Since Kamala Harris lost to Donald Trump last year, various Democratic elected officials, think tanks, and strategists have been frantically attempting to find some magic political formula that will allow the party to beat Republicans once more.
The latest version comes from a new big donor–backed strategist group called Welcome, and it’s titled “Deciding to Win.” The advice, surprise surprise, is that Democrats should punch the left—abandon progressive policy like Medicare for All, stop talking about LGBTQ people and climate change, and focus on milquetoast kitchen-table issues. It’s the same thing we’ve already seen a hundred times this year.
But I happen to have some other polling, provided exclusively to the Prospect by Data for Progress, that sheds a different light on what Democrats should have done. To sum up, a large majority of American voters are greatly dissatisfied with the state of things, most especially the economy. It turns out that median voters were catastrophically misled about the stakes of the election last year. Addressing that problem is a prerequisite for any messaging to break through, regardless of content.
Data for Progress approaches this from some interesting directions. They asked whether affordability or the decline in social connection between Americans is worse, and found more than two-thirds answering the former. Only men who voted for Trump had a majority more worried about community, and that not by much. Interestingly, young people were also more concerned about community than older folks—perhaps reflecting the disproportionate hold that the solitary viewing of social media wields over young people, young men in particular. Insofar as voting for Trump was hoped to heal the cultural malaise caused by Silicon Valley tech platforms, it hasn’t worked.
Data for Progress also asked about the state of the country and affordability directly; 60 percent said life in this country is getting worse, and 43 percent said they were extremely worried about the rising cost of living. These are not Ronald Reagan-in-1983 numbers. People are unhappy; they aren’t getting what they want. “The cost of living has understandably dominated the conversation over the last year, and while voters see America’s problems as more economic than cultural, they do express high levels of concern about the divisiveness in our culture and other challenges associated with modernity,” said Ryan O’Donnell, interim executive director at Data for Progress.
These results complicate a foundational assumption behind “Deciding to Win,” namely, that voters make decisions like policy wonks do—by sitting down with a list of positions and picking the candidate who is closest to their own preferences. The authors insist that “substantive positioning does affect electoral outcomes,” and cite several academic papers to this effect, as well as their own analysis that endorsing various pieces of legislation could change one’s vote total by no more than 9 percent, though most only registered in the low single digits.
Still, all the research cited in this section of “Deciding to Win,” except for one preprint, examines evidence from 2018 or before—long before Elon Musk dumped an ocean of toxic waste in American political discourse. And I would like to pop the hood on how “expectations” were calculated, as we’ll see below. But whatever the case, these effects of changing policy stances are not huge numbers. I’m highly skeptical of the efficacy of policy messaging on its own, but I wouldn’t have said that, say, endorsing free college—a massive reform to higher education—would make no difference to one’s vote total whatsoever. A swing of about 1 percent, which is what they find, sounds right.
But quibbles aside, it simply cannot be the case that the candidate with the most moderate messaging always wins, because unlike Harris, Donald Trump ran a paint-blisteringly extreme campaign in 2024, and he won anyway. He ran on tariffs, cuts to health care programs, and big tax cuts for the rich. All of those things were and are quite unpopular. His de facto campaign platform of Project 2025 was deeply unpopular. His Supreme Court nominees had repealed Roe v. Wade (extremely unpopular), and he was convicted of 34 felonies. (I can’t find any polling outfit that has even tried to ask this question, but I think it’s a safe bet most would answer “Should the president be a literal convicted felon?” in the negative.)
Tariffs deserve particular attention, because while voters do tend to blame the incumbent party for what’s happening, in 2024 they hated inflation, and tariffs are quite literally a “make prices go up button” that Trump was promising to press, loudly and repeatedly. Voters also loathed the fact that Joe Biden was extremely old, but apparently did not mind the same being true of Trump, despite his repeatedly glitching out on stage for many minutes, and his opponent being almost 20 years younger. Either voters didn’t hear about what Trump said and did—perhaps because of how the mainstream media compulsively laundered or obscured his extremism—or didn’t understand it, or didn’t believe it.
This also handily explains why Trump’s popularity has plummeted since taking office, even on his previously strongest subjects. The Economist has him at 18 points underwater, his lowest level ever. A critical mass of people who decided the election were bamboozled by America’s horrendously broken information environment, and don’t like the results of their own choices.
There are more objections one could make about “Deciding to Win.” There is the unstated implication, as usual for this sort of operation, that Kamala Harris ran as some sort of transgender antifa climate activist, when she actually ran as a proud gun-owning moderate who is friends with prominent conservatives. Conversely, Joe “At Least Three Genders” Biden ran a markedly progressive campaign in 2020 (his later behavior on Gaza notwithstanding) and won handily.
Then there is some funny business going on with Welcome’s Medicare for All polling. They find it 11 points underwater, which as Carl Beijer points out, is between 28 and 51 points worse than other polls from reputable sources. That might be because they claimed in their question that it would require a payroll tax more than twice as big as the one in Bernie Sanders’s bill. (What it would actually require would be a policy question; there is already enough tax revenue to fund universal Medicare if we paid similar prices as other rich nations.)
It’s almost like this is a cynical exercise in cooking the political books designed to head off anything that might upset the billionaire donor class that pays for this kind of stuff.
But whatever the case, I’d advise said billionaires to start standing up some media institutions that can combat the far right’s death grip on the American information environment before funding yet another centrist think tank (which could justly be called the Center for Better Things Aren’t Possible). We’ll never get a means-tested tax credit for small-business entrepreneurs if Americans never hear about it. But if they do, they’re sure to love it.