I’ve been wondering for a while about the wildly varying poll numbers for Robert F. Kennedy Jr., and the more I learn the more I think his standing will soon fade, as I explained at New York:

Robert F. Kennedy Jr. needs just one more really strong poll showing to meet part of CNN’s criteria for participation in its June 27 presidential debate. The network stipulates that participants secure at least 15 percent in four national surveys no later than June 20 from a specific list of approved pollsters, and Kennedy has three (a Quinnipiac poll and a CNN poll in April, and a Marquette Law School poll in May). He may still fail to make the stage because CNN also requires that participants be on the ballot in states representing at least 270 electoral votes, and the Kennedy campaign is in a fight it may not win over how ballot access is confirmed. But still, the idea that Kennedy is polling as well as any non-major-party candidate since Ross Perot is a good advertisement for his viability.

There are, however, two reasons Kennedy’s standing in the race may be significantly overstated by the best of his polls. The first is a matter of history and, well, common sense: as Election Day nears, voters begin to focus on the most viable options and become less likely to “waste their votes” on candidates with slim odds of actually winning. So even the strongest of non-major-party presidential candidates in living memory — Perot in 1992 and 1996, John Anderson in 1980, and George Wallace in 1968 — all lost ground by late summer of the election year and finished well below their peak in polls. It’s one key reason RFK Jr. is frantic to get into a debate with the Democratic and Republican nominees (as Perot, the strongest indie candidate ever, did in 1992); he needs a game-changing development to forestall the otherwise inevitable late-cycle swoon.

But there’s another reason that polls showing Kennedy in the mid-to-high teens could overstate his actual support: They are counterbalanced by other polls showing him performing much more poorly. Indeed, all three June polls testing five major- and minor-party candidates (Biden, Trump, Kennedy, Cornel West, and Jill Stein) place RFK Jr. a lot lower: 6 percent at Emerson, 4 percent at Yahoo News, and 3 percent at Economist/YouGov.

Why are there such wild gyrations in Kennedy’s standing in the polls? There’s no infallible answer, but the New York Times’ Ruth Igielnik offered a persuasive explanation last month: Question order in presidential polls has a big effect on non-major-party candidates:

“[M]any reputable pollsters ask both versions of the question: one that poses a simple head-to-head contest between major-party candidates, and one that includes third-party candidates who may be on the ballot.

“And which question gets asked first is where the difference comes in. …

“Our experiment worked like this: All respondents were shown both the long and short questions, but half were shown the full list first, and the other half were first shown the two-way race.

“Among those who saw the long list first, Mr. Kennedy garnered 7 percent of the vote.

“But among those respondents who encountered the head-to-head contest before seeing the full list, Mr. Kennedy’s support shot up six percentage points to 13 percent.”

That’s a very big difference. What explains it?

“[I]t is at least partly related to a phenomenon that pollsters call expressive responding. This is when people might use a survey response to show their frustration or express a particular feeling that’s not exactly what is being asked.

“In this case, many respondents seem to be using the second question to convey frustration with the choices for president in the first question, whether or not their answers reflect their full views.”

As you probably know, frustration “with the choices for president” is famously high this year. Igielnik goes on to show that most of the recent polls showing Kennedy with double-digit support are those that ask first about the head-to-head Biden-Trump contest before including the other candidates in a second question, while those that present the full list of candidates right off the bat tend to show much lower support for the conspiracy theorist with the famous name.

When voters actually vote, of course, they are going to see the full list of candidates without first encountering some frustrating presentation of the major-party choices alone. So the odds are good that Kennedy will underperform his best polls. Indeed, putting together the two factors we’ve discussed, it’s not surprising to learn that RFK Jr.’s standing in the RealClearPolitics polling averages has steadily drifted downward from nearly 17 percent last November to 13 percent as recently as March to 8.6 percent today. Defying history by making a serious run at Biden and Trump will take a lot of doing for Kennedy and isn’t a very good bet.