Owners describing a fault almost always say where they were. That detail turns out to predict the outcome better than the fault itself does.
| Where it happened | Complaints | Crashes | Crash rate |
|---|---|---|---|
| Junction | 14,384 | 1,831 | 12.7% |
| Car park | 17,972 | 2,210 | 12.3% |
| Driveway | 18,937 | 2,249 | 11.9% |
| Garage | 20,475 | 2,124 | 10.4% |
| Merging | 25,720 | 2,631 | 10.2% |
| In traffic | 53,186 | 4,292 | 8.1% |
| Motorway | 112,724 | 5,833 | 5.2% |
The motorway appears in eight times more complaints than junctions do and produces less than half the crash rate.
Speed is not the variable
The intuition says the fast road is the dangerous one. The data says the opposite, and the reason is what surrounds the car rather than how quickly it is going.
A motorway gives a failing vehicle almost everything it needs. Traffic moves one way, lanes are wide, there is usually a shoulder, and a car that loses power at seventy still has several hundred metres of momentum to steer with. The driver has time to make a decision.
A junction gives none of that. The failure happens with cross traffic already committed to its own path, with no shoulder, and with a decision required in under a second. Losing the engine while pulling across a road is a different problem from losing it in lane three.
Car parks and driveways sit near the top, which is stranger
Speeds there are walking pace, and yet the crash rates are 12.3% and 11.9% — more than double the motorway.
What those places have is obstacles at close range and people on foot. A car that moves unexpectedly in a car park does not need speed to hit something; everything is already within a few metres.
This is also where unintended movement concentrates — the rollaway complaints, the vehicles that crept out of driveways, the ones that went forward when reverse was selected. We found the same pattern in parking brake complaints, which record a 13.7% crash rate, the highest of any braking component.
The severity runs the other way
Crash rate is not the whole picture, and the injury figures show why.
Motorway incidents produce 5,601 injuries across 5,833 crashes — close to one injury per crash. Car park incidents produce 1,220 injuries across 2,210 crashes, roughly one per two.
So the junction and the car park are where a fault is most likely to hit something, and the motorway is where hitting something hurts most. Both facts matter and neither cancels the other.
What to do with a car that starts misbehaving
The instinct is to keep going to somewhere convenient — the next exit, the nearest car park, home.
The data argues against all three. The risk concentrates where the vehicle has to interact with cross traffic, pedestrians and obstacles at close range, which is exactly what "somewhere convenient" means.
Stopping in a straight line on a wide road, with hazards on, is the option the numbers favour. It is inconvenient and it is where the crash rate is lowest.
And if the fault is one the manufacturer already knows about, the repair is free — a free VIN check shows the open campaigns for that specific vehicle.
Frequently asked questions
Where do car faults most often cause crashes?
At junctions. 12.7% of complaints mentioning an intersection also record a crash, against 5.2% for complaints mentioning a motorway — despite the motorway appearing in eight times more reports. Car parks follow at 12.3% and driveways at 11.9%.
Why is a motorway fault less likely to end in a collision?
Space and predictability. Traffic moves in one direction, lanes are wide, there is usually a hard shoulder, and a driver who loses power has hundreds of metres to steer somewhere safe. A junction offers none of those — the failure happens with cross traffic already committed.
Are low-speed faults actually dangerous?
The crash rate says yes, though the severity is different. Car parks and driveways record 12.3% and 11.9% crash rates, but far fewer injuries per crash than motorway incidents — 1,220 injuries across 2,210 car park crashes against 5,601 across 5,833 motorway ones.
What does this mean for how I react to a fault?
Do not aim for the next exit if the car is misbehaving in traffic. The data says the risk concentrates where the vehicle has to interact with cross traffic and pedestrians, so stopping in a straight line on a wide road is safer than continuing to somewhere convenient.
Keep reading
All articles →
36,815 Owners Say the Same Fault Came Back After Repair
A repair either works or it does not, and the complaint database records a large number that did not. What makes these different from ordinary faults is that somebody had already been paid to solve them.

70,474 Complaints End With the Car on a Flatbed
A breakdown is not the same as a defect, and most faults never strand anyone. These are the ones that did — and the age at which they did it is younger than most people would guess.

24,701 Complaints Mention a Spouse. 875 Mention Being Pregnant.
A complaint form asks what failed. A large number of people answer by explaining who was sitting next to them, which turns out to change how the report reads and what it is useful for.
Mark ForCar as a preferred source and our data shows up first in your Google results. One click, no account.

