15 ms·

The linked Dangerous by Design 2021 report [1] shows the data on increased pedestrian deaths: > The number of people struck and killed by drivers nationwide wh

The linked Dangerous by Design 2021 report [1] shows the data on increased pedestrian deaths: > The number of people struck and killed by drivers nationwide while walking increased by an astonishing 45 percent over the last decade (2010-2019). [1] https://smartgrowthamerica.org/dangerous-by-design/ https://smartgrowthamerica.org/dangerous-by-design/

unclebucknasty 5y ago
Interestingly, the article you linked doesn't mention SUVs or refer to vehicle size at all. It attributes the increase to an emphasis on speed over safety: >state and local transportation agencies place a higher value on speed (and avoiding delay) than they do on safety...When faced with decisions that would elevate and prioritize safety for people walking but increase delay for vehicles, the decision-makers’ true priorities are laid bare. The original linked article, however, does point to another article on a govt site that gets into vehicle size. [0] But, in total the conclusions vacillate between the issue being lack of visibility (implying more accidents) and increased lethality owed to mass and, possibly the higher profile of the vehicle. Of course, you could conclude it's both more accidents and increased lethality, but the data doesn't seem to bear that out. For instance, the report cites a meta study that found a 2x-3x increase in fatalities when trucks or SUVs are involved, but the report also states the fatality rate for those vehicles is 40%. The original article further notes sales of trucks and SUVs have outpaced smaller vehicles since at least '07 and other assessments show these larger vehicles now make up 50% of vehicles on the road. So, the data seems to show disproportionately fewer fatalities for these larger vehicles. [0] https://www.federalregister.gov/documents/2015/12/16/2015-31323/new-car-assessment-program https://www.federalregister.gov/documents/2015/12/16/2015-31...
- kadoban 5y ago
  > For instance, the report cites a meta study that found a 2x-3x increase in fatalities when trucks or SUVs are involved, but the report also states the fatality rate for those vehicles is 40%. I don't understand. How is that data contradicting the hypothesis? Why is there a "but" there?
  - unclebucknasty 5y ago
    >How is that data contradicting the hypothesis? Doesn't necessarily contradict the hypothesis on its own. My overall point is that the article cites different data, some of which is tangential to its hypothesis (while purporting to support it). Other data outright contradicts the hypothesis. Overall, the article was sloppy--just kind of throwing stuff against the wall, IMO. >Why is there a "but" there? The gov't report notes that trucks/SUVs made up 33% of the total vehicle fleet (sorry, I didn't quote that important bit in my previous comment, but it's in the report), while they make up 40% of all fatalities. That doesn't translate to a 2x-3x higher rate for trucks/SUVs.
    - scythe 5y ago
      Probably some fraction of "trucks/SUVs" are professional drivers who usually have a much lower crash rate than private drivers. But the SUV looks twice as deadly in the direct comparison of collisions, while there isn't much reason to expect the type of car has a causal effect on the chance of a collision, ceteris paribus. So we expect a truck/SUV would be twice as deadly in the same hands as a car.
      - throwaway0a5e 5y ago
        Professional drivers tend to have a high fender bender rate just from sheer exposure to traffic and a very low "texted their way into a stopped traffic at 60mph" rate because if you behave like that you will have close calls first and get fired over them.
