You've supplied ancedotes. That is not evidence. The plural of ancedote is
not "data". You could have examined the actual published medical evidence:
Goldacre (of Bad Science fame) provides an excellent summary of the actual data, the summary of which is that there is no good data to support the idea that helmet wearing reduces injury.
And here is another issue. Two in fact. The first being that most helmets are exceedingly limited in what they protect: the conventional design will not protect the face: full face helmets are not popular even in mtbing.
The other issue is that to induce severe brain injury requires on average 10 times as much energy than a helmet rated to EN1072 will provide. (Snell rated helmets are a little better, but not significantly). It therefore is highly unlikely that wearing a helmet would have changed the outcome for your coma victim. Thank you for at least removing the unnecessarily nasty last line.
ETA:
@JoeyB, the Goldacre editorial I've linked to above is a very good place to start for a short but lucid review of the available evidence.