We have reached that point in the 2024 election cycle when it is too late to be early and too early to be late. With summer approaching and the first debate ever between an incumbent president and a convicted felon looming, this seems like the ideal moment to propose 10 rules for following the campaigns. 

1) Remember that presidential polls on the eve of the election can be very wrong. 

Pollsters have been muffing it since the Literary Digest survey in 1936 awarded the election to Alf Landon, who ended up carrying just two states against President Franklin D. Roosevelt. In 2012, the final Gallup Poll gave a 1-point lead to Mitt Romney, who lost to Barack Obama by 4 points. And then, as Democrats remember with horror, there were the misfiring 2016 polls.

Five months before the election, horse-race polls have the accuracy of a nearsighted archer in a thunderstorm. 

2) Never trust a leaked poll from inside a campaign.

With local newspapers dying, there is a shortage of public polling on contested House races and even some Senate battles. That void creates an irresistible temptation to rely on internal campaign polling provided to the press. 

Even if the numbers are 100 percent accurate, there is no way of knowing if the poll is an outlier, since no campaign releases all its surveys. In short: If a poll is leaked, there is always a reason beyond the generosity of a campaign.

3) Be cautious about electoral models, since it is impossible to replicate the idiosyncratic factors that are shaping the race between Joe Biden and Donald Trump. 

Politics is far different than it was when Grover Cleveland ran against incumbent Benjamin Harrison in 1892 in a replay of the 1888 election. Or when Ronald Reagan deprived Jimmy Carter of a second term in 1980.

Since Fox News was launched a month before the 1996 election, there have been just three other campaigns involving incumbents (2004, 2012 and 2020). Social media has an even shorter political history: Only two presidents have sought reelection since Facebook was opened to a general audience in 2006. 

4) Don’t overreact to what’s shown on Fox News and MSNBC.

A recent paper by political scientists David Broockman and Joshua Kalla found that most voters have better things to do with their time than obsessively watch cable TV. The same, alas, cannot be said about most political reporters. 

Using proprietary Nielsen data, Broockman and Kalla found that only 30 percent of self-identified Republicans watch Fox News for more than an hour per month (or two minutes per day). For Democrats, only 27 percent watch either CNN or MSNBC for a similar amount of time. 

5) The effects of TV ads last about as long as a budget deodorant in a heat wave.

In their book on the 2012 campaign, “The Gamble,” political scientists John Sides and Lynn Vavreck found that “the effects of advertising decayed rapidly. We found no significantly meaningful impact of ads aired five days before … respondents were interviewed.”  

Media consultants undoubtedly vigorously disagree. What seems certain is that a supposedly “killer ad” aired in June or September will be forgotten by Election Day. 

6) Remember that swing voters see TV ads amid the clutter of other commercials. 

Political cognoscenti may be monitoring YouTube feeds for the latest TV spots, but that isn’t how ordinary voters view them. If you really want to gauge the impact of a single campaign ad, spend an evening this fall watching network TV in, say, Green Bay, Wis. You will soon realize that your head is being pounded by negative spots with scary music and voice-of-doom narration.

7) Understand there is, in effect, a super PAC tax.

The most successful campaign reform in history predates Watergate. Beginning 60 days before an election, it grants candidates for any public office (including county coroner) the lowest advertising rate that a broadcast TV station offers for that time slot.

In contrast, there is no discount for a super PAC, a party committee or a cause group. They have to pay the going rate, which in a hot political media market like Las Vegas or Milwaukee might be five times what a candidate pays. All money in politics is not alike. 

8) Remember there has never been a June presidential debate.

Be skeptical about any glib after-action assessments of who won and what it ultimately means. In Denver in early October 2012, Obama had probably the worst debate of any candidate in history, including Richard Nixon in 1960. A month later, Obama easily won reelection. 

9) Political conventions are infomercials — and nothing more.

There has not been a contested moment at a convention since the 1980s. These days, they have only one purpose: free prime-time advertising for each party. 

Every moment in Milwaukee (Republicans) or Chicago (Democrats) that is off-message because of demonstrations, embarrassing flubs or out-of-control speakers is a major loss. The goal is to have a convention totally devoid of news value but compelling enough to bring home wayward partisans.

10) Third-party candidates generally shed support in the waning days of a campaign.

Never underestimate the potency of the wasted-vote syndrome. No matter what voters tell pollsters, they often rethink their political choices when they realize that their favored candidate cannot win. 

What this means is that it is hard to assess the impact of candidates like Robert F. Kennedy Jr. What will be particularly tricky is predicting in advance whom he will take votes from. After all, political analysts are still arguing over whether Ross Perot’s third-party candidacy in 1992 gave Bill Clinton the White House. 

Walter Shapiro is a staff writer for The New Republic and a lecturer in political science at Yale. He is a veteran of USA Today, Time, Newsweek and The Washington Post.