      - unclebucknasty 5y ago
        >Probably some fraction of "trucks/SUVs" are professional drivers who usually have a much lower crash rate than private drivers. I read the crash rate as already factored-in. Here's the fuller quote from the report linked [0] by the article that I'm referencing: The majority of fatal pedestrian crashes involve light vehicles. About one-third of pedestrians who are injured are struck by an SUV or pickup truck, which corresponds closely to the make-up of SUVs and pickups in the U.S. vehicle fleet. However, SUVs and pickups account for closer to 40 percent of pedestrian fatalities, which suggests that injuries may be more severe when sustained in collisions with these vehicles. Results from a meta-analysis of 12 independent injury data studies showed that pedestrians are 2-3 times more likely to suffer a fatality when struck by an SUV or pickup truck than when struck by a passenger car. So, they're saying the accident-rate is proportional—i.e. maps to the percentage of SUVs and pickups on the road. It's only the fatality rate that is increased. And, sure, 40% would be disproportionately high vs 33%, but it doesn't represent a rate that's 2x-3x higher. >there isn't much reason to expect the type of car has a causal effect on the chance of a collision Well, the original article actually suggests that reduced visibility in SUVs/trucks may contribute to higher crash rates. That's exactly the kind of thing I found sloppy. Just a statement with no data to back it up (in fact, the gov't report they linked contradicts it, as noted above). The article itself opens with this beauty: You don’t need to read a research paper to know that more people are driving trucks and SUVs (sport utility vehicles) compared to 20 years ago, that today’s trucks and SUVs are significantly bigger than they were 10-20 years ago, or that they’re more dangerous for people walking, biking, or getting around with an assistive device. It's intuitive. It's actually not intuitive that they're more dangerous for pedestrians. For instance, they could come equipped with more sensors and other tech that accounts for their size and actually makes them safer than other vehicles. Or, drivers of larger vehicles could be more cautious. Or, their increased size may make them more visible, hence, pedestrians are less likely to step out in front of them. And, what's the point of noting that "it's intuitive" anyway? Not the best way to start an article that should be (and purports to be) data-driven. In general, the article suggests a series of loosely assembled "intuition" and facts, then draws these hard conclusions that I'm not sure we're certain about yet. I'm not disputing their claims. Just saying it could be much tighter and reads a bit like an agenda loosely packaged in data and appeals to "intuition". I think the real answer is that we need more data and are still figuring this out, especially if we're looking for actionable solutions. [0] https://www.federalregister.gov/documents/2015/12/16/2015-31323/new-car-assessment-program https://www.federalregister.gov/documents/2015/12/16/2015-31...
dzhiurgis 5y ago
I blame more smartphones and opioid crisis
- bottled_poe 5y ago
  I would be surprised if the increase in drivers using smart phones was not a significant factor. Opioid crisis seems less likely, where’s the data?
  - sokoloff 5y ago
    I wouldn’t be surprised if pedestrian use of phones (and to a lesser extent, medication/intoxicants) was a contributing factor in a significantly higher fraction of mishaps than in 2010.
  - dzhiurgis 5y ago
    Since no one in this thread has any data - I’m pulling it from my ass

