Cerberus said:
And you're getting those numbers from what source exactly?
Mathematics?
The type of bike helmets worn in road cycling are basically barely meeting the testing protocol standards. They are only capable of absorbing the energy of a fall over type impact: any additional energy is going to be transmitted to the head. If you aren't moving and just fall over, your head will hit the ground at roughly 12mph, with about 1300 joules of energy. If your head is going 30mph when it impacts something, that's about 6150 joules. The helmet will potentially subtract its 1300 (if worn properly, perfectly positioned, etc.), leaving you with a 4850 joule impact. That's the same as from a helmetless 27mph impact.
This should be common sense. Hold your helmet in your hand. Ask yourself how hard would it be to tear that thing to shreds? We can easily do it with our bare hands or just by stepping on it lightly, right? Well, that minor amount of energy is all that is going to be removed from our crash scenario by the helmet. Our body will have to deal with the rest: luckily it's pretty good at doing that.
Do you think Voigt Crash this year would have worked out as relatively well as it did if he hadn't been wearing a helmet?
Yes. He might have had more roadrash type injuries, but I believe his actual head injury - brain injury - would have been basically the same. Bike helmets are not designed to lessen concussion type injuries. The worst brain injuries are caused by the brain rotating in the skull and ironically/tragically, helmets actually increase the rotational energy.
If you're interested in reading more about bike helmets, the wikipedia article is pretty objective and the claims made in the article are well supported by footnotes to the direct sources. There are over 100 footnotes and you can always go read some of those articles if you want to see the research directly.
http://en.wikipedia.org/wiki/Bicycle_helmets