Why Your Bike Lane Ends at Every Intersection (And How the Dutch Fixed This)

TL;DR;

• In most North American cities, bike lanes “give up” at intersections because older design manuals prioritized turning cars and vehicle throughput over continuous protection for cyclists.¹²
• The default fix has been the mixing zone: drop the bike lane and force riders to merge with turning traffic—something only the most confident riders enjoy.³⁴
• Dutch engineers solved this decades ago with protected intersections that keep physical separation, slow turning speeds, and make priority unambiguous using setbacks, corner islands, and “shark teeth” yield markings.⁵⁶
• These junctions support very high cycling mode share (over 25% of trips nationwide in the Netherlands) while delivering far lower per-km injury and fatality rates than car-centric countries.¹⁵
• Until your city upgrades the hardware of its streets, riders still benefit from better “software”: defensive positioning, good lights, and audible warnings like car-horn-loud bicycle horns that drivers instinctively respond to.⁷

“We shouldn’t ask, ‘Why are Dutch intersections so safe?’ but ‘Why did we ever accept anything less?’”
— Paraphrasing many Dutch traffic engineers

1. Why your bike lane dies at the corner

If you ride in almost any North American city, the pattern is familiar:

• Mid-block: a decent painted or even fully protected bike lane.
• 100 feet from the intersection: the protection ends, the lane jogs toward traffic, maybe some green paint appears.
• At the corner: you’re suddenly negotiating with turning cars in the exact place crashes are most likely.

This isn’t accidental; it’s baked into how streets were designed for decades.

1.1 We designed for cars first, then tried to fit bikes in

Post-war traffic engineering in the US, Canada, and much of the UK centered on a simple hierarchy: keep cars moving, then everything else.¹ Intersections were optimized for:

• High turning speeds (wide curb radii, sweeping right turns).
• Extra turn lanes at corners, often carved out of curb space.
• Signal timing favoring vehicle throughput and “level of service” grades.

When bikes began to be re-introduced in design manuals, the dominant paradigm was vehicular cycling: riders were encouraged to “act as drivers” and merge into traffic at intersections to avoid right-hook collisions.² Instead of fixing the geometry, we asked human beings on 15 kg vehicles to behave like 2-ton steel boxes.

The result: many design guides literally told engineers to drop the bike lane before the intersection and merge riders into a shared lane with turning cars.

1.2 The “mixing zone” problem

Today this practice has a friendly name: the mixing zone.

A mixing zone is created when a bike lane disappears shortly before the intersection and turns into a shared lane from which drivers can turn right, often marked only with a “sharrow.”³⁴ On paper, it’s supposed to encourage eye contact and negotiation. In reality, it:

• Concentrates conflict exactly where sight lines are worst and speeds are higher.
• Depends on driver courtesy and attention in a world of distraction and big vehicles.
• Feels terrifying to risk-averse riders, effectively excluding kids, elders, and many women from cycling.¹

Research from New York and elsewhere comparing mixing zones with fully separated signal phases has found more predictable operations and improved safety when conflicts are separated in time or space rather than left to informal weaving.⁴⁸

1.3 Intersections are where crashes cluster

Mid-block protected bike lanes do a good job reducing crash risk where they exist, but intersection design often undermines those gains. The US National Transportation Safety Board’s 2019 bicyclist safety report highlights intersections as key conflict points and calls for better integration of separated bike lanes with intersection treatments.⁸ Dutch crash analyses similarly show that four-way junctions on 50 km/h roads are where many collisions concentrate, even if severity per crash is not always higher than on straight segments.⁹

In other words: if your protection ends at the corner, you’ve left the riskiest part of the trip unsolved.

