Why the Polls May Be Feeding Liberals Another Blue Mirage

6 minute read

This article is part of The D.C. Brief, TIME’s politics newsletter. Sign up here to get stories like this sent to your inbox.

The easiest explanation for a natural mirage is that light moves faster through hot air than cold air, and rays will bend to find the path of least resistance. Phantom sightings of watering holes in deserts, for instance, are as encouraged by the psyche as the physics of light on the horizon chasing hotter air. Desert mirages are something of a lazy trope in cartoons, but the phenomenon happens plenty in politics, too, and some Democrats are worrying that they’re seeing oases where none may exist. The source of those illusions? All those polls suggesting President Joe Biden’s party may not be as screwed as they had thought.

By historical benchmarks, Democrats should definitely be in trouble. Since World War II—with one exception, after 9/11—the party in power at the White House gets flayed in their first at bat with voters picking their members of the House. It doesn’t matter if it’s Ronald Reagan in 1982 or Bill Clinton in 1994; the first term of a President leaves the electorate ready to reject the latest experiment.

At the moment, though, there are at least some cold pockets of air that suggest the recent optimism coming from the Democratic side may not be just figments of liberal imaginations. While the money race is competitive and the feeling in the field is that the ground games on both sides are evenly matched, Republicans are stumbling toward the general election with some lousy nominees, particularly for Senate.

Nowhere is the faith in the mirage more pronounced than in polling, a political junky’s favorite cheap buzz. Spend any time looking at the numbers these days and you’re likely seeing positive trends for Democratic candidates. Sens. Mark Kelly of Arizona, Michael Bennet of Colorado, and Maggie Hassan of New Hampshire have consistently led their Republican challengers. Pennsylvania seems increasingly tough for GOP nominee Dr. Mehmet Oz, while Wisconsin is, by all signs, a true jump-ball race between incumbent Republican Sen. Ron Johnson and the Democrat, Lieutenant Governor Mandela Barnes. Democratic Sens. Raphael Warnock and Catharine Cortez Masto are inches ahead of their rivals in Georgia and Nevada. And in Ohio, Florida, and North Carolina, states where Donald Trump won by 8 points, 3 points, and 1 point, respectively, the Democratic candidates for Senate remain within striking distance.

So why aren’t Democrats giddy? Put simply: responsible Democrats know they have been here before—seeing the palm trees in the desert, ordering cocktails with matching umbrellas—and walked away big losers.

The perceived flaws in polling have been manifest in conversations about recent elections. In recent cycles, pollsters supposedly missed the silent Trump supporters who either refused to participate in surveys or lied about their true allegiances. Pew’s deep dive into the problem is a must-read for why nuance matters a ton in this space. Even though the polls were collectively off by about four points, according to an industry autopsy, they weren’t entirely afield; they just missed some of the shading. Still, the hue matters when billions of dollars, control of Washington, and maybe the future of democracy itself is in the balance.

(Also, don’t discount the political impact of COVID-19 during the 2020 cycle. That was a huge pocket of icy air that bent a lot of perceptions, created plenty of mirages, and, according to a government report, changed votes.)

These concerns about an over-reliance on polling isn’t new. Some have suggested it’s time for all of us to quit being polling junkies hooked on the heroin of crosstabs. Others have said the polls themselves are in need of a complete overhaul, or should be abandoned entirely.

Tempting, sure. But maybe the answer is a more-informed use of the numbers.

Armchair pundits have invested far too much confidence in surveys that—by their very definition—have a baked-in margin of error, and expectations that occasionally a garbage reading gets taken. In that, it’s on the consumer to understand the product; no one would blame Elon Musk for someone who drove their Tesla into the sea to find it didn’t float. Similarly, blaming pollsters for their output when it is misapplied isn’t fair.

That said, the polls could be off once again, and maybe even worse than in 2020, which was the biggest polling misfire since 1980. Even the most seasoned pollsters concede their work in recent years has missed parts of the electorate. In the wake of the last midterm results that caught Democrats flat footed, their top field general told her colleagues that all of the numbers were wrong: “I also want to say the thing we’re all feeling: I’m furious. Something went wrong here across the entire political world,” Rep. Cheri Bustos said during a tense call.

It’s entirely possible we’re in the midst of a sequel, of course. But a better use of time would be for the political junkies to diversify their crack: cut some early-vote numbers into the polling, blend in some new-registration figures.

In this, it’s hard to ignore the lyrics that roared this weekend at the Kennedy Center, which capped its yearlong celebration of the performing arts hub’s 50th birthday with a performance of the work that opened it back in 1971. In Leonard Bernstein’s Mass, the chorus belts a lyric that seems as diagnostic today as it was in the wake of the turbulent 1960s, even if the drug of choice for Washington insiders these days is data: “Half of the people are stoned and the other half are waiting for the next election. Half of the people are drowned and the other half are swimming in the wrong direction.” For a lot of D.C. this election season, there’s a whole lot of overlap in those two groups.

Make sense of what matters in Washington. Sign up for the D.C. Brief newsletter.

More Must-Reads From TIME

Write to Philip Elliott at philip.elliott@time.com