Are Pollsters Gaslighting Us?
Updated: Oct 27, 2021
Could the 2020 presidential race be a WHOLE lot closer than “experts” are telling us?
If this election season were a movie, it would look a lot like one we rented four years ago.
Once again, experts have given a litany of reasons why Donald Trump can’t possibly win the White House: shifting state demographics; suburban white women fed up with his sexist antics; people of color fed up with his racist antics; his botched handling of the pandemic; and an economy thrashed by lockdowns.
Once again, Trump started off double digits behind in the polls with a seemingly impossible chance of closing the gap. Once again his opponent’s formidable lead has evaporated, and they’re fighting to stay outside the margin of error to best the Orange Menace.
Polling experts tell us it’s natural for the gap to narrow as Election Day nears and undecideds slip off the fence. But is that really what's happening this year?
A handful of rogue pollsters have claimed throughout the 2020 election season that this race has always been closer than most polls — specifically, the ones relied upon by mainstream media — have been telling us. These fringe “experts” are among the very few who came even close to projecting a Trump win in 2016.
One pollster in particular, the Trafalgar Group, not only predicted a Trump win four years ago, but even anticipated that he would narrowly take Pennsylvania, Michigan, Wisconsin, Florida and North Carolina. Trafalgar now sees a similar pattern taking shape in 2020. Their polling has consistently shown a much tighter race in the battleground states. Why?
Trafalgar believes it’s seeing tell-tale signs that most polling firms are either missing or refusing to acknowledge. They insist that shy Trump voters aren’t mythical beasts; they actually exist. And if they were reluctant to admit their support for Trump in 2016, the current climate has made them even less visible now. When anyone can be “cancelled”on a moment’s notice, and people are assaulted for wearing MAGA apparel, is it surprising that shy voters might want to keep an even lower profile in 2020?
Trafalgar has also picked up on the fact that while Trump may be losing some suburban white moms and working class whites, he hasn’t lost college educated white voters. Moreover, he’s managed to eat into the Democratic firewall by gaining ground among a surprising group: people of color.
In 2016, Trump garnered 29% of the Latino vote, and he’s increased that support in key states: even mainstream polls give him the support of 38% of Latino voters in Texas and 43% in Florida. While Trump received only 8% of the black vote in 2016, he’s expected to take between 12% and 21% this year. These are seismic shifts that are ringing alarm bells among Democratic party elites.
Does this mean that Trafalgar is right and Trump will win? Of course not. But given their uncanny ability to spot what other pollsters completely missed in 2016, it might mean we shouldn’t be shocked if Trump does win. It might also give us reason to ask common sense questions the mainstream media have avoided as they’ve clung to traditional polling data:
Did it really make sense that in a country so radically polarized, Joe Biden would ever have had a double digit lead over Trump at any point?
Did it really make sense that 1/4 of the voters who supported Trump in 2016 suddenly abandoned him when Biden received the nomination, and then decided to return to the fold in the final week?
How does Biden —who was once a shoo-in — suddenly find himself almost neck-and-neck with a president who stokes white supremacy, caused 230,000 COVID-19 deaths, and destroyed the economy? Nothing has changed on any of these fronts in the past six months. If anything, these issues have made Trump less “appealing” to voters.
So why does polling data now reveal a race that’s suddenly so much closer?
Here’s something else to consider: on September 1, when Trump was still almost double-digits behind in the polls, Michael Bloomberg’s data firm, Hawkfish, warned of a "red mirage,"predicting that “it’s highly likely that President Trump will appear to have won” on election night, "even if he ultimately loses when all the votes are counted."
How could a candidate who was so woefully behind in the polls appear to have won on election night?
This doesn’t make sense, and it points to what a growing number of Americans are starting to suspect: our electoral reality may have been distorted. This race is, and probably always has been, much closer than most pollsters have told us.
It’s not clear whether this distortion of our reality is intentional or inadvertent. Moreover, it doesn’t mean that most polls are always wrong. But given the way things seem to play out every election season, it does suggest that polling may not only be gauging voter sentiment, but also steering voter expectations and election outcomes.
