgo and Pffft’s strategy of dismissing the data out of hand.

    And your #5, amounts to “trust me, I can come up with lots more hypothetical scenarios for denying these data.” Occam’s Razor, I’d say.