Running in the dark
Evidence: limited
Almost no research is on runners specifically. The useful findings are borrowed from controlled studies of cyclist and pedestrian visibility and from pedestrian collision statistics, which transfer in principle but were not measured on runners. The advice on seeing the ground and on personal safety is practical reasoning, not evidence-graded.
For much of the year many runners have little choice but to run in the dark, before work or after it. It brings two problems: being seen by drivers, and seeing the ground yourself. The first has some real science behind it, most of it done on cyclists and pedestrians rather than runners.
Being seen: it is about motion, not brightness
The intuitive move is to pull on a reflective vest, and it is close to the least effective thing you can do. In a night-driving study on a closed road, retroreflective strips on the moving joints, the ankles and knees, let drivers recognise a cyclist far sooner than a reflective vest carrying the same amount of material (Wood et al. 2012). The reason lies in how human vision works. A bright patch on the torso reads as an unidentified blob, whereas material on the limbs moves in the distinctive up-and-down rhythm of a person, and that biological motion, which researchers call biomotion, is what makes a driver see a human and react. Reflective material on the shoes, ankles and knees therefore beats a static vest, and both cyclists and pedestrians tend to overestimate how visible they are and to undervalue the limb markers that work best.
The material has to match the light. Fluorescent colours such as high-visibility yellow and orange work by reacting to ultraviolet light, so they help in daylight and at dusk but do almost nothing in true darkness. A daytime study found fluorescent material also worked best on the moving limbs: a fluorescent jersey plus fluorescent leg covers was recognised about 3.3 times farther away than the jersey alone (Edewaard et al. 2020). In daylight and at dusk, fluorescent colours help. In the dark, only retroreflective material and your own lights make you visible, and wherever you can, they should go on the parts that move.
Why it matters, and choosing your route
The stakes are real, though the numbers come from pedestrians rather than runners. Most pedestrian road deaths happen in darkness, night-time pedestrian fatalities have risen faster than daytime ones, and being struck in the dark, and on unlit roads in particular, sharply raises the odds that a collision is fatal (pedestrian fatality analysis, 2022). That supports the obvious defensive choices: prefer well-lit, low-traffic routes; on roads without a path, run facing the traffic so you can see and react to oncoming vehicles; and assume drivers have not seen you until it is clear they have.
Seeing the ground
A light is not only about being seen; it is about not falling. On unlit, uneven ground the risk of tripping is real, and it rises on the soft, irregular surfaces of trails and parks, though no study measures the fall rate of running in the dark specifically, so this is footing sense rather than data (see running surfaces). A head torch is the common answer. A lamp carried lower, on the chest or waist, casts longer shadows across the ground than one at eye level, which makes bumps, roots and kerbs stand out instead of being washed flat by a beam shining straight along your line of sight. On technical trail this matters far more than on pavement, where a light is mainly for being seen.
Staying safe
Beyond traffic and footing, the ordinary precautions apply, and they are practical advice rather than anything the evidence weighs in on. Run a route you know, tell someone where you are going or share your location, keep the volume low or run without headphones so you can hear what is around you, and carry a phone and some identification. None of this is specific to running fitness; it is simply how to come home safely from a run in the dark.