Roads are one of the great accomplishments of powerful, long-lasting, and well-organized civilizations. Medieval roads were generally awful, unpaved, and poorly kept up. The idea we have of ancient highways stretches back into the classical world where the Roman road system was one of their most celebrated accomplishments, bridging the great distances behind the imperial borders with safe, easy-to-travel, easy-to-use, well-maintained roads. They are a hallmark of organizational ability, for they require labor and material at a constant rate to be maintained. The best medieval folks could generally manage was a stone bridge or two.
The maintenance of roads and highways are a measure of state power, and the selection of which tracks and trails are chosen to become roads is predicated on a number of factors: political, mercantile, and most of all, military. The swift transport of troops is the primary reason for a state to construct and maintain roads, with almost all other considerations being mostly secondary. Of course, once the major military routes are paved and a road-maintenance technology and labor organization is developed and secured, they can be expanded to cover trade routes.
Indeed, the presence of tracks can be a good measure of traffic and can even hint at the size of a settlement. Major markets and traffic centers tend to have a number of trails snaking out from them in all directions, while smaller towns will generally have much fewer and require fewer links. This is a function of economics, and one which has been worked out by greater minds than I. Breaking and bulking centers scale in size from the local, to regional, to extra-regional. Each stage of center has additional road-attachments that lead to and from both its periphery and to the larger markets, operating on a scale that seems almost fractal and only breaks down at the most local of levels.
As farms are attached to a town center by tracks, that town center is attached to its local market by a trail, and that market attached to all town centers in its periphery. Furthermore, that market-center is attached to a regional marketplace, which in turn may be attached to an extra-regional market center, which in turn is attached by trade-routes to other extra-regional centers.
Thus, a center with a lot of roads or tracks snaking off of it will likely be important to the surrounding population. Indeed, one might roughly estimate the size and variety of goods available in a settlement based on the number of roads spiraling off to subsequent marketing centers.
Of course, all this data is based on stuff I read three years ago in graduate school, and it may very well have changed since then. If anyone out there knows a better economic theory to simulate a medieval trade economy, I'm all ears!