In the general campaign – late August to Election Day – three phases stand out. Later chapters reveal that the first phase corresponded to predictions from forecasting models that emphasize the economy, that the second was induced by a shift in evaluations of candidates' personal characteristics, and that the third turned on a battle over issues. In this chapter, however, the focus is on what vote intentions tell us by themselves. In the first phase, Al Gore was clearly ahead, about as far ahead as forecasting models said he should be. In the second he was clearly behind, almost as far behind as earlier he was ahead. In the last phase, he drew even. The transition from the first phase to the second was abrupt and unmistakable. The transition from the second phase to the third is more debatable. Within each phase, less consequential shifts also occurred, but none lasted more than a few days and none fundamentally altered the course of the campaign.

Movement, whether enduring or temporary, in the candidates' vote shares was asymmetric. Shifts in Al Gore's share were not just the mirror image of shifts in George W. Bush's share. Until the very end, when the system appeared to undergo a qualitative transformation, Gore's share was more mobile than Bush's. Change in the relative standing of the two candidates stemmed heavily from the ebb and flow in Gore's own support.