Could the Florida election disaster have been avoided by competition among polling stations?

We study a model of competition among polling stations where polling stations have to invest in counting precision to attract voters. The benchmark is the (current) monopolistic case in which there is no incentive to provide particularly good counting technologies. Unsurprisingly, we find that competition enhances precision when compared to a monopoly. The somewhat surprising part of our result is that the optimal number of polling stations is two.