<p>When reality and theory collide, one of them has to budge. For as long as we have been predicting transport demand, we have been able to take certain things more or less for granted. This article questions whether those assumptions are now out of date, and whether we need to change our current approach to demand modelling.</p>