2. How the Dutch fixed the intersection

The Netherlands tried the “cars first” model too. In the 1960s–70s they widened roads, removed tram lines, and treated bikes as second-class traffic. Cycling levels fell and deaths rose sharply, sparking the Stop de Kindermoord (“Stop the Child Murder”) movement and a political shift toward what’s now called Sustainable Safety.⁵¹⁰

Instead of asking cyclists to adapt to car-centric intersections, Dutch engineers rebuilt the intersection around human vulnerability.

2.1 Core principles: Sustainable Safety at junctions

Dutch Sustainable Safety policy boils down to a few principles that matter a lot at intersections:¹⁰

1. Functionality: Through-traffic belongs on distributor roads; local roads should be low-speed, low-volume access streets.
2. Homogeneity of mass and speed: Where bikes and cars must interact, speeds and directions should be predictable and moderated.
3. Predictability & recognizability: Road users should be able to “read” priority at a glance.
4. Forgiving design: Human error is inevitable; the street should prevent it from becoming fatal.

A protected intersection is just those principles drawn in paint, concrete, and signals.

2.2 Anatomy of a protected intersection

A classic Dutch protected junction keeps cyclists and pedestrians physically and visually separated from turning cars while still fitting into a fairly standard urban footprint.⁵⁶

Key ingredients:

1. Set-back crossings (“bend-out”)
   The cycle track and crosswalk are pulled back from the main carriageway by about one car length. This creates a waiting space for turning drivers to stop and look for people crossing without blocking through traffic. It also slows turning speeds and improves angles of visibility.

2. Corner refuge islands
   Concrete islands at each corner tighten the turning radius and create a protective pocket for people waiting to cross. They also physically separate the path of cyclists from that of turning vehicles, so conflicts happen slowly and at right angles instead of shallow, high-speed cuts.⁵

3. Continuous cycle tracks with clear priority
   The bike path doesn’t disappear; it bends around the corner. Pavement color (often red asphalt) and strong markings make it visually continuous. Priority is made explicit with “shark teeth” yield triangles and side-street stop or yield control.⁵⁶

4. Signal timing that protects the vulnerable
   On larger junctions, Dutch signals often give bikes and pedestrians a fully protected phase or at least a leading interval, so they’re clear of conflicts before turning traffic moves.⁶¹¹ Phasing is designed to keep conflicts simple, not to maximize car throughput at all costs.

5. Moderate approach speeds
   Approaching lanes are narrow and often slightly deflected; roundabouts or raised tables may substitute for traditional signalized junctions on lower-speed roads.⁵

The result is an intersection that feels easy to use even for an eight-year-old on a bike or an eighty-year-old with a walker—hence the Dutch mantra: design for “8 to 80.”

2.3 What changes at your corner?

You don’t need a blank slate to get most of the benefits. A typical North American four-lane arterial with parking and a painted bike lane can usually host a Dutch-style protected intersection simply by reallocating space.⁶

Here’s a simplified comparison:

Protected intersections are now being piloted and built across North America, inspired directly by Dutch examples.⁶¹² Design guides from FHWA, NACTO, and several states now include them as standard tools rather than exotic experiments.³⁸¹¹

3. Why this matters for safety and ridership

3.1 High cycling levels and low risk

International comparisons consistently show that countries with extensive, high-quality cycling infrastructure—including safe intersections—achieve both high bicycle mode share and relatively low per-km injury risk.¹¹⁰¹³

Pucher and Buehler’s classic analysis of the Netherlands, Denmark, and Germany found that protected facilities along busy roads and at intersections are central to making everyday cycling “irresistible” and safe across ages and genders.¹ More recent work on cycling safety in the Netherlands underscores that further gains will come from tackling remaining conflicts at junctions and high-speed rural roads.¹⁰

In short: the intersection is the last big piece of the safety puzzle, and the Dutch have already shown how to fit it.

3.2 From “strong and fearless” to “all ages and abilities”

Surveys in North America often categorize potential riders into groups: “strong and fearless,” “enthused and confident,” and the much larger “interested but concerned.” The mixing-zone model effectively designs only for the first group.

