Unit distance seems reasonable as long as conditions are genuinely like for like. The links Nick Finnigan gave suggests conditions are not too dissimilar then.
However the reason for my scepticism about vehicle distance comes from looking at French data at where the data is broken down by department. We are not told about vehicle types but we get 100 data points varying between rural, lightly populated areas where people drive a lot to densely populated urban areas where distances are less.
3 accidents in a weekWhat is it about this time of year and road accidents? In the last week I have witnessed 2 accidents and come across a third just after it happened. Normally I can go for...
Plot things and a few things stand out. When looking at data for all road users, baneities increase as people drive further. Yet SI remain constant at between 2k-3kmillion head over more than a 4 to 1 vehicle-distance range
Looking at pedestrians, banes are a steady 10-20million head even though vehicle distance varies over the same range.
So I think it is necessary to home in closely to where the pedestrians are, not just where vehicles go.
It is plausible that a head position the same height as the standing person provides some advantage. Bus & coach figures don't look too good though but I quess one unfortunate bus queue adjusts figures significantly.
But it is just one hypothesis.
Another could be that pedestrians are less likely to linger in front of something large and intimidating.