unclebucknasty 5y ago

Interestingly, the article you linked doesn't mention SUVs or refer to vehicle size at all. It attributes the increase to an emphasis on speed over safety: >state and local transportation agencies place a higher value on speed (and avoiding delay) than they do on safety...When faced with decisions that would elevate and prioritize safety for people walking but increase delay for vehicles, the decision-makers’ true priorities are laid bare. The original linked article, however, does point to another article on a govt site that gets into vehicle size. [0] But, in total the conclusions vacillate between the issue being lack of visibility (implying more accidents) and increased lethality owed to mass and, possibly the higher profile of the vehicle. Of course, you could conclude it's both more accidents and increased lethality, but the data doesn't seem to bear that out. For instance, the report cites a meta study that found a 2x-3x increase in fatalities when trucks or SUVs are involved, but the report also states the fatality rate for those vehicles is 40%. The original article further notes sales of trucks and SUVs have outpaced smaller vehicles since at least '07 and other assessments show these larger vehicles now make up 50% of vehicles on the road. So, the data seems to show disproportionately fewer fatalities for these larger vehicles. [0] https://www.federalregister.gov/documents/2015/12/16/2015-31323/new-car-assessment-program https://www.federalregister.gov/documents/2015/12/16/2015-31...

kadoban 5y ago
> For instance, the report cites a meta study that found a 2x-3x increase in fatalities when trucks or SUVs are involved, but the report also states the fatality rate for those vehicles is 40%. I don't understand. How is that data contradicting the hypothesis? Why is there a "but" there?
- unclebucknasty 5y ago
  >How is that data contradicting the hypothesis? Doesn't necessarily contradict the hypothesis on its own. My overall point is that the article cites different data, some of which is tangential to its hypothesis (while purporting to support it). Other data outright contradicts the hypothesis. Overall, the article was sloppy--just kind of throwing stuff against the wall, IMO. >Why is there a "but" there? The gov't report notes that trucks/SUVs made up 33% of the total vehicle fleet (sorry, I didn't quote that important bit in my previous comment, but it's in the report), while they make up 40% of all fatalities. That doesn't translate to a 2x-3x higher rate for trucks/SUVs.
  - scythe 5y ago
    Probably some fraction of "trucks/SUVs" are professional drivers who usually have a much lower crash rate than private drivers. But the SUV looks twice as deadly in the direct comparison of collisions, while there isn't much reason to expect the type of car has a causal effect on the chance of a collision, ceteris paribus. So we expect a truck/SUV would be twice as deadly in the same hands as a car.
    - throwaway0a5e 5y ago
      Professional drivers tend to have a high fender bender rate just from sheer exposure to traffic and a very low "texted their way into a stopped traffic at 60mph" rate because if you behave like that you will have close calls first and get fired over them.
    - unclebucknasty 5y ago
      >Probably some fraction of "trucks/SUVs" are professional drivers who usually have a much lower crash rate than private drivers. I read the crash rate as already factored-in. Here's the fuller quote from the report linked [0] by the article that I'm referencing: The majority of fatal pedestrian crashes involve light vehicles. About one-third of pedestrians who are injured are struck by an SUV or pickup truck, which corresponds closely to the make-up of SUVs and pickups in the U.S. vehicle fleet. However, SUVs and pickups account for closer to 40 percent of pedestrian fatalities, which suggests that injuries may be more severe when sustained in collisions with these vehicles. Results from a meta-analysis of 12 independent injury data studies showed that pedestrians are 2-3 times more likely to suffer a fatality when struck by an SUV or pickup truck than when struck by a passenger car. So, they're saying the accident-rate is proportional—i.e. maps to the percentage of SUVs and pickups on the road. It's only the fatality rate that is increased. And, sure, 40% would be disproportionately high vs 33%, but it doesn't represent a rate that's 2x-3x higher. >there isn't much reason to expect the type of car has a causal effect on the chance of a collision Well, the original article actually suggests that reduced visibility in SUVs/trucks may contribute to higher crash rates. That's exactly the kind of thing I found sloppy. Just a statement with no data to back it up (in fact, the gov't report they linked contradicts it, as noted above). The article itself opens with this beauty: You don’t need to read a research paper to know that more people are driving trucks and SUVs (sport utility vehicles) compared to 20 years ago, that today’s trucks and SUVs are significantly bigger than they were 10-20 years ago, or that they’re more dangerous for people walking, biking, or getting around with an assistive device. It's intuitive. It's actually not intuitive that they're more dangerous for pedestrians. For instance, they could come equipped with more sensors and other tech that accounts for their size and actually makes them safer than other vehicles. Or, drivers of larger vehicles could be more cautious. Or, their increased size may make them more visible, hence, pedestrians are less likely to step out in front of them. And, what's the point of noting that "it's intuitive" anyway? Not the best way to start an article that should be (and purports to be) data-driven. In general, the article suggests a series of loosely assembled "intuition" and facts, then draws these hard conclusions that I'm not sure we're certain about yet. I'm not disputing their claims. Just saying it could be much tighter and reads a bit like an agenda loosely packaged in data and appeals to "intuition". I think the real answer is that we need more data and are still figuring this out, especially if we're looking for actionable solutions. [0] https://www.federalregister.gov/documents/2015/12/16/2015-31323/new-car-assessment-program https://www.federalregister.gov/documents/2015/12/16/2015-31...

dzhiurgis 5y ago

I blame more smartphones and opioid crisis

bottled_poe 5y ago
I would be surprised if the increase in drivers using smart phones was not a significant factor. Opioid crisis seems less likely, where’s the data?
- sokoloff 5y ago
  I wouldn’t be surprised if pedestrian use of phones (and to a lesser extent, medication/intoxicants) was a contributing factor in a significantly higher fraction of mishaps than in 2010.
- dzhiurgis 5y ago
  Since no one in this thread has any data - I’m pulling it from my ass