In their efforts to create a position in a market, and to maintain that position, firms make positioning investments of various sorts, in R&D, plant, advertising, and location, or more generally, in product development and maintenance. The heart of this paper is the hypothesis that the success of these positioning investments is not assured. In an environment where the success of positioning investments is stochastic, the positioning game played by firms that compete to serve a market is necessarily dynamic. We model the positioning and operating decisions of firms in an environment of this sort. When the market is large enough to support at least one active firm, the expected number of firms serving the market at a point in time is a nearly continuous function of market size, in sharp contrast to the familiar integer valued step function seen in classic models of market structure. As a result, equilibrium expected total surplus and expected consumer surplus are higher than standard non-stochastic models would
suggest, especially in circumstances where the expected number of firms is small. This suggests that classic models of market structure are not always a sound guide for policy.