Clocks Fall Back on November 1. What the Research Says About Walking in the Dark

Economists and traffic researchers use the fall clock change as a natural experiment. Two risks move into the evening with the dark, and both can be measured.

On Saturday, October 31, the sun sets in New York City at about 5:53 p.m. At 2 a.m. the clocks fall back an hour, and on Sunday sunset arrives at about 4:51 p.m. A person who walks the dog at 5:30 leaves in daylight on Saturday and in darkness on Sunday, and in New York the sun does not set after 5:30 p.m. again until about mid-February. The jump is the same size elsewhere: in Fort Myers, sunset moves from about 6:45 p.m. to about 5:44 p.m.

Twice a year, daylight moves an hour against fixed schedules, and researchers use that jump to measure what light does to public safety. Two lines of research matter most to anyone who walks, runs or waits for a bus after work. One finds that reported robberies rise around sunset when the clocks fall back. The other finds that fatal crashes involving pedestrians and bicyclists rise over the following weeks. Both describe averages across many places, not the odds on your street, and both point to the same response: a lit, visible, planned evening route, with a loud personal alarm or a flashlight as backup where your route has dark gaps.

Where the risk goes

The clock change moves darkness; it does not remove daylight. When the clocks fall back, an hour of light shifts into the early morning, and the late afternoon and evening hours, when many people are commuting, running errands or walking the dog, get darker. Researchers have tied that shift to two kinds of harm. They use different data and measure different things, so we cover them separately.

Channel 1: robbery follows the dark

The most cited study comes from economists Jennifer Doleac of the University of Virginia and Nicholas Sanders of Cornell. Using national incident-based crime data from 2005 to 2008 across 558 jurisdictions, they compared the days just before and after each clock change, and compared the years daylight saving time was extended (2007 and 2008) with the same dates in 2005 and 2006. When daylight saving time begins in spring, robbery rates over the whole day fall by about 7 percent, and by about 27 percent in the evening hour that gains sunlight. The fall change runs the other way: when daylight saving time ends, reported robberies rise about 7 percent, and the increase concentrates around and just after sunset (Doleac and Sanders, Review of Economics and Statistics, 2015). The authors estimated that the 2007 extension saved at least $59 million a year in avoided social costs.

Doleac's own summary explains why the timing matters. "The timing of sunset is pretty close to the time many of us leave work," she writes, and walking to cars or homes in the dark "makes us easier targets for street criminals" (Doleac, UVA Batten School). The study found no matching rise in the morning, even though mornings get darker too. Robbery is the crime one would expect to respond most to light, because it happens face to face in public places.

These numbers carry limits. They are percentage changes on a small base: robbery is rare for any one person on any one walk. The data come from 2005 to 2008. And they describe averages across hundreds of jurisdictions, so your street can sit far from the mean. What the study establishes is direction and timing: the added risk concentrates in the hours just after sunset.

Channel 2: the road

The traffic evidence comes from federal fatal-crash records. A team from the Insurance Institute for Highway Safety analyzed U.S. fatal crashes for the five weeks before and after every clock change from 2010 to 2019. In the five weeks after the fall change, fatal crashes involving pedestrians and bicyclists were up 13 percent, while fatal crashes involving vehicle occupants were down 7.1 percent. The spring change did the reverse: pedestrian and bicyclist crashes down 24 percent, occupant crashes up 12 percent (Woods, Weast and Monfort, Journal of Safety Research, 2025).

Counted across the whole study, the clock changes were associated with a net 26 fewer fatal pedestrian and bicyclist crashes and a net 29 more fatal vehicle-occupant crashes. The authors tied the pedestrian and bicyclist change to light conditions, but linked only 2 of the extra occupant crashes to light, so most of the occupant increase is not explained by it (IIHS, April 11, 2024). The authors recommend better pedestrian and bicyclist infrastructure and vehicle technology that helps avoid crashes.

The longer trend is improving. The Governors Highway Safety Association reports that drivers killed 6,732 people on foot in 2025, 7 percent fewer than in 2024 and the third straight annual decline. Even so, about 76 percent of 2024 pedestrian deaths with known lighting conditions occurred after dark, and pedestrian distraction was cited in 29.2 percent of 2024 fatal pedestrian crashes, up from 15.3 percent in 2019 (GHSA, July 14, 2026).

The two changes hurt in different ways

Spring forward costs an hour of sleep and puts darkness back into the morning commute. Fall back returns the hour of sleep and moves darkness to the evening. In the IIHS data, the spring change was followed by more fatal vehicle-occupant crashes and fewer pedestrian and bicyclist crashes, and the fall change by the reverse. For the person walking the dog, waiting at the bus stop, or crossing a road after work, the fall change is the one that adds darkness to their hours.

What the evidence says to do

The research supports a few practical steps. None guarantees safety, and none removes risk.

None of these studies says your evening walk is dangerous. They say risk shifts into predictable hours, and that preparation can match it. For more on choosing, see the best self-defense items to carry, whether personal alarms actually work, and why people still carry pepper spray when crime falls.

The argument the data feeds

These studies feed the debate over changing clocks twice a year. The Sunshine Protection Act, which would let states adopt year-round daylight saving time, passed the House 308 to 117 on July 14, 2026 (Inquirer, July 15, 2026). The Senate has not passed it, so the clocks change on schedule this year. Nineteen states have laws that would adopt permanent daylight saving time if federal law allowed it. Supporters point to benefits like the evening-light effects above. Critics, including Sen. Tom Cotton, point to dark winter mornings, with sunrise as late as 9 a.m. in some places (KRIS-TV, July 2026). The IIHS results show the trade: more evening light helps people on foot and shifts risk toward morning crashes.

Congress decides whether that trade is ever settled. Until then the change arrives on November 1 at 2 a.m. Set your clocks, and then set your route, your visibility and your attention for the new dark.

The evidence, in one table

Sources

• Doleac, J. and Sanders, N., "Under the Cover of Darkness: How Ambient Light Influences Criminal Activity," Review of Economics and Statistics 97(5), 2015; Cornell Chronicle, "Fall back and watch out: Robbers thrive on standard time," Oct. 29, 2015; Doleac, "Fighting crime with Daylight Saving Time," UVA Batten School.
• Woods, A.N., Weast, R.A. and Monfort, S.S., "Daylight saving time and fatal crashes: the impact of changing light conditions," Journal of Safety Research, February 2025; IIHS, "Changing the clocks helps pedestrians but hurts motorists," April 11, 2024.
• Governors Highway Safety Association, "U.S. pedestrian deaths fall for third straight year," July 14, 2026.
• The Philadelphia Inquirer, report on the House vote on the Sunshine Protection Act, July 15, 2026; KRIS-TV, "House passes Sunshine Protection Act," July 2026.
• Sunset times are calculated from standard solar position formulas for New York City, Chicago and Fort Myers; timeanddate.com lists times for any city and may differ by a minute or two.