The Passenger Side Was Tested Fifteen Years After the Driver's. It Scores Better.

The small overlap test is the hardest crash evaluation in common use. Only a quarter of the front of the car hits the barrier, outside the main structural rails, so the energy goes straight into the wheel, the firewall and the footwell.

It was introduced on the driver's side. The passenger side was added much later, on the reasonable-sounding assumption that a symmetrical car would behave symmetrically.

It does not, and the direction of the difference is the surprise.

The side that was tested second fails less often than the side that was tested first.

The gap is a timing artefact, not engineering

The explanation is not that the passenger side is built stronger. It is that the passenger side was never rated during the years when cars were failing.

When the driver's side test arrived, the existing fleet had not been designed for it, and failures were widespread. Manufacturers then spent years reinforcing that corner. By the time the passenger side test appeared, the structural fixes were already in the cars — and in many designs they had been applied to both sides, because it is cheaper to make a symmetrical part than two different ones.

So the passenger side inherited a decade of work it was never measured for. The 10% figure describes cars that were already fixed.

Where the two sides genuinely differ

Not everything transferred. The cases that fail on the passenger side and pass on the driver's side usually involve the airbag rather than the structure.

Driver airbags are engineered around a fixed seating position: the wheel is in one place and the occupant is roughly in front of it. A passenger sits further from the dashboard, with a wider range of positions, and the curtain and knee bag coverage is less specific. A structure that holds up can still allow enough movement for the head to miss the airbag entirely.

What this says about safety testing generally

The pattern here repeats across the whole rating system, and it is worth naming because it changes how you should read any safety score.

Cars improve on the thing being measured, quickly, and do not improve on the thing that is not. Head restraints went from 88% failing to zero once rated. Headlights halved once the award required them. The rear seat failed immediately when a dummy was finally put in it, after decades of nobody looking.

A high safety rating is evidence about the tested scenarios and silence about everything else. That is not a criticism of the rating — it is what a rating is.

For a used buyer

Check which version of the test the rating came from. A car from before the passenger side evaluation existed has a driver-side result only, and its passenger corner has never been independently measured.

For most model years the structure is shared and the result would be similar. But it is an assumption, not a measurement, and the difference matters most on exactly the cars where it is hardest to check: older ones.