Remember kids... always wear a helmet. (The almighty bikeradar helmet thread)

But I think it warps the issue and relies on emotions rather than data, which is what is seriously lacking in this debate.

There are far more variables that affect the outcome of a bike accident than say a car crash. Most car crashes happen in predictable ways - run into something, get hit from the side, etc, (with the speeds varying) and with passengers sitting in the same position so you can replicate what will happen in a car crash knowing that it will apply in many cases.

There is no typical bike accident - you can hit / be hit from any angle, be standing, sitting, leaning, hit the ground in a multitude of ways, land on different materials etc. Testing one thing will hardly have any relevance to other cases.

So for that reason, it's best to rely on raw emotion, hearsay and first person experience.