Skip to main content

The NY Times article on pedestrian fatalities (https://www.nytimes.com/interactive/2023/12/11/upshot/nighttime-deaths.html) is still heavy on my mind. So much so that I looked at the FARS data myself and created a chart of when fatalities occurred in relation to the time of sunset (i.e. a more explicit version of the NYT chart faceted by month)*. It's wild how concentrated fatalities are right after sunset. Blog post forthcoming.

* My chart only contains 2016-2021 data; the NYT chart goes back to 2000.
#VisionZero #rstats #dataViz

Chart titled "Crashes are highly concentrated right after sunset." The chart is a histogram with time since sunset on the x axis and the number of pedestrian fatalities on the y axis. There is a huge spike of fatalities starting right after sunset.
Screenshot of the NYT chart, "Pedestrian deaths by time of day." It's line chart faceted by month, showing how the peak of pedestrian deaths shifts over the course of the seasons.

Why Are So Many American Pedestrians Dying at Night?

Nothing resembling this pattern has occurred in other comparably wealthy countries.
Emily Badger (The New York Times)
#rstats #dataviz #visionzero
in reply to Harald Kliems

Bill Ricker
Bill Ricker
• •
that civil twilight was the singularly most dangerous time was taught in drivers-ed some 50 years ago.
These graphs beautifully, horrifically communicate that, but is this news to transit planners?
( It's discussed whenever DST is debated. )
in reply to Harald Kliems

Matthias Flor
Matthias Flor
• •
Wow, that is a crazily strong "signal" in the data!
This entry was edited (11 months ago)
in reply to Matthias Flor

Harald Kliems
Harald Kliems
• •
@mattflor Indeed. What I find weird is that the number don't start rising earlier, like in the half hour before sunset. We've all seen (and groaned about) the "driver was blinded by the setting sun and therefore couldn't avoid the crash" narrative. But the ped fatality data does not show that.
@Matthias Flor
in reply to Harald Kliems

Bill Ricker
Bill Ricker
• •

@mattflor

(Endured sunset 🌇 👀 yesterday. )
Guess-
▪️we drivers are all too aware of our impaired vision when driving into sunset, and mostly accommodate with extra caution &/or polarizers;
▪️ ¾ of sunset rush-hour trips are mostly not into sun ;
▪️ while immediately after sunset, we have a Dunning-Krueger belief in adequacy of illumination, before all headlights & streetlights are on.

@Matthias Flor
in reply to Bill Ricker

Paul Wermer
Paul Wermer
• •

@BRicker @mattflor expanding on that - as a pedestrian at just after sunset (so twilight) I can see pretty well. But I'm always surprised (as a driver) how much less visible pedestrians are to me at twilight (since as a pedestrian I don't have that experience).

I suspect drivers don't realize how quickly their vision degrades at twilight, and that's a key factor (exacerbated by rushing home to dinner, frustration with commute traffic, on the cell phone to let some know you'll be late because of aforesaid commute traffic, ....)

@Bill Ricker @Matthias Flor
in reply to Paul Wermer

Bill Ricker
Bill Ricker
• •

@PaulWermer @mattflor
When drivers over-estimate their ability to see into twilight shadows, and pedestrians over-estimate how visible their clothing is in twilight shadows, the carnage is so predictable.

Sometimes they're even the same people - certain blocks where delivery drivers double-park and jay-walk (lunge, dash across 2 lanes of traffic) for pickups. One might think knowledge would transfer.

@Matthias Flor @Paul Wermer
in reply to Bill Ricker

Bill Ricker
Bill Ricker
• •

@PaulWermer @mattflor

(I became a MUCH more paranoid cyclist when i got my car drivers license!
Why don't they?)

@Matthias Flor @Paul Wermer
⇧