Think about how polling usually unfolds during each election cycle:
First Stage: Shortly after the primaries end, polling firms use methodology and assumptions they’ve relied upon for decades, giving one candidate a sizable, often double-digit lead over their opponent and establishing them as the frontrunner. This daunting lead sets expectations early, dampens the enthusiasm of their opponents’ supporters, and creates a sense of inevitability. It also gives the leading candidate something invaluable: an early psychological advantage.
Second Stage: Two to three months before the election, the race tightens, and the frontrunner’s double digit lead slips into high single digits. This signals to their supporters that they can’t take victory for granted and encourages them to double-down on their fundraising and get-out-the-vote efforts. Despite this narrowing gap, however, the frontrunner is still widely expected to prevail. Their opponent is repeatedly reminded that their chances of winning are slim to none, but they’re encouraged to keep playing, anyway. This stage increases support for the frontrunner, giving them a “real” advantage while simultaneously demoralizing their opponent's supporters.
Third Stage: In the final week, polls tighten even more. The frontrunner’s lead slips into the low single digits, still outside the margin of error but still giving them overwhelming odds of winning. Their opponent continues to fight valiantly, even though no one really expects them to win. The frontrunner’s supporters accept that a narrow victory, while not the ideal outcome, is likely. But more importantly, they see a loss as inconceivable. They see their candidate — who has been the consistent frontrunner — as destined, and even entitled, to win. This expectation gives the frontrunner the moral/righteous advantage on Election Day.
What the frontrunner’s supporters never consider is the possibility that most polls — all using more or less the same methodology —could ever be wrong. It never occurs to them that their candidate’s lead in key states may actually be within the margin of error, or that they may even be trailing their opponent.
They’re never forced to contemplate these possibilities because in most election cycles their expectations are rarely challenged. Because the longer a candidate leads in the polls, the more people come to accept that they will win, and the harder it becomes to imagine that they could possibly lose. Since everyone wants to be part of a winning team, even voters who initially resisted the frontrunner eventually join their team because polling has influenced their expectations.
This is why the frontrunner usually wins.
In most election cycles, this process plays out seamlessly. But what if this year it's different? What if 2020 is more like 2016 and less like other years? What if enough “shy” Trump supporters and newly-turned voters of color turn out in sufficient numbers to once again defy expectations created by polling?
Trafalgar’s unique data sets and common sense inconsistencies of what’s clearly observable on the ground suggest that this year, polling may have once again created expectations that don’t reflect reality. And in a country teetering on the verge of civil unrest, this is extremely dangerous. Because when voters’ polling-based expectations aren’t met, chaos can ensue.
This is what happened in 2000, when George W. Bush eked out a win in the closest election in U.S. history, a race that was never “supposed” to be so close. That’s when we learned just how politically divided America had become.
It happened again in 2016 when most polls grossly underestimated the size of Trump’s working class base. The disconnect left Clinton supporters, who had been assured of victory, shocked and traumatized.
If most polls have once again underestimated the breadth of Trump’s voters and set unrealistic expectations for Biden’s supporters, we could be in for more shock and trauma.
Even worse, mainstream media is feeding these expectations with the “red mirage” narrative that may prompt Trump to prematurely declare victory before all mail-in votes are counted. The expectation has been set that any edge Trump appears to have will ultimately be defeated by mail-in ballots — because in a close race, the expectation is that the candidate who’s been leading in the polls can’t lose, and the one who’s been lagging can’t win.
If this race is as close as Trafalgar (and common sense) suggests, then voters should emotionally prepare themselves for either outcome. But that’s not what’s happening. Biden’s advantage in major polls has created — and continues to create — a sense of inevitability and entitlement, and the expectation that Trump can't possibly win. This narrative could spell disaster for our country.
If polling threatens to create unrealistic expectations in an election year that may once again defy conventional wisdom, it may doom us to a repeat of 2000: an election rife with recounts, litigation, and even more division.
It may be time to re-assess the value and need for public polling. While it can help create a more dramatic and exciting election season — fueling TV ratings, encouraging discussion of issues, and engaging voters— in an increasingly polarized America, it could also end up doing more harm than good.
It may be time to ask ourselves whether public polling is not only gauging voter sentiment, but also steering it; if it’s not only assessing our preferences, but also shaping them.
In an increasingly polarized America, public polling may guarantee that no matter which candidate wins, we all lose.
The world is getting strange, and I do my best to help you make sense of it. If you enjoyed this, sign up here to get my latest posts as soon as they're available!