The # of clicks is just one factor used to assess the effectiveness of a UI - it is more an estimator of intuitiveness, as are tests of navigation through a text-less design. If the goal is to reduce the time it takes to perform common tasks, then the the number of clicks required shouldn't matter so much as the discoverability of components, design principles, the opportunity to convert novice users to power users, recognition over recall, etc. You should really design for what people actually do, as Andy points out, instead of optimize a route that isn't really taken as improving a user's actual task pipeline would benefit him most in the end anwyay.

If design were truly concerned with solely the reduction of steps it takes to perform a task, then surely backpack toilets would be the norm. Tongue Out