In a 1997 article in the American Political Science Review, I contend that the success of the new, directional model of electoral choice is illusory rather than real. When appropriately evaluated, the classical Downsian proximity model remains the better theory. In a subsequent article, in the Journal of Politics, Macdonald, Rabinowitz, and Listhaug (1998) contend that I am nowhere right. Soon thereafter, the empty center in the controversy between us is seized by Lewis and King (1999), who contend that it is impossible to know who is right, and by Merrill and Grofman (1999), who contend that we are both partly right. In this article, I examine the arguments of my directional opponents. In so doing, I also keep an indirect, and occasionally direct, eye on the arguments advanced by the two center parties. The results explain why my initial contention remains more justifiable than those subsequently offered by other contenders. I conclude by spelling out the principal reasons why the directional theory itself, as well as the methods to which it owes its apparent success, represent steps in the wrong direction in our search for a better understanding of the process of electoral choice. Any additional comments I may have after having read the text by Macdonald et al. expected to appear next to mine can be found in Westholm 2001d.