In a campaign spending contest framework, we investigate the potency of frequently cited sources of incumbency advantage in generating the observed patterns of campaign expenditure and reelection rates. Since one Dollar of extra spending should not change a certain loser to a certain winner, we extend the literature by allowing the electoral benefit of visibility to be stochastic. The model provides an explanation of earlier empirical findings that district variation in the cost of visibility does not influence incumbents’ victory probability. Furthermore, in contrast to previous literature, the model predicts that campaign finance legislation can increase challengers’ expected payoffs.