Will late Trump or Biden TV ads change the race?

If you are unlucky enough to live in one of the dozen or so states where the campaigns of President Donald Trump and former Vice President Joe Biden are duking it out on the airwaves, you've probably already had your fill of political ads. Hang on, there are less than two months to go.

The unfortunate part of all the presidential advertising is that most of it is a waste of money.

The ads present their own nominees in the best possible light as they attack the opponent's judgment, integrity, values and records. But after what appears to be an interminable presidential race, most voters know how they feel about Trump and Biden and probably how they will vote.

And those who are torn or unenthusiastic (or flippable, even at this last date) are going to be affected more by real news — Bob Woodward's book "Rage," the presidential debates or other new information that hits between now and November 3 — rather than 30-second TV ads.

People understand most ads are little more than each side trying to sell a product. But in this case, the voters have seen the products over a long enough time and in enough situations to make up their own minds.

My assessment (recently on Twitter) that presidential ads are overrated drew a quick response from veteran Democratic message-maker David Doak, who observed that while the ads "may be less effective at persuasion per se, they can be helpful on 'controlling the definition' of what the election is all about (focusing the electorate on the issues that favor your candidacy), and at countering false negative attacks."

Both of David's points strike me as important.

I've often said that most elections are "about" what the election is about. Is it an election about "law and order," or the economy or the government's response to the coronavirus? Or maybe it's about the character and integrity of the sitting president or health care. You give me the answer, and I'll tell you who is likely to win.

So, David's point is on the mark.

My only response is that if both campaigns are comparably funded and well-run (yes, I know, that's quite a caveat), then the two campaigns simply offset each other's definitional argument in the ads — and voters end up deciding what is more important to them, as they would have without both sides spending millions of dollars.

David's second point is also correct. Ads are one way of rebutting misleading or untrue charges during a campaign. If Trump says something untrue about Biden (for example, that the former vice president wants to defund the police or impose Medicare-for-all), Biden can respond with TV ads correcting the record.

But the attack/correct the record/attack/correct the record cycle that David points to also can merely produce more confusion about what the candidates stand for, not less. Each side, after all, can both attack and defend in TV spots, forcing voters to default to their own views of which candidate is most appealing and trustworthy.

Obviously, if one side has financial resources and the other doesn't, traditional 30-second ads can have an impact, as they did in 1996, when Bob Dole's presidential campaign ran out of cash late in the primary season.

Campaign ads are also useful telling us the lay of the electoral land, including where the race will be decided.

And political ads certainly can have more of an impact down ballot, in races where the candidates are less well known. Trump has been on television almost every day for the last four years, and that has helped voters to come to conclusions about him. But nominees for Congress, the state legislature or local office often don't have that visibility, so ads can provide information that most voters don't already have.

There is always the chance that a compelling ad can impact the narrative of the race. Not all TV spots are equally effective in moving voters. And even if ads in September and October move just a handful of voters in a handful of crucial states, they could prove to be decisive in the outcome if the presidential election is as close as the elections of 2000 and 2016 were.

Just don't count on it.

Stuart Rothenberg wrote this piece for CQ-Roll Call.