By contrast, protected intersections:

• Remove the need to merge into fast traffic.
• Make priority legible with consistent geometry and markings.
• Work for children biking independently and for older adults on e-bikes or mobility devices.

That shift is critical because cycling safety improves when more people ride and when a broader slice of the population is represented.¹⁰¹³

3.3 Until the hardware changes: what riders can do now

You shouldn’t have to “hack” your way through dangerous intersections—but while many cities upgrade slowly, there are still tools that can improve your odds:

• Lane positioning: Taking the lane early on narrow approaches can sometimes reduce close passes and ambiguous merges.
• Defensive timing: Hang back from large vehicles at corners; don’t assume a turn signal (or the lack of one) reflects what will actually happen.
• Visibility: Good lights and reflective elements help, especially in the dark, wet conditions where many crashes occur.⁸

And because drivers’ brains are trained to respond to certain sounds, an audible warning in the “language of cars” can be a last-line defense when someone starts turning across your path. Car-horn-loud bicycle horns (like the Loud Mini) are specifically designed to trigger that rapid, instinctive braking response drivers already have for horns, not for polite bells.⁷ Used sparingly and only in real conflicts, they’re a way of borrowing drivers’ own safety reflexes while we push for streets that won’t require them.

4. What cities can do tomorrow

The Dutch approach isn’t magic; it’s a design pattern that can be copied, adapted, and improved.

1. Stop giving up at the intersection Follow NACTO’s blunt advice: “Don’t give up at the intersection.” Continue the bikeway’s protection all the way to the crossing instead of dropping riders into mixing zones.¹⁴

2. Pilot low-cost protected corners Use modular islands, planters, and temporary curbs to tighten turns and create set-back crossings. Measure yielding, speeds, and conflicts before and after.

3. Align signals with human vulnerability Retime phases to give bikes and pedestrians leading intervals or protected phases where volumes justify it. Resist the urge to optimize only for car delay—small reductions in turning speeds translate into huge gains in survivability.¹¹¹⁵

4. Update local design guides Explicitly include protected intersection templates, with details on radii, setback distances, and markings. The FHWA Separated Bike Lane guide and updated NACTO Urban Bikeway Design Guide both provide solid starting points.³¹¹¹⁴

5. Design for the “interested but concerned” Evaluate intersections not by how confident riders feel, but by whether a cautious 12-year-old or their grandparent would comfortably use them. If not, it’s not done.

Intersections are where your bike lane’s promise is tested. The Dutch have already shown that the lane doesn’t have to end there—it can carry you safely through.

FAQ

Q1. Aren’t protected intersections bad for car traffic?
A. When designed well, they can keep turning movements organized and reduce unpredictable weaving, which often improves overall flow even while lowering speeds. Some junctions may see slightly longer car delays, but the tradeoff is fewer serious crashes and a street that works for more people.

Q2. Do you always need signals for a Dutch-style intersection?
A. No. On lower-volume streets, Dutch practice often uses compact single-lane roundabouts or priority-controlled junctions with set-back crossings and clear yield markings instead of signals; the geometric protection remains, just without the lights.

Q3. Is green paint in the conflict zone “good enough”?
A. Colored pavement helps mark the bikeway, but by itself it doesn’t slow turning speeds or physically separate movements. It’s far more effective when combined with geometric changes—corner islands, tighter radii, and setbacks—rather than as a paint-only fix.

Q4. What if there isn’t enough space for a textbook protected intersection?
A. Many elements scale: you can still tighten turn radii, pull the bike crossing back a few meters, and use small islands or flexible posts to clarify paths and priorities, even on constrained streets.

Q5. Do riders still need bells or horns at protected intersections?
A. Yes. Good design reduces conflict but doesn’t eliminate human error. A polite bell is great for people walking or rolling; a louder, car-like horn can be a lifesaver when a driver begins turning across your right-of-way and needs to react now.⁷

References