>>  Yerbut I don't think it helps to say "cycling is safer than walking
>>  because KSI per mile travelled is less" when KSI per hour may well be
>>  more. "Cycling is generally no more dangerous than walking" might be a
>>  better angle, then there's less reason getting bogged down in stats
>>  trying to prove one is "safer" than the other.

> You mean Phil should have said something like "the whole picture is that
> neither cycling nor walking are particulary dangerous. One of them is
> probably more dangerous, but it will be a close thing.", like, err, Phil
> did?