Seriously: there's empirical problems with the study, granted, given the arbitrary comparison of Friday 6th with Friday 13th without accounting for other external variables which would influence traffic volume and driving behaviour, but I thought it was interesting.