According to the Mayor of New York City, De Blasio: (my emphasis)

Excessive speed contributes to 25 percent of roadway fatalities on New York City Streets and is the leading cause of motor vehicle crashes. Reducing vehicle speed from 30 to 25 mph doubles the likelihood of a pedestrian surviving a crash.

He seems to be saying that the expected number of pedestrian deaths would go down to half of what it is currently (which the numbers I've seen don't agree on, so I'll ask a separate question for that.)

What percentage of current pedestrian deaths would be expected to be avoided with a 5mph reduction in the speed limit?

Here's a more explicit quote with a subtly different claim, from later in the article: (my emphasis)

“Because of the leadership of Mayor de Blasio, this life-saving measure is now law,” said Department of Transportation Commissioner Polly Trottenberg. “Dropping the default speed limit by 5 mph may not seem like a lot, but it cuts the chances of pedestrians dying from being hit by a vehicle in half.

  • 7
    He says that reducing the (actual) vehicle speed, not the (nominal/legal) speed limit, has that effect. – ChrisW Oct 30 '14 at 15:21
  • @ChrisW Is the added quote better? And also, it makes a difference how fast most crashes involving deaths occur at. I'm not so sure most are with cars going 30mph. – ike Oct 30 '14 at 15:32
  • Yes, thank you, the added quote is better (i.e. that claim better matches your question). – ChrisW Oct 30 '14 at 16:39
  • 1
    It would need more data and some serious calculation. The "cut in half" is claimed to happen if instead of being hit by a car at 30mph, you are hit by one at 25mph. We don't know if this is the travel speed or impact speed - cars travelling at 30mph usually don't hit pedestrians at that speed because the driver brakes. We'd need to know the effect at different speeds. And we'd need to know how a change in speed limit affects the actual speed. Worst case, pedestrians might get careless if cars drive slower. – gnasher729 Oct 30 '14 at 16:39
  • 1
    Both claims are also about the likelihood of a person struck be a vehicle surviving at the respective speeds, and not about aggregate pedestrian deaths as the question seems to be asking. – Flimzy Nov 2 '14 at 12:47

The two quotes say subtly different things. The second quote is much harder to examine, because talking about changes in speed limits without any information on compliance levels is useless.

So I'll tackle the first quote, that risk of fatality halves when vehicle speed is 11.2 m/s (25mph, 40km/h) instead of 13.4 m/s (30mph, 48km/h).

And that's true, at least according to a recent study used by the UK's Department for Transport in its official research.

enter image description here

is one logistic regression equation for probability of fatality, when vehicle speed was v km/h.

Roughly speaking, that works out at 3.5% for 25 mph, and 7% for 30 mph.

Source: Relationship between Speed and Risk of Fatal Injury: Pedestrians and Car Occupants, UK Department for Transport research report, 2010, quoting: Rosen E. and Sander, U. (2009) Pedestrian fatality risk as a function of car impact speed. Accident Analysis and Prevention, 41, 536–542.

  • Could you try to address the question of whether most deaths are at 30mph or not? I had a hard time finding that info. – ike Oct 30 '14 at 20:34
  • 1
    @ike - no part of the quotes you've given refer to that, they don't claim it, and they don't imply it. So it's out of scope here. FWIW, I'd expect most pedestrian deaths to be happening when the motor vehicle speed is above 30mph, because the risk of fatality in case of collision ramps up very quickly above 30mph, so though there might be fewer collisions above 30mph than below it, the higher fatality rate would probably outweigh the lower incidence rate. This would, of course, totally depend on speed limit compliance rates. – EnergyNumbers Oct 30 '14 at 20:37
  • 1
    Many accidents wouldn't even happen at lower speed. You want to cross the street, no car in sight, so you start crossing. A car going 30mph wouldn't be close enough to hit you unless you are really careless, but you might miss a car going 50mph and get hit. – gnasher729 Oct 31 '14 at 16:39
  • 1
    @gnasher729 I'm guessing you have nothing to back that up, right? – Andy Nov 25 '14 at 0:51
  • It's always like this. They quote studies on death rates for the actual collision speeds, and not for "streets with a particular speed limit". Probably because factors such as traffic volume, road width, curb design (parking allowed or not, fenced off or not, visibility etc), lighting and the overall location have a vastly greater effect on deaths per vehicle passing than the speed limit itself. If a clear correlation existed, it would be the one quoted, so I'm guessing that it doesn't. – RomanSt Dec 26 '14 at 9:46

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .