From NPR (of all places!):

But empirical studies, which seek to measure the effectiveness of ads across campaigns, suggest that these campaigns may be most effective when voters are unfamiliar with a candidate — which won’t happen this fall. When voters know a candidate fairly well, the ads don’t usually do much.

“When voters are confronted with inconvenient facts, it is oftentimes difficult to persuade them that those facts are, in fact, facts,” said George Washington political scientist John Sides. “When supporters of President Obama see negative information about Obama, they don’t think it is true. To the extent it is true, they find ways to explain it or rationalize it — they discount it.”

The same, of course, will be true for the GOP nominee come fall.

Sides argues that candidates stick with negative ads in part because they’re locked in an arms race and are unwilling to unilaterally disarm. Candidates who think they’ll be attacked by negative ads believe they need to return fire.

Another reason why campaign finance cases don’t bother me as much as they maybe should.