You sit at a red light. The intersection is empty. No cars, no pedestrians, no reason for you to be stationary except a coloured lamp on a pole. You wait 90 seconds, the light changes, and you proceed — mildly irritated and vaguely certain that this particular light is conspiring against you. It isn’t. But the system governing its behaviour is more complex, more politically negotiated, and more deliberately imperfect than almost anyone sitting at a red light would guess.

Fixed-Time vs. Actuated: Two Philosophies

Traffic signals operate on one of two fundamental principles. Fixed-time signals follow a pre-programmed cycle: green for a set duration, amber, red for a set duration, repeat. The durations are determined by traffic engineering studies that measure vehicle volumes, pedestrian flows, and intersection geometry, then calculate optimal phase lengths using formulas derived from queueing theory and fluid dynamics. Fixed-time signals are simple, reliable, and cheap. They are also rigid — a 90-second red phase runs its full course whether the queue contains 30 cars or none.

Actuated signals use sensors — typically induction loops cut into the road surface, or increasingly, video cameras and radar detectors — to detect the presence of vehicles and adjust phase lengths in real time. When a car pulls up to an empty side street, the sensor registers its presence and triggers a phase change. When the side street is empty, the main road retains its green phase. Actuated signals are more efficient than fixed-time systems but more expensive to install, maintain, and calibrate.

Most urban intersections in the UK and Europe use some combination of both: fixed background cycles with actuated modifications. The base timing is pre-programmed, but sensors can extend or shorten individual phases within limits set by the traffic engineer. The result is a system that adapts to real conditions but within boundaries it cannot exceed — a constrained optimisation that frustrates drivers precisely because they can see it’s almost responsive but never fully so.

The Green Wave and Why It Barely Works

The “green wave” is the traffic engineering holy grail: coordinating a sequence of signals along a corridor so that a vehicle travelling at a designated speed encounters green lights at every intersection. The concept is elegant and, under ideal conditions, achievable. Time the offsets between successive signals to match the travel time between intersections at a target speed — typically 30 to 50 km/h in urban areas — and traffic flows without stopping.

In practice, green waves are fragile. They work in one direction at a time: optimising southbound flow on a corridor typically disrupts the timing for northbound traffic. They assume uniform travel speeds, which real traffic doesn’t maintain. They break down at intersections with crossing traffic from side streets, where conflicting demands force compromises in phase timing. And they require signal spacing that matches the target speed — a condition that historical street layouts, designed long before traffic signals existed, rarely provide.

London’s SCOOT system (Split Cycle Offset Optimisation Technique), deployed across approximately 12,000 signals, continuously adjusts cycle lengths, phase splits, and offsets based on real-time data from road sensors. It achieves an estimated 12 percent reduction in average delay compared to fixed-time plans. Twelve percent is significant at city scale — thousands of cumulative hours saved daily — but imperceptible to the individual driver who is still sitting at a red light wondering why this intersection hates them specifically.

The Pedestrian Negotiation

Every traffic signal negotiates between competing demands: through traffic, turning traffic, pedestrians, cyclists, and increasingly, bus and tram priority. Pedestrian phases are the most politically contentious element of signal timing because they impose the largest time cost on vehicle flow. A pedestrian crossing phase at a busy intersection typically requires 8 to 15 seconds of green walk time plus 5 to 10 seconds of flashing clearance, during which all conflicting vehicle movements must stop. At a four-way intersection with pedestrian phases in all directions, pedestrian time can consume 30 to 40 percent of the total cycle.

The “beg button” — the push-button at pedestrian crossings that requests a green walk signal — is a source of enduring public frustration. Many pedestrians suspect the buttons don’t work, and in some cities, they’re correct. In New York City, the Department of Transportation acknowledged in the early 2000s that the majority of pedestrian push buttons in Manhattan had been deactivated following the installation of automated timing systems, but the physical buttons were left in place because removing them was deemed more expensive than leaving non-functional hardware on the poles.

In the UK, most pelican and puffin crossings have functioning push buttons, but the system imposes minimum wait times and maximum frequency limits to prevent pedestrian phases from disrupting vehicle flow excessively. Press the button and you may wait 30 to 90 seconds, depending on where the vehicle cycle is when the request is registered. The wait is not random. It is a calculated minimum designed to protect vehicle throughput at the expense of pedestrian convenience — a political decision embedded in an algorithm and disguised as a technical constraint.

Why Some Intersections Feel Personal

The conviction that a particular traffic light is unusually slow or deliberately obstructive is nearly universal, and it is not entirely irrational. Signal timing reflects political and engineering priorities that vary by location. An intersection near a school may have longer pedestrian phases. An approach to a motorway junction may have extended green times for the main direction. A recently developed residential area may have shorter cycles to calm traffic. Each of these decisions is defensible in isolation, but the driver experiencing them sequentially perceives not a series of local optimisations but a conspiracy of red lights.

Cognitive bias amplifies the perception. Psychologists have documented that people overestimate the duration of unpleasant waits (red lights) and underestimate the duration of pleasant non-events (green lights). A study published in Transportation Research found that drivers consistently estimated red-light wait times as 30 to 50 percent longer than actual duration when measured objectively. The light isn’t longer than you think. You think it’s longer than it is, because waiting at an empty intersection with a red light is one of the most vivid daily experiences of powerlessness over time.

The Infrastructure Nobody Thanks

Before traffic signals, intersections were governed by right-of-way conventions, police officers, and nerve. The first electric traffic signal was installed in Cleveland, Ohio, in 1914. Within fifty years, signalised intersections had reduced urban traffic fatalities in equipped areas by estimates ranging from 15 to 40 percent, depending on the study and the city. The traffic light is one of the most effective safety interventions in the history of urban infrastructure, and its daily operation saves lives at a rate that would make any other public health measure envious.

None of this is visible to the driver at the red light. What’s visible is an empty intersection, a coloured lamp, and 90 seconds of enforced stillness. The system behind the lamp — the sensors, the algorithms, the political negotiations, the pedestrian safety calculations, the coordinated offsets along the corridor — is invisible by design. Traffic signals work best when nobody thinks about them. The frustration you feel at the red light is, paradoxically, evidence that the system is working: you’re alive, the intersection is safe, and the worst thing that happened to you today was waiting 90 seconds for a lamp to